#include "include/stringify.h"
#include <boost/algorithm/string.hpp>
#include <boost/lexical_cast.hpp>
-#include <boost/regex.hpp>
+#include <regex>
// Definitions for enums
#include "common/perf_counters.h"
+// rbd feature validation
+#include "librbd/Features.h"
+
+namespace {
+class printer : public boost::static_visitor<> {
+ ostream& out;
+public:
+ explicit printer(ostream& os)
+ : out(os) {}
+ template<typename T>
+ void operator()(const T& v) const {
+ out << v;
+ }
+ void operator()(boost::blank blank) const {
+ return;
+ }
+ void operator()(bool v) const {
+ out << (v ? "true" : "false");
+ }
+ void operator()(double v) const {
+ out << std::fixed << v << std::defaultfloat;
+ }
+ void operator()(const Option::size_t& v) const {
+ out << v.value;
+ }
+ void operator()(const std::chrono::seconds v) const {
+ out << v.count();
+ }
+};
+}
+
+ostream& operator<<(ostream& os, const Option::value_t& v) {
+ printer p{os};
+ v.apply_visitor(p);
+ return os;
+}
void Option::dump_value(const char *field_name,
const Option::value_t &v, Formatter *f) const
if (boost::get<boost::blank>(&v)) {
// This should be nil but Formatter doesn't allow it.
f->dump_string(field_name, "");
- } else if (type == TYPE_UINT) {
- f->dump_unsigned(field_name, boost::get<uint64_t>(v));
- } else if (type == TYPE_INT) {
- f->dump_int(field_name, boost::get<int64_t>(v));
- } else if (type == TYPE_STR) {
- f->dump_string(field_name, boost::get<std::string>(v));
- } else if (type == TYPE_FLOAT) {
- f->dump_float(field_name, boost::get<double>(v));
- } else if (type == TYPE_BOOL) {
- f->dump_bool(field_name, boost::get<bool>(v));
- } else {
- f->dump_stream(field_name) << v;
+ return;
+ }
+ switch (type) {
+ case TYPE_INT:
+ f->dump_int(field_name, boost::get<int64_t>(v)); break;
+ case TYPE_UINT:
+ f->dump_unsigned(field_name, boost::get<uint64_t>(v)); break;
+ case TYPE_STR:
+ f->dump_string(field_name, boost::get<std::string>(v)); break;
+ case TYPE_FLOAT:
+ f->dump_float(field_name, boost::get<double>(v)); break;
+ case TYPE_BOOL:
+ f->dump_bool(field_name, boost::get<bool>(v)); break;
+ default:
+ f->dump_stream(field_name) << v; break;
}
}
return 0;
}
+int Option::parse_value(
+ const std::string& raw_val,
+ value_t *out,
+ std::string *error_message,
+ std::string *normalized_value) const
+{
+ std::string val = raw_val;
+
+ int r = pre_validate(&val, error_message);
+ if (r != 0) {
+ return r;
+ }
+
+ if (type == Option::TYPE_INT) {
+ int64_t f = strict_si_cast<int64_t>(val.c_str(), error_message);
+ if (!error_message->empty()) {
+ return -EINVAL;
+ }
+ *out = f;
+ } else if (type == Option::TYPE_UINT) {
+ uint64_t f = strict_si_cast<uint64_t>(val.c_str(), error_message);
+ if (!error_message->empty()) {
+ return -EINVAL;
+ }
+ *out = f;
+ } else if (type == Option::TYPE_STR) {
+ *out = val;
+ } else if (type == Option::TYPE_FLOAT) {
+ double f = strict_strtod(val.c_str(), error_message);
+ if (!error_message->empty()) {
+ return -EINVAL;
+ } else {
+ *out = f;
+ }
+ } else if (type == Option::TYPE_BOOL) {
+ if (strcasecmp(val.c_str(), "false") == 0) {
+ *out = false;
+ } else if (strcasecmp(val.c_str(), "true") == 0) {
+ *out = true;
+ } else {
+ int b = strict_strtol(val.c_str(), 10, error_message);
+ if (!error_message->empty()) {
+ return -EINVAL;
+ }
+ *out = (bool)!!b;
+ }
+ } else if (type == Option::TYPE_ADDR) {
+ entity_addr_t addr;
+ if (!addr.parse(val.c_str())){
+ return -EINVAL;
+ }
+ *out = addr;
+ } else if (type == Option::TYPE_ADDR) {
+ entity_addrvec_t addr;
+ if (!addr.parse(val.c_str())){
+ return -EINVAL;
+ }
+ *out = addr;
+ } else if (type == Option::TYPE_UUID) {
+ uuid_d uuid;
+ if (!uuid.parse(val.c_str())) {
+ return -EINVAL;
+ }
+ *out = uuid;
+ } else if (type == Option::TYPE_SIZE) {
+ Option::size_t sz{strict_iecstrtoll(val.c_str(), error_message)};
+ if (!error_message->empty()) {
+ return -EINVAL;
+ }
+ *out = sz;
+ } else if (type == Option::TYPE_SECS) {
+ try {
+ *out = parse_timespan(val);
+ } catch (const invalid_argument& e) {
+ *error_message = e.what();
+ return -EINVAL;
+ }
+ } else {
+ ceph_abort();
+ }
+
+ r = validate(*out, error_message);
+ if (r != 0) {
+ return r;
+ }
+
+ if (normalized_value) {
+ *normalized_value = to_str(*out);
+ }
+ return 0;
+}
+
void Option::dump(Formatter *f) const
{
- f->open_object_section("option");
f->dump_string("name", name);
f->dump_string("type", type_to_str(type));
dump_value("min", min, f);
dump_value("max", max, f);
+ f->dump_bool("can_update_at_runtime", can_update_at_runtime());
+
+ f->open_array_section("flags");
+ if (has_flag(FLAG_RUNTIME)) {
+ f->dump_string("option", "runtime");
+ }
+ if (has_flag(FLAG_NO_MON_UPDATE)) {
+ f->dump_string("option", "no_mon_update");
+ }
+ if (has_flag(FLAG_STARTUP)) {
+ f->dump_string("option", "startup");
+ }
+ if (has_flag(FLAG_CLUSTER_CREATE)) {
+ f->dump_string("option", "cluster_create");
+ }
+ if (has_flag(FLAG_CREATE)) {
+ f->dump_string("option", "create");
+ }
f->close_section();
}
+std::string Option::to_str(const Option::value_t& v)
+{
+ return stringify(v);
+}
+
+void Option::print(ostream *out) const
+{
+ *out << name << " - " << desc << "\n";
+ *out << " (" << type_to_str(type) << ", " << level_to_str(level) << ")\n";
+ if (!boost::get<boost::blank>(&daemon_value)) {
+ *out << " Default (non-daemon): " << stringify(value) << "\n";
+ *out << " Default (daemon): " << stringify(daemon_value) << "\n";
+ } else {
+ *out << " Default: " << stringify(value) << "\n";
+ }
+ if (!enum_allowed.empty()) {
+ *out << " Possible values: ";
+ for (auto& i : enum_allowed) {
+ *out << " " << stringify(i);
+ }
+ *out << "\n";
+ }
+ if (!boost::get<boost::blank>(&min)) {
+ *out << " Minimum: " << stringify(min) << "\n"
+ << " Maximum: " << stringify(max) << "\n";
+ }
+ *out << " Can update at runtime: "
+ << (can_update_at_runtime() ? "true" : "false") << "\n";
+ if (!services.empty()) {
+ *out << " Services: " << services << "\n";
+ }
+ if (!tags.empty()) {
+ *out << " Tags: " << tags << "\n";
+ }
+ if (!see_also.empty()) {
+ *out << " See also: " << see_also << "\n";
+ }
+
+ if (long_desc.size()) {
+ *out << "\n" << long_desc << "\n";
+ }
+}
+
constexpr unsigned long long operator"" _min (unsigned long long min) {
return min * 60;
}
Option("host", Option::TYPE_STR, Option::LEVEL_BASIC)
.set_description("local hostname")
.set_long_description("if blank, ceph assumes the short hostname (hostname -s)")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_service("common")
.add_tag("network"),
Option("fsid", Option::TYPE_UUID, Option::LEVEL_BASIC)
.set_description("cluster fsid (uuid)")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_service("common")
.add_tag("service"),
.set_description("path to MonMap file")
.set_long_description("This option is normally used during mkfs, but can also "
"be used to identify which monitors to connect to.")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_service("mon")
- .add_tag("mkfs"),
+ .set_flag(Option::FLAG_CREATE),
Option("mon_host", Option::TYPE_STR, Option::LEVEL_BASIC)
.set_description("list of hosts or addresses to search for a monitor")
"list of IP addresses or hostnames. Hostnames are "
"resolved via DNS and all A or AAAA records are "
"included in the search list.")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_service("common"),
Option("mon_dns_srv_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("ceph-mon")
.set_description("name of DNS SRV record to check for monitor addresses")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_service("common")
.add_tag("network")
.add_see_also("mon_host"),
// lockdep
Option("lockdep", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_description("enable lockdep lock dependency analyzer")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_service("common"),
Option("lockdep_force_backtrace", Option::TYPE_BOOL, Option::LEVEL_DEV)
.add_service("common")
.add_see_also("admin_socket"),
- Option("crushtool", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_description("name of the 'crushtool' utility")
- .add_service("mon"),
-
// daemon
Option("daemonize", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
.set_daemon_default(true)
.set_description("whether to daemonize (background) after startup")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_service({"mon", "mgr", "osd", "mds"})
.add_tag("service")
.add_see_also({"pid_file", "chdir"}),
Option("chdir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_description("path to chdir(2) to after daemonizing")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_service({"mon", "mgr", "osd", "mds"})
.add_tag("service")
.add_see_also("daemonize"),
.add_service({"mon", "mgr", "osd", "mds"})
.add_tag("service"),
+ Option("crash_dir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("/var/lib/ceph/crash")
+ .set_description("Directory where crash reports are archived"),
+
// restapi
Option("restapi_log_level", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_description("default set by python code"),
Option("erasure_code_dir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default(CEPH_PKGLIBDIR"/erasure-code")
.set_description("directory where erasure-code plugins can be found")
- .add_service({"mon", "osd"})
- .set_safe(),
+ .add_service({"mon", "osd"}),
// logging
Option("log_file", Option::TYPE_STR, Option::LEVEL_BASIC)
.set_default("")
.set_daemon_default("/var/log/ceph/$cluster-$name.log")
.set_description("path to log file")
- .add_see_also({"log_to_stderr",
+ .add_see_also({"log_to_file",
+ "log_to_stderr",
"err_to_stderr",
"log_to_syslog",
"err_to_syslog"}),
.set_description("recent log entries to keep in memory to dump in the event of a crash")
.set_long_description("The purpose of this option is to log at a higher debug level only to the in-memory buffer, and write out the detailed log messages only if there is a crash. Only log entries below the lower log level will be written unconditionally to the log. For example, debug_osd=1/5 will write everything <= 1 to the log unconditionally but keep entries at levels 2-5 in memory. If there is a seg fault or assertion failure, all entries will be dumped to the log."),
+ Option("log_to_file", Option::TYPE_BOOL, Option::LEVEL_BASIC)
+ .set_default(true)
+ .set_description("send log lines to a file")
+ .add_see_also("log_file"),
+
Option("log_to_stderr", Option::TYPE_BOOL, Option::LEVEL_BASIC)
.set_default(true)
.set_daemon_default(false)
.set_description("send critical error log lines to stderr"),
Option("log_stderr_prefix", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_description("String to prefix log messages with when sent to stderr"),
+ .set_description("String to prefix log messages with when sent to stderr")
+ .set_long_description("This is useful in container environments when combined with mon_cluster_log_to_stderr. The mon log prefixes each line with the channel name (e.g., 'default', 'audit'), while log_stderr_prefix can be set to 'debug '.")
+ .add_see_also("mon_cluster_log_to_stderr"),
Option("log_to_syslog", Option::TYPE_BOOL, Option::LEVEL_BASIC)
.set_default(false)
.set_description("port number for the remote graylog server")
.add_see_also("log_graylog_host"),
+ Option("log_coarse_timestamps", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .set_description("timestamp log entries from coarse system clock "
+ "to improve performance")
+ .add_service("common")
+ .add_tag("performance")
+ .add_tag("service"),
// unmodified
Option("clog_to_monitors", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("default=true")
- .set_description(""),
+ .set_description("Make daemons send cluster log messages to monitors"),
Option("clog_to_syslog", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("false")
- .set_description(""),
+ .set_description("Make daemons send cluster log messages to syslog"),
Option("clog_to_syslog_level", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("info")
- .set_description(""),
+ .set_description("Syslog level for cluster log messages")
+ .add_see_also("clog_to_syslog"),
Option("clog_to_syslog_facility", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("default=daemon audit=local0")
- .set_description(""),
+ .set_description("Syslog facility for cluster log messages")
+ .add_see_also("clog_to_syslog"),
Option("clog_to_graylog", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("false")
- .set_description(""),
+ .set_description("Make daemons send cluster log to graylog"),
Option("clog_to_graylog_host", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("127.0.0.1")
- .set_description(""),
+ .set_description("Graylog host to cluster log messages")
+ .add_see_also("clog_to_graylog"),
Option("clog_to_graylog_port", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("12201")
- .set_description(""),
+ .set_description("Graylog port number for cluster log messages")
+ .add_see_also("clog_to_graylog"),
Option("mon_cluster_log_to_stderr", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description("Send cluster log messages to stderr (prefixed by channel)"),
+ .add_service("mon")
+ .set_description("Make monitor send cluster log messages to stderr (prefixed by channel)")
+ .add_see_also("log_stderr_prefix"),
Option("mon_cluster_log_to_syslog", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("default=false")
- .set_description(""),
+ .add_service("mon")
+ .set_description("Make monitor send cluster log messages to syslog"),
Option("mon_cluster_log_to_syslog_level", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("info")
- .set_description(""),
+ .add_service("mon")
+ .set_description("Syslog level for cluster log messages")
+ .add_see_also("mon_cluster_log_to_syslog"),
Option("mon_cluster_log_to_syslog_facility", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("daemon")
- .set_description(""),
+ .add_service("mon")
+ .set_description("Syslog facility for cluster log messages")
+ .add_see_also("mon_cluster_log_to_syslog"),
+
+ Option("mon_cluster_log_to_file", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .add_service("mon")
+ .set_description("Make monitor send cluster log messages to file")
+ .add_see_also("mon_cluster_log_file"),
Option("mon_cluster_log_file", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("default=/var/log/ceph/$cluster.$channel.log cluster=/var/log/ceph/$cluster.log")
- .set_description(""),
+ .add_service("mon")
+ .set_description("File(s) to write cluster log to")
+ .set_long_description("This can either be a simple file name to receive all messages, or a list of key/value pairs where the key is the log channel and the value is the filename, which may include $cluster and $channel metavariables")
+ .add_see_also("mon_cluster_log_to_file"),
Option("mon_cluster_log_file_level", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("debug")
- .set_description(""),
+ .add_service("mon")
+ .set_description("Lowest level to include is cluster log file")
+ .add_see_also("mon_cluster_log_file"),
Option("mon_cluster_log_to_graylog", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("false")
- .set_description(""),
+ .add_service("mon")
+ .set_description("Make monitor send cluster log to graylog"),
Option("mon_cluster_log_to_graylog_host", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("127.0.0.1")
- .set_description(""),
+ .add_service("mon")
+ .set_description("Graylog host for cluster log messages")
+ .add_see_also("mon_cluster_log_to_graylog"),
Option("mon_cluster_log_to_graylog_port", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("12201")
- .set_description(""),
+ .add_service("mon")
+ .set_description("Graylog port for cluster log messages")
+ .add_see_also("mon_cluster_log_to_graylog"),
Option("enable_experimental_unrecoverable_data_corrupting_features", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description(""),
+ .set_description("Enable named (or all with '*') experimental features that may be untested, dangerous, and/or cause permanent data loss"),
Option("plugin_dir", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default(CEPH_PKGLIBDIR)
- .set_description("")
- .set_safe(),
+ .add_service({"mon", "osd"})
+ .set_description("Base directory for dynamically loaded plugins"),
+ // XIO
Option("xio_trace_mempool", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
.set_description(""),
.set_default(128)
.set_description(""),
- Option("xio_mp_max_64", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("xio_mp_max_64", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(65536)
.set_description(""),
- Option("xio_mp_max_256", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("xio_mp_max_256", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(8192)
.set_description(""),
- Option("xio_mp_max_1k", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("xio_mp_max_1k", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(8192)
.set_description(""),
- Option("xio_mp_max_page", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("xio_mp_max_page", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(4096)
.set_description(""),
- Option("xio_mp_max_hint", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("xio_mp_max_hint", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(4096)
.set_description(""),
.set_default("rdma")
.set_description(""),
- Option("xio_max_send_inline", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("xio_max_send_inline", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(512)
.set_description(""),
+ // Compressor
Option("compressor_zlib_isal", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("Use Intel ISA-L accelerated zlib implementation if available"),
Option("compressor_zlib_level", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("Zlib compression level to use"),
- Option("async_compressor_enabled", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("qat_compressor_enabled", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
-
- Option("async_compressor_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("snappy")
- .set_description(""),
-
- Option("async_compressor_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(2)
- .set_description(""),
-
- Option("async_compressor_thread_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(5)
- .set_description(""),
-
- Option("async_compressor_thread_suicide_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(30)
- .set_description(""),
+ .set_description("Enable Intel QAT acceleration support for compression if available"),
Option("plugin_crypto_accelerator", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("crypto_isal")
- .set_description(""),
+ .set_description("Crypto accelerator library to use"),
- Option("mempool_debug", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("mempool_debug", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.set_description(""),
Option("key", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
.set_description("Authentication key")
.set_long_description("A CephX authentication key, base64 encoded. It normally looks something like 'AQAtut9ZdMbNJBAAHz6yBAWyJyz2yYRyeMWDag=='.")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_see_also("keyfile")
.add_see_also("keyring"),
.set_default("")
.set_description("Path to a file containing a key")
.set_long_description("The file should contain a CephX authentication key and optionally a trailing newline, but nothing else.")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_see_also("key"),
Option("keyring", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default(
"/etc/ceph/$cluster.$name.keyring,/etc/ceph/$cluster.keyring,"
- "/etc/ceph/keyring,/etc/ceph/keyring.bin,"
+ "/etc/ceph/keyring,/etc/ceph/keyring.bin,"
#if defined(__FreeBSD)
"/usr/local/etc/ceph/$cluster.$name.keyring,"
"/usr/local/etc/ceph/$cluster.keyring,"
- "/usr/local/etc/ceph/keyring,/usr/local/etc/ceph/keyring.bin,"
+ "/usr/local/etc/ceph/keyring,/usr/local/etc/ceph/keyring.bin,"
#endif
)
.set_description("Path to a keyring file.")
.set_long_description("A keyring file is an INI-style formatted file where the section names are client or daemon names (e.g., 'osd.0') and each section contains a 'key' property with CephX authentication key as the value.")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_see_also("key")
.add_see_also("keyfile"),
Option("heartbeat_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("Frequency of internal heartbeat checks (seconds)"),
Option("heartbeat_file", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description(""),
+ .set_description("File to touch on successful internal heartbeat")
+ .set_long_description("If set, this file will be touched every time an internal heartbeat check succeeds.")
+ .add_see_also("heartbeat_interval"),
Option("heartbeat_inject_failure", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
Option("perf", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("Enable internal performance metrics")
+ .set_long_description("If enabled, collect and expose internal health metrics"),
Option("ms_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("async+posix")
- .set_description("")
- .set_safe(),
+ .set_description("Messenger implementation to use for network communication"),
Option("ms_public_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description(""),
+ .set_description("Messenger implementation to use for the public network")
+ .set_long_description("If not specified, use ms_type")
+ .add_see_also("ms_type"),
Option("ms_cluster_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description(""),
+ .set_description("Messenger implementation to use for the internal cluster network")
+ .set_long_description("If not specified, use ms_type")
+ .add_see_also("ms_type"),
+
+ Option("ms_mon_cluster_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
+ .set_default("crc")
+ .set_description("Connection modes (crc, secure) for intra-mon connections in order of preference")
+ .add_see_also("ms_mon_service_mode")
+ .add_see_also("ms_mon_client_mode")
+ .add_see_also("ms_service_mode")
+ .add_see_also("ms_cluster_mode")
+ .add_see_also("ms_client_mode"),
+
+ Option("ms_mon_service_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
+ .set_default("crc")
+ .set_description("Allowed connection modes (crc, secure) for connections to mons")
+ .add_see_also("ms_service_mode")
+ .add_see_also("ms_mon_cluster_mode")
+ .add_see_also("ms_mon_client_mode")
+ .add_see_also("ms_cluster_mode")
+ .add_see_also("ms_client_mode"),
+
+ Option("ms_mon_client_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
+ .set_default("crc")
+ .set_description("Connection modes (crc, secure) for connections from clients to monitors in order of preference")
+ .add_see_also("ms_mon_service_mode")
+ .add_see_also("ms_mon_cluster_mode")
+ .add_see_also("ms_service_mode")
+ .add_see_also("ms_cluster_mode")
+ .add_see_also("ms_client_mode"),
+
+ Option("ms_cluster_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
+ .set_default("crc")
+ .set_description("Connection modes (crc, secure) for intra-cluster connections in order of preference")
+ .add_see_also("ms_service_mode")
+ .add_see_also("ms_client_mode"),
+
+ Option("ms_service_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
+ .set_default("crc")
+ .set_description("Allowed connection modes (crc, secure) for connections to daemons")
+ .add_see_also("ms_cluster_mode")
+ .add_see_also("ms_client_mode"),
+
+ Option("ms_client_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
+ .set_default("crc")
+ .set_description("Connection modes (crc, secure) for connections from clients in order of preference")
+ .add_see_also("ms_cluster_mode")
+ .add_see_also("ms_service_mode"),
Option("ms_tcp_nodelay", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("Disable Nagle's algorithm and send queued network traffic immediately"),
- Option("ms_tcp_rcvbuf", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("ms_tcp_rcvbuf", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description(""),
+ .set_description("Size of TCP socket receive buffer"),
- Option("ms_tcp_prefetch_max_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("ms_tcp_prefetch_max_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(4_K)
- .set_description(""),
+ .set_description("Maximum amount of data to prefetch out of the socket receive buffer"),
Option("ms_initial_backoff", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.2)
- .set_description(""),
+ .set_description("Initial backoff after a network error is detected (seconds)"),
Option("ms_max_backoff", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(15.0)
- .set_description(""),
+ .set_description("Maximum backoff after a network error before retrying (seconds)")
+ .add_see_also("ms_initial_backoff"),
- Option("ms_crc_data", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("ms_crc_data", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
- .set_description(""),
+ .set_description("Set and/or verify crc32c checksum on data payload sent over network"),
- Option("ms_crc_header", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("ms_crc_header", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
- .set_description(""),
+ .set_description("Set and/or verify crc32c checksum on header payload sent over network"),
- Option("ms_die_on_bad_msg", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("ms_die_on_bad_msg", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .set_description("Induce a daemon crash/exit when a bad network message is received"),
- Option("ms_die_on_unhandled_msg", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("ms_die_on_unhandled_msg", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .set_description("Induce a daemon crash/exit when an unrecognized message is received"),
- Option("ms_die_on_old_message", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("ms_die_on_old_message", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .set_description("Induce a daemon crash/exit when a old, undecodable message is received"),
- Option("ms_die_on_skipped_message", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("ms_die_on_skipped_message", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .set_description("Induce a daemon crash/exit if sender skips a message sequence number"),
- Option("ms_dispatch_throttle_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(100 << 20)
- .set_description(""),
+ Option("ms_die_on_bug", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ .set_default(false)
+ .set_description("Induce a crash/exit on various bugs (for testing purposes)"),
+
+ Option("ms_dispatch_throttle_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(100_M)
+ .set_description("Limit messages that are read off the network but still being processed"),
+
+ Option("ms_msgr2_sign_messages", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description("Sign msgr2 frames' payload")
+ .add_see_also("ms_msgr2_encrypt_messages"),
+
+ Option("ms_msgr2_encrypt_messages", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description("Encrypt msgr2 frames' payload")
+ .add_see_also("ms_msgr2_sign_messages"),
+
+ Option("ms_bind_ipv4", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .set_description("Bind servers to IPv4 address(es)")
+ .add_see_also("ms_bind_ipv6"),
Option("ms_bind_ipv6", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("Bind servers to IPv6 address(es)")
+ .add_see_also("ms_bind_ipv4"),
+
+ Option("ms_bind_prefer_ipv4", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description("Prefer IPV4 over IPV6 address(es)"),
+
+ Option("ms_bind_msgr1", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .set_description("Bind servers to msgr1 (legacy) protocol address(es)")
+ .add_see_also("ms_bind_msgr2"),
+
+ Option("ms_bind_msgr2", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .set_description("Bind servers to msgr2 (nautilus+) protocol address(es)")
+ .add_see_also("ms_bind_msgr1"),
Option("ms_bind_port_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(6800)
- .set_description(""),
+ .set_description("Lowest port number to bind daemon(s) to"),
Option("ms_bind_port_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(7300)
- .set_description(""),
+ .set_description("Highest port number to bind daemon(s) to"),
Option("ms_bind_retry_count", Option::TYPE_INT, Option::LEVEL_ADVANCED)
#if !defined(__FreeBSD__)
// FreeBSD does not use SO_REAUSEADDR so allow for a bit more time per default
.set_default(6)
#endif
- .set_description(""),
+ .set_description("Number of attempts to make while bind(2)ing to a port"),
Option("ms_bind_retry_delay", Option::TYPE_INT, Option::LEVEL_ADVANCED)
#if !defined(__FreeBSD__)
// FreeBSD does not use SO_REAUSEADDR so allow for a bit more time per default
.set_default(6)
#endif
- .set_description(""),
+ .set_description("Delay between bind(2) attempts (seconds)"),
Option("ms_bind_before_connect", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("Call bind(2) on client sockets"),
Option("ms_tcp_listen_backlog", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(512)
- .set_description(""),
+ .set_description("Size of queue of incoming connections for accept(2)"),
- Option("ms_rwthread_stack_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("ms_rwthread_stack_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1_M)
- .set_description(""),
+ .set_description("Size of stack for SimpleMessenger read/write threads"),
Option("ms_tcp_read_timeout", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(900)
- .set_description(""),
+ .set_description("Time before an idle TCP connection is closed (seconds)"),
- Option("ms_pq_max_tokens_per_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("ms_pq_max_tokens_per_priority", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(16777216)
.set_description(""),
- Option("ms_pq_min_cost", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("ms_pq_min_cost", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(65536)
.set_description(""),
Option("ms_inject_socket_failures", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(0)
- .set_description(""),
+ .set_description("Inject a socket failure every Nth socket operation"),
Option("ms_inject_delay_type", Option::TYPE_STR, Option::LEVEL_DEV)
.set_default("")
- .set_description("")
- .set_safe(),
+ .set_description("Entity type to inject delays for")
+ .set_flag(Option::FLAG_RUNTIME),
Option("ms_inject_delay_msg_type", Option::TYPE_STR, Option::LEVEL_DEV)
.set_default("")
- .set_description(""),
+ .set_description("Message type to inject delays for"),
Option("ms_inject_delay_max", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(1)
- .set_description(""),
+ .set_description("Max delay to inject"),
Option("ms_inject_delay_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0)
Option("ms_inject_internal_delays", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0)
- .set_description(""),
+ .set_description("Inject various internal delays to induce races (seconds)"),
Option("ms_dump_on_send", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("Hexdump message to debug log on message send"),
Option("ms_dump_corrupt_message_level", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1)
- .set_description(""),
+ .set_description("Log level at which to hexdump corrupt messages we receive"),
Option("ms_async_op_threads", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(3)
- .set_description(""),
+ .set_min_max(1, 24)
+ .set_description("Threadpool size for AsyncMessenger (ms_type=async)"),
Option("ms_async_max_op_threads", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
-
- Option("ms_async_set_affinity", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(true)
- .set_description(""),
-
- Option("ms_async_affinity_cores", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("")
- .set_description(""),
+ .set_description("Maximum threadpool size of AsyncMessenger")
+ .add_see_also("ms_async_op_threads"),
Option("ms_async_rdma_device_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
.set_default(false)
.set_description(""),
- Option("ms_async_rdma_buffer_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("ms_async_rdma_buffer_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(128_K)
.set_description(""),
.set_description(""),
Option("ms_async_rdma_receive_buffers", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(1024)
+ .set_default(32768)
+ .set_description(""),
+
+ Option("ms_async_rdma_receive_queue_len", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(4096)
+ .set_description(""),
+
+ Option("ms_async_rdma_support_srq", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
.set_description(""),
Option("ms_async_rdma_port_num", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_description("The maximum number of consecutive failed accept() calls before "
"considering the daemon is misconfigured and abort it."),
+ Option("ms_async_rdma_cm", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description(""),
+
+ Option("ms_async_rdma_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("ib")
+ .set_description(""),
+
Option("ms_dpdk_port_id", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description(""),
Option("ms_dpdk_coremask", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("1")
+ .set_default("0xF") //begin with 0x for the string
.set_description("")
- .set_safe(),
+ .add_see_also("ms_async_op_threads"),
Option("ms_dpdk_memory_channel", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("4")
Option("ms_dpdk_host_ipv4_addr", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description("")
- .set_safe(),
+ .set_description(""),
Option("ms_dpdk_gateway_ipv4_addr", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description("")
- .set_safe(),
+ .set_description(""),
Option("ms_dpdk_netmask_ipv4_addr", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description("")
- .set_safe(),
+ .set_description(""),
Option("ms_dpdk_lro", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
Option("inject_early_sigterm", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .set_description("send ourselves a SIGTERM early during startup"),
+
+ // MON
+ Option("mon_enable_op_tracker", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .add_service("mon")
+ .set_description("enable/disable MON op tracking"),
+
+ Option("mon_op_complaint_time", Option::TYPE_SECS, Option::LEVEL_ADVANCED)
+ .set_default(30)
+ .add_service("mon")
+ .set_description("time after which to consider a monitor operation blocked "
+ "after no updates"),
+
+ Option("mon_op_log_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(5)
+ .add_service("mon")
+ .set_description("max number of slow ops to display"),
+
+ Option("mon_op_history_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(20)
+ .add_service("mon")
+ .set_description("max number of completed ops to track"),
+
+ Option("mon_op_history_duration", Option::TYPE_SECS, Option::LEVEL_ADVANCED)
+ .set_default(600)
+ .add_service("mon")
+ .set_description("expiration time in seconds of historical MON OPS"),
+
+ Option("mon_op_history_slow_op_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(20)
+ .add_service("mon")
+ .set_description("max number of slow historical MON OPS to keep"),
+
+ Option("mon_op_history_slow_op_threshold", Option::TYPE_SECS, Option::LEVEL_ADVANCED)
+ .set_default(10)
+ .add_service("mon")
+ .set_description("duration of an op to be considered as a historical slow op"),
Option("mon_data", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.set_default("/var/lib/ceph/mon/$cluster-$id")
- .set_description(""),
+ .add_service("mon")
+ .set_description("path to mon database"),
Option("mon_initial_members", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
+ .add_service("mon")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
+ .set_flag(Option::FLAG_CLUSTER_CREATE)
.set_description(""),
Option("mon_compact_on_start", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
+ .add_service("mon")
.set_description(""),
Option("mon_compact_on_bootstrap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
+ .add_service("mon")
.set_description(""),
Option("mon_compact_on_trim", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
+ .add_service("mon")
.set_description(""),
+ /* -- mon: osdmap prune (begin) -- */
+ Option("mon_osdmap_full_prune_enabled", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .add_service("mon")
+ .set_description("enables pruning full osdmap versions when we go over a given number of maps")
+ .add_see_also("mon_osdmap_full_prune_min")
+ .add_see_also("mon_osdmap_full_prune_interval")
+ .add_see_also("mon_osdmap_full_prune_txsize"),
+
+ Option("mon_osdmap_full_prune_min", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(10000)
+ .add_service("mon")
+ .set_description("minimum number of versions in the store to trigger full map pruning")
+ .add_see_also("mon_osdmap_full_prune_enabled")
+ .add_see_also("mon_osdmap_full_prune_interval")
+ .add_see_also("mon_osdmap_full_prune_txsize"),
+
+ Option("mon_osdmap_full_prune_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(10)
+ .add_service("mon")
+ .set_description("interval between maps that will not be pruned; maps in the middle will be pruned.")
+ .add_see_also("mon_osdmap_full_prune_enabled")
+ .add_see_also("mon_osdmap_full_prune_interval")
+ .add_see_also("mon_osdmap_full_prune_txsize"),
+
+ Option("mon_osdmap_full_prune_txsize", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(100)
+ .add_service("mon")
+ .set_description("number of maps we will prune per iteration")
+ .add_see_also("mon_osdmap_full_prune_enabled")
+ .add_see_also("mon_osdmap_full_prune_interval")
+ .add_see_also("mon_osdmap_full_prune_txsize"),
+ /* -- mon: osdmap prune (end) -- */
+
Option("mon_osd_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(500)
- .set_description(""),
+ .add_service("mon")
+ .set_description("maximum number of OSDMaps to cache in memory"),
Option("mon_cpu_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(4)
- .set_description(""),
+ .add_service("mon")
+ .set_description("worker threads for CPU intensive background work"),
- Option("mon_osd_mapping_pgs_per_chunk", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mon_osd_mapping_pgs_per_chunk", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(4096)
- .set_description(""),
+ .add_service("mon")
+ .set_description("granularity of PG placement calculation background work"),
Option("mon_osd_max_creating_pgs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1024)
- .set_description(""),
+ .add_service("mon")
+ .set_description("maximum number of PGs the mon will create at once"),
+
+ Option("mon_osd_max_initial_pgs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(1024)
+ .add_service("mon")
+ .set_description("maximum number of PGs a pool will created with")
+ .set_long_description("If the user specifies more PGs than this, the cluster will subsequently split PGs after the pool is created in order to reach the target."),
Option("mon_tick_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .add_service("mon")
+ .set_description("interval for internal mon background checks"),
Option("mon_session_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(300)
- .set_description(""),
+ .add_service("mon")
+ .set_description("close inactive mon client connections after this many seconds"),
- Option("mon_subscribe_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mon_subscribe_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(1_day)
- .set_description(""),
+ .add_service("mon")
+ .set_description("subscribe interval for pre-jewel clients"),
Option("mon_delta_reset_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(10)
- .set_description(""),
-
+ .add_service("mon")
+ .add_service("mon")
+ .set_description("window duration for rate calculations in 'ceph status'"),
+
Option("mon_osd_laggy_halflife", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1_hr)
- .set_description(""),
+ .add_service("mon")
+ .set_description("halflife of OSD 'lagginess' factor"),
Option("mon_osd_laggy_weight", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.3)
- .set_description(""),
+ .set_min_max(0.0, 1.0)
+ .add_service("mon")
+ .set_description("how heavily to weight OSD marking itself back up in overall laggy_probability")
+ .set_long_description("1.0 means that an OSD marking itself back up (because it was marked down but not actually dead) means a 100% laggy_probability; 0.0 effectively disables tracking of laggy_probability."),
Option("mon_osd_laggy_max_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(300)
- .set_description(""),
+ .add_service("mon")
+ .set_description("cap value for period for OSD to be marked for laggy_interval calculation"),
Option("mon_osd_adjust_heartbeat_grace", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .add_service("mon")
+ .set_description("increase OSD heartbeat grace if peers appear to be laggy")
+ .set_long_description("If an OSD is marked down but then marks itself back up, it implies it wasn't actually down but was unable to respond to heartbeats. If this option is true, we can use the laggy_probability and laggy_interval values calculated to model this situation to increase the heartbeat grace period for this OSD so that it isn't marked down again. laggy_probability is an estimated probability that the given OSD is down because it is laggy (not actually down), and laggy_interval is an estiate on how long it stays down when it is laggy.")
+ .add_see_also("mon_osd_laggy_halflife")
+ .add_see_also("mon_osd_laggy_weight")
+ .add_see_also("mon_osd_laggy_max_interval"),
Option("mon_osd_adjust_down_out_interval", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .add_service("mon")
+ .set_description("increase the mon_osd_down_out_interval if an OSD appears to be laggy")
+ .add_see_also("mon_osd_adjust_heartbeat_grace"),
Option("mon_osd_auto_mark_in", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_description("mark any OSD that comes up 'in'"),
Option("mon_osd_auto_mark_auto_out_in", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .add_service("mon")
+ .set_description("mark any OSD that comes up that was automatically marked 'out' back 'in'")
+ .add_see_also("mon_osd_down_out_interval"),
Option("mon_osd_auto_mark_new_in", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .add_service("mon")
+ .set_description("mark any new OSD that comes up 'in'"),
Option("mon_osd_destroyed_out_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(600)
- .set_description(""),
+ .add_service("mon")
+ .set_description("mark any OSD 'out' that has been 'destroy'ed for this long (seconds)"),
Option("mon_osd_down_out_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(600)
- .set_description(""),
+ .add_service("mon")
+ .set_description("mark any OSD 'out' that has been 'down' for this long (seconds)"),
Option("mon_osd_down_out_subtree_limit", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("rack")
- .set_description(""),
+ .set_flag(Option::FLAG_RUNTIME)
+ .add_service("mon")
+ .set_description("do not automatically mark OSDs 'out' if an entire subtree of this size is down")
+ .add_see_also("mon_osd_down_out_interval"),
Option("mon_osd_min_up_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.3)
- .set_description(""),
+ .add_service("mon")
+ .set_description("do not automatically mark OSDs 'out' if fewer than this many OSDs are 'up'")
+ .add_see_also("mon_osd_down_out_interval"),
Option("mon_osd_min_in_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.75)
- .set_description(""),
+ .add_service("mon")
+ .set_description("do not automatically mark OSDs 'out' if fewer than this many OSDs are 'in'")
+ .add_see_also("mon_osd_down_out_interval"),
Option("mon_osd_warn_op_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(32)
- .set_description(""),
+ .add_service("mgr")
+ .set_description("issue REQUEST_SLOW health warning if OSD ops are slower than this age (seconds)"),
Option("mon_osd_err_op_age_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(128)
- .set_description(""),
-
- Option("mon_osd_max_split_count", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(32)
- .set_description(""),
-
- Option("mon_osd_allow_primary_temp", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
- .set_description(""),
-
- Option("mon_osd_allow_primary_affinity", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
- .set_description(""),
+ .add_service("mgr")
+ .set_description("issue REQUEST_STUCK health error if OSD ops are slower than is age (seconds)"),
- Option("mon_osd_prime_pg_temp", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("mon_osd_prime_pg_temp", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
- .set_description(""),
+ .add_service("mon")
+ .set_description("minimize peering work by priming pg_temp values after a map change"),
- Option("mon_osd_prime_pg_temp_max_time", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mon_osd_prime_pg_temp_max_time", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(.5)
- .set_description(""),
+ .add_service("mon")
+ .set_description("maximum time to spend precalculating PG mappings on map change (seconds)"),
Option("mon_osd_prime_pg_temp_max_estimate", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.25)
- .set_description(""),
-
- Option("mon_osd_pool_ec_fast_read", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_description("calculate all PG mappings if estimated fraction of PGs that change is above this amount"),
Option("mon_stat_smooth_intervals", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(6)
Option("mon_election_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .add_service("mon")
+ .set_description("maximum time for a mon election (seconds)"),
Option("mon_lease", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .add_service("mon")
+ .set_description("lease interval between quorum monitors (seconds)")
+ .set_long_description("This setting controls how sensitive your mon quorum is to intermittent network issues or other failures."),
Option("mon_lease_renew_interval_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.6)
- .set_description(""),
+ .set_min_max((double)0.0, (double).9999999)
+ .add_service("mon")
+ .set_description("multiple of mon_lease for the lease renewal interval")
+ .set_long_description("Leases must be renewed before they time out. A smaller value means frequent renewals, while a value close to 1 makes a lease expiration more likely.")
+ .add_see_also("mon_lease"),
Option("mon_lease_ack_timeout_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(2.0)
- .set_description(""),
+ .set_min_max(1.0001, 100.0)
+ .add_service("mon")
+ .set_description("multiple of mon_lease for the lease ack interval before calling new election")
+ .add_see_also("mon_lease"),
Option("mon_accept_timeout_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(2.0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("multiple of mon_lease for follower mons to accept proposed state changes before calling a new election")
+ .add_see_also("mon_lease"),
Option("mon_clock_drift_allowed", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.050)
- .set_description(""),
+ .add_service("mon")
+ .set_description("allowed clock drift (in seconds) between mons before issuing a health warning"),
Option("mon_clock_drift_warn_backoff", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .add_service("mon")
+ .set_description("exponential backoff factor for logging clock drift warnings in the cluster log"),
Option("mon_timecheck_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(300.0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("frequency of clock synchronization checks between monitors (seconds)"),
Option("mon_timecheck_skew_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(30.0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("frequency of clock synchronization (re)checks between monitors while clocks are believed to be skewed (seconds)")
+ .add_see_also("mon_timecheck_interval"),
Option("mon_pg_stuck_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(60)
.set_long_description("see doc/control.rst under dump_stuck for more info")
.add_service("mgr"),
- Option("mon_pg_min_inactive", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(1)
- .set_description(""),
-
Option("mon_pg_warn_min_per_osd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(30)
+ .add_service("mgr")
.set_description("minimal number PGs per (in) osd before we warn the admin"),
Option("mon_max_pg_per_osd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_min(1)
.set_default(250)
- .set_description("Max number of PGs per OSD the cluster will allow"),
+ .add_service("mgr")
+ .set_description("Max number of PGs per OSD the cluster will allow")
+ .set_long_description("If the number of PGs per OSD exceeds this, a "
+ "health warning will be visible in `ceph status`. This is also used "
+ "in automated PG management, as the threshold at which some pools' "
+ "pg_num may be shrunk in order to enable increasing the pg_num of "
+ "others."),
+
+ Option("mon_target_pg_per_osd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_min(1)
+ .set_default(100)
+ .set_description("Automated PG management creates this many PGs per OSD")
+ .set_long_description("When creating pools, the automated PG management "
+ "logic will attempt to reach this target. In some circumstances, it "
+ "may exceed this target, up to the ``mon_max_pg_per_osd`` limit. "
+ "Conversely, a lower number of PGs per OSD may be created if the "
+ "cluster is not yet fully utilised"),
Option("mon_pg_warn_max_object_skew", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(10.0)
Option("mon_cache_target_full_warn_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.66)
- .set_description(""),
+ .add_service("mgr")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
+ .set_flag(Option::FLAG_CLUSTER_CREATE)
+ .set_description("issue CACHE_POOL_NEAR_FULL health warning when cache pool utilization exceeds this ratio of usable space"),
Option("mon_osd_full_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.95)
- .set_description(""),
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
+ .set_flag(Option::FLAG_CLUSTER_CREATE)
+ .set_description("full ratio of OSDs to be set during initial creation of the cluster"),
Option("mon_osd_backfillfull_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.90)
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
+ .set_flag(Option::FLAG_CLUSTER_CREATE)
.set_description(""),
Option("mon_osd_nearfull_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.85)
- .set_description(""),
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
+ .set_flag(Option::FLAG_CLUSTER_CREATE)
+ .set_description("nearfull ratio for OSDs to be set during initial creation of cluster"),
Option("mon_osd_initial_require_min_compat_client", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("jewel")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
+ .set_flag(Option::FLAG_CLUSTER_CREATE)
.set_description(""),
Option("mon_allow_pool_delete", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_description("allow pool deletions"),
Option("mon_fake_pool_delete", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_description("fake pool deletions by renaming the rados pool"),
Option("mon_globalid_prealloc", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(10000)
- .set_description(""),
+ .add_service("mon")
+ .set_description("number of globalid values to preallocate")
+ .set_long_description("This setting caps how many new clients can authenticate with the cluster before the monitors have to perform a write to preallocate more. Large values burn through the 64-bit ID space more quickly."),
Option("mon_osd_report_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(900)
- .set_description(""),
+ .add_service("mon")
+ .set_description("time before OSDs who do not report to the mons are marked down (seconds)"),
- Option("mon_force_standby_active", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("mon_warn_on_msgr2_not_enabled", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .add_service("mon")
+ .set_description("issue MON_MSGR2_NOT_ENABLED health warning if monitors are all running Nautilus but not all binding to a msgr2 port")
+ .add_see_also("ms_bind_msgr2"),
Option("mon_warn_on_legacy_crush_tunables", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .add_service("mgr")
+ .set_description("issue OLD_CRUSH_TUNABLES health warning if CRUSH tunables are older than mon_crush_min_required_version")
+ .add_see_also("mon_crush_min_required_version"),
Option("mon_crush_min_required_version", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("firefly")
- .set_description(""),
+ .add_service("mgr")
+ .set_description("minimum ceph release to use for mon_warn_on_legacy_crush_tunables")
+ .add_see_also("mon_warn_on_legacy_crush_tunables"),
Option("mon_warn_on_crush_straw_calc_version_zero", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .add_service("mgr")
+ .set_description("issue OLD_CRUSH_STRAW_CALC_VERSION health warning if the CRUSH map's straw_calc_version is zero"),
Option("mon_warn_on_osd_down_out_interval_zero", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .add_service("mgr")
+ .set_description("issue OSD_NO_DOWN_OUT_INTERVAL health warning if mon_osd_down_out_interval is zero")
+ .set_long_description("Having mon_osd_down_out_interval set to 0 means that down OSDs are not marked out automatically and the cluster does not heal itself without administrator intervention.")
+ .add_see_also("mon_osd_down_out_interval"),
Option("mon_warn_on_cache_pools_without_hit_sets", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .add_service("mgr")
+ .set_description("issue CACHE_POOL_NO_HIT_SET health warning for cache pools that do not have hit sets configured"),
Option("mon_warn_on_pool_no_app", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
- .set_description("Enable POOL_APP_NOT_ENABLED health check"),
+ .add_service("mgr")
+ .set_description("issue POOL_APP_NOT_ENABLED health warning if pool has not application enabled"),
- Option("mon_min_osdmap_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(500)
- .set_description(""),
+ Option("mon_warn_on_misplaced", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .add_service("mgr")
+ .set_description("Issue a health warning if there are misplaced objects"),
+
+ Option("mon_max_snap_prune_per_epoch", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(100)
+ .add_service("mon")
+ .set_description("max number of pruned snaps we will process in a single OSDMap epoch"),
- Option("mon_max_pgmap_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mon_min_osdmap_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(500)
- .set_description(""),
+ .add_service("mon")
+ .set_description("min number of OSDMaps to store"),
Option("mon_max_log_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(500)
- .set_description(""),
+ .add_service("mon")
+ .set_description("max number of past cluster log epochs to store"),
Option("mon_max_mdsmap_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(500)
- .set_description(""),
+ .add_service("mon")
+ .set_description("max number of FSMaps/MDSMaps to store"),
Option("mon_max_mgrmap_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(500)
- .set_description(""),
+ .add_service("mon")
+ .set_description("max number of MgrMaps to store"),
Option("mon_max_osd", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(10000)
- .set_description(""),
+ .add_service("mon")
+ .set_description("max number of OSDs in a cluster"),
Option("mon_probe_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(2.0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("timeout for querying other mons during bootstrap pre-election phase (seconds)"),
- Option("mon_client_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mon_client_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(100ul << 20)
- .set_description(""),
+ .add_service("mon")
+ .set_description("max bytes of outstanding client messages mon will read off the network"),
+
+ Option("mon_daemon_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(400ul << 20)
+ .add_service("mon")
+ .set_description("max bytes of outstanding mon messages mon will read off the network"),
Option("mon_mgr_proxy_client_bytes_ratio", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(.3)
+ .add_service("mon")
.set_description("ratio of mon_client_bytes that can be consumed by "
"proxied mgr commands before we error out to client"),
Option("mon_log_max_summary", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(50)
- .set_description(""),
-
- Option("mon_daemon_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(400ul << 20)
- .set_description(""),
+ .add_service("mon")
+ .set_description("number of recent cluster log messages to retain"),
Option("mon_max_log_entries_per_event", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(4096)
- .set_description(""),
+ .add_service("mon")
+ .set_description("max cluster log entries per paxos event"),
Option("mon_reweight_min_pgs_per_osd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(10)
+ .add_service("mgr")
.set_description(""),
- Option("mon_reweight_min_bytes_per_osd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mon_reweight_min_bytes_per_osd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(100_M)
+ .add_service("mgr")
.set_description(""),
Option("mon_reweight_max_osds", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(4)
+ .add_service("mgr")
.set_description(""),
Option("mon_reweight_max_change", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0.05)
- .set_description(""),
-
- Option("mon_health_data_update_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
- .set_default(60.0)
+ .add_service("mgr")
.set_description(""),
Option("mon_health_to_clog", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .add_service("mon")
+ .set_description("log monitor health to cluster log"),
Option("mon_health_to_clog_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1_hr)
- .set_description(""),
+ .add_service("mon")
+ .set_description("frequency to log monitor health to cluster log")
+ .add_see_also("mon_health_to_clog"),
- Option("mon_health_to_clog_tick_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mon_health_to_clog_tick_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(60.0)
+ .add_service("mon")
.set_description(""),
- Option("mon_health_preluminous_compat", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
- .set_description("Include health warnings in preluminous JSON fields"),
-
- Option("mon_health_preluminous_compat_warning", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(true)
- .set_description("Warn about the health JSON format change in preluminous JSON fields"),
-
Option("mon_health_max_detail", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(50)
+ .add_service("mon")
.set_description("max detailed pgs to report in health detail"),
Option("mon_health_log_update_period", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(5)
- .set_description("Minimum time in seconds between log messages about "
+ .add_service("mon")
+ .set_description("minimum time in seconds between log messages about "
"each health check")
.set_min(0),
Option("mon_data_avail_crit", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .add_service("mon")
+ .set_description("issue MON_DISK_CRIT health error when mon available space below this percentage"),
Option("mon_data_avail_warn", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(30)
- .set_description(""),
+ .add_service("mon")
+ .set_description("issue MON_DISK_LOW health warning when mon available space below this percentage"),
- Option("mon_data_size_warn", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mon_data_size_warn", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(15_G)
- .set_description(""),
+ .add_service("mon")
+ .set_description("issue MON_DISK_BIG health warning when mon database is above this size"),
- Option("mon_warn_not_scrubbed", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(0)
- .set_description(""),
+ Option("mon_warn_pg_not_scrubbed_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.5)
+ .set_min(0)
+ .set_description("Percentage of the scrub max interval past the scrub max interval to warn")
+ .set_long_description("")
+ .add_see_also("osd_scrub_max_interval"),
- Option("mon_warn_not_deep_scrubbed", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(0)
- .set_description(""),
+ Option("mon_warn_pg_not_deep_scrubbed_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.75)
+ .set_min(0)
+ .set_description("Percentage of the deep scrub interval past the deep scrub interval to warn")
+ .set_long_description("")
+ .add_see_also("osd_deep_scrub_interval"),
Option("mon_scrub_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1_day)
- .set_description(""),
+ .add_service("mon")
+ .set_description("frequency for scrubbing mon database"),
Option("mon_scrub_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(5_min)
- .set_description(""),
+ .add_service("mon")
+ .set_description("timeout to restart scrub of mon quorum participant does not respond for the latest chunk"),
Option("mon_scrub_max_keys", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(100)
- .set_description(""),
+ .add_service("mon")
+ .set_description("max keys per on scrub chunk/step"),
Option("mon_scrub_inject_crc_mismatch", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0.0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("probability for injecting crc mismatches into mon scrub"),
Option("mon_scrub_inject_missing_keys", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0.0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("probability for injecting missing keys into mon scrub"),
- Option("mon_config_key_max_entry_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(4_K)
- .set_description(""),
+ Option("mon_config_key_max_entry_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(64_K)
+ .add_service("mon")
+ .set_description("Defines the number of bytes allowed to be held in a "
+ "single config-key entry"),
Option("mon_sync_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(60.0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("timeout before canceling sync if syncing mon does not respond"),
- Option("mon_sync_max_payload_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mon_sync_max_payload_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1_M)
- .set_description(""),
+ .add_service("mon")
+ .set_description("target max message payload for mon sync"),
- Option("mon_sync_debug", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("mon_sync_debug", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_description("enable extra debugging during mon sync"),
Option("mon_inject_sync_get_chunk_delay", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("inject delay during sync (seconds)"),
- Option("mon_osd_min_down_reporters", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mon_osd_min_down_reporters", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(2)
- .set_description(""),
+ .add_service("mon")
+ .set_description("number of OSDs from different subtrees who need to report a down OSD for it to count")
+ .add_see_also("mon_osd_reporter_subtree_level"),
Option("mon_osd_reporter_subtree_level", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("host")
- .set_description(""),
+ .add_service("mon")
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("in which level of parent bucket the reporters are counted"),
Option("mon_osd_snap_trim_queue_warn_on", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(32768)
+ .add_service("mon")
.set_description("Warn when snap trim queue is that large (or larger).")
.set_long_description("Warn when snap trim queue length for at least one PG crosses this value, as this is indicator of snap trimmer not keeping up, wasting disk space"),
- Option("mon_osd_force_trim_to", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mon_osd_force_trim_to", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("force mons to trim osdmaps through this epoch"),
- Option("mon_mds_force_trim_to", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mon_mds_force_trim_to", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("force mons to trim mdsmaps/fsmaps through this epoch"),
Option("mon_mds_skip_sanity", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_description("skip sanity checks on fsmap/mdsmap"),
- Option("mon_fixup_legacy_erasure_code_profiles", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(true)
- .set_description("Automatically adjust ruleset-* to crush-* so that legacy apps can set modern erasure code profiles without modification"),
+ Option("mon_debug_extra_checks", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ .set_default(false)
+ .add_service("mon")
+ .set_description("Enable some additional monitor checks")
+ .set_long_description(
+ "Enable some additional monitor checks that would be too expensive "
+ "to run on production systems, or would only be relevant while "
+ "testing or debugging."),
+
+ Option("mon_debug_block_osdmap_trim", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ .set_default(false)
+ .add_service("mon")
+ .set_description("Block OSDMap trimming while the option is enabled.")
+ .set_long_description(
+ "Blocking OSDMap trimming may be quite helpful to easily reproduce "
+ "states in which the monitor keeps (hundreds of) thousands of "
+ "osdmaps."),
Option("mon_debug_deprecated_as_obsolete", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_description("treat deprecated mon commands as obsolete"),
Option("mon_debug_dump_transactions", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_description("dump paxos transactions to log")
+ .add_see_also("mon_debug_dump_location"),
Option("mon_debug_dump_json", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_description("dump paxos transasctions to log as json")
+ .add_see_also("mon_debug_dump_transactions"),
Option("mon_debug_dump_location", Option::TYPE_STR, Option::LEVEL_DEV)
.set_default("/var/log/ceph/$cluster-$name.tdump")
- .set_description(""),
+ .add_service("mon")
+ .set_description("file to dump paxos transactions to")
+ .add_see_also("mon_debug_dump_transactions"),
- Option("mon_debug_no_require_luminous", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ Option("mon_debug_no_require_mimic", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_flag(Option::FLAG_CLUSTER_CREATE)
+ .set_description("do not set mimic feature for new mon clusters"),
+
+ Option("mon_debug_no_require_nautilus", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ .set_default(false)
+ .add_service("mon")
+ .set_flag(Option::FLAG_CLUSTER_CREATE)
+ .set_description("do not set nautilus feature for new mon clusters"),
Option("mon_debug_no_require_bluestore_for_ec_overwrites", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_description("do not require bluestore OSDs to enable EC overwrites on a rados pool"),
Option("mon_debug_no_initial_persistent_features", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_flag(Option::FLAG_CLUSTER_CREATE)
+ .set_description("do not set any monmap features for new mon clusters"),
Option("mon_inject_transaction_delay_max", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(10.0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("max duration of injected delay in paxos"),
Option("mon_inject_transaction_delay_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("probability of injecting a delay in paxos"),
+
+ Option("mon_inject_pg_merge_bounce_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
+ .set_default(0)
+ .add_service("mon")
+ .set_description("probability of failing and reverting a pg_num decrement"),
Option("mon_sync_provider_kill_at", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("kill mon sync requester at specific point"),
Option("mon_sync_requester_kill_at", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
- .set_description(""),
+ .add_service("mon")
+ .set_description("kill mon sync requestor at specific point"),
Option("mon_force_quorum_join", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .add_service("mon")
+ .set_description("force mon to rejoin quorum even though it was just removed"),
Option("mon_keyvaluedb", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("rocksdb")
- .set_description(""),
+ .set_enum_allowed({"leveldb", "rocksdb"})
+ .set_flag(Option::FLAG_CREATE)
+ .add_service("mon")
+ .set_description("database backend to use for the mon database"),
Option("mon_debug_unsafe_allow_tier_with_nonempty_snaps", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
+ .add_service("mon")
.set_description(""),
Option("mon_osd_blacklist_default_expire", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1_hr)
+ .add_service("mon")
.set_description("Duration in seconds that blacklist entries for clients "
"remain in the OSD map"),
Option("mon_mds_blacklist_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(1_day)
.set_min(1_hr)
+ .add_service("mon")
.set_description("Duration in seconds that blacklist entries for MDS "
"daemons remain in the OSD map"),
Option("mon_osd_crush_smoke_test", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .add_service("mon")
+ .set_description("perform a smoke test on any new CRUSH map before accepting changes"),
+
+ Option("mon_smart_report_timeout", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(5)
+ .add_service("mon")
+ .set_description("Timeout (in seconds) for smarctl to run, default is set to 5"),
+
+
+ // PAXOS
Option("paxos_stash_full_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(25)
+ .add_service("mon")
.set_description(""),
Option("paxos_max_join_drift", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(10)
+ .add_service("mon")
.set_description(""),
Option("paxos_propose_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1.0)
+ .add_service("mon")
.set_description(""),
Option("paxos_min_wait", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0.05)
+ .add_service("mon")
.set_description(""),
Option("paxos_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(500)
+ .add_service("mon")
.set_description(""),
Option("paxos_trim_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(250)
+ .add_service("mon")
.set_description(""),
Option("paxos_trim_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(500)
+ .add_service("mon")
.set_description(""),
Option("paxos_service_trim_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(250)
+ .add_service("mon")
.set_description(""),
Option("paxos_service_trim_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(500)
+ .add_service("mon")
.set_description(""),
Option("paxos_kill_at", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
+ .add_service("mon")
.set_description(""),
+
+ // AUTH
+
Option("auth_cluster_required", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("cephx")
- .set_description(""),
+ .set_description("authentication methods required by the cluster"),
Option("auth_service_required", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("cephx")
- .set_description(""),
+ .set_description("authentication methods required by service daemons"),
Option("auth_client_required", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("cephx, none")
- .set_description(""),
+ .set_flag(Option::FLAG_MINIMAL_CONF)
+ .set_description("authentication methods allowed by clients"),
Option("auth_supported", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description(""),
+ .set_description("authentication methods required (deprecated)"),
Option("max_rotating_auth_attempts", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(10)
- .set_description(""),
+ .set_description("number of attempts to initialize rotating keys before giving up"),
+
+ Option("rotating_keys_bootstrap_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(30)
+ .set_description("timeout for obtaining rotating keys during bootstrap phase (seconds)"),
+
+ Option("rotating_keys_renewal_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(10)
+ .set_description("timeout for updating rotating keys (seconds)"),
Option("cephx_require_signatures", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
.set_description(""),
Option("mon_client_hunt_parallel", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(2)
+ .set_default(3)
.set_description(""),
Option("mon_client_hunt_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_description(""),
Option("mon_client_hunt_interval_backoff", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
- .set_default(2.0)
+ .set_default(1.5)
.set_description(""),
Option("mon_client_hunt_interval_min_multiple", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1000)
.set_description(""),
- Option("mon_max_pool_pg_num", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mon_max_pool_pg_num", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(65536)
.set_description(""),
.set_default(10)
.set_description(""),
- Option("objecter_tick_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("objecter_tick_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(5.0)
.set_description(""),
Option("objecter_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(10.0)
- .set_description(""),
+ .set_description("Seconds before in-flight op is considered 'laggy' and we query mon for the latest OSDMap"),
- Option("objecter_inflight_op_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("objecter_inflight_op_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(100_M)
- .set_description(""),
+ .set_description("Max in-flight data in bytes (both directions)"),
Option("objecter_inflight_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1024)
- .set_description(""),
+ .set_description("Max in-flight operations"),
- Option("objecter_completion_locks_per_session", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("objecter_completion_locks_per_session", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(32)
.set_description(""),
.set_default(false)
.set_description(""),
- Option("objecter_retry_writes_after_first_reply", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("objecter_retry_writes_after_first_reply", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
Option("filer_max_purge_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(10)
- .set_description(""),
+ .set_description("Max in-flight operations for purging a striped range (e.g., MDS journal)"),
Option("filer_max_truncate_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(128)
- .set_description(""),
+ .set_description("Max in-flight operations for truncating/deleting a striped sequence (e.g., MDS journal)"),
Option("journaler_write_head_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(15)
- .set_description(""),
+ .set_description("Interval in seconds between journal header updates (to help bound replay time)"),
- Option("journaler_prefetch_periods", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ // * journal object size
+ Option("journaler_prefetch_periods", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(10)
- .set_description(""),
+ .set_min(2) // we need at least 2 periods to make progress.
+ .set_description("Number of striping periods to prefetch while reading MDS journal"),
+ // * journal object size
+ Option("journaler_prezero_periods", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(5)
+ // we need to zero at least two periods, minimum, to ensure that we
+ // have a full empty object/period in front of us.
+ .set_min(2)
+ .set_description("Number of striping periods to zero head of MDS journal write position"),
+
+ // -- OSD --
Option("osd_calc_pg_upmaps_aggressively", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("try to calculate PG upmaps more aggressively, e.g., "
"by doing a fairly exhaustive search of existing PGs "
"that can be unmapped or upmapped"),
Option("osd_calc_pg_upmaps_max_stddev", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1.0)
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("standard deviation below which there is no attempt made "
"while trying to calculate PG upmaps"),
Option("osd_calc_pg_upmaps_local_fallback_retries", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(100)
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Maximum number of PGs we can attempt to unmap or upmap "
"for a specific overfull or underfull osd per iteration "),
- Option("journaler_prezero_periods", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_numa_prefer_iface", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .set_flag(Option::FLAG_STARTUP)
+ .set_description("prefer IP on network interface on same numa node as storage")
+ .add_see_also("osd_numa_auto_affinity"),
+
+ Option("osd_numa_auto_affinity", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .set_flag(Option::FLAG_STARTUP)
+ .set_description("automatically set affinity to numa node when storage and network match"),
+
+ Option("osd_numa_node", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(-1)
+ .set_flag(Option::FLAG_STARTUP)
+ .set_description("set affinity to a numa node (-1 for none)")
+ .add_see_also("osd_numa_auto_affinity"),
+
+ Option("osd_smart_report_timeout", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("Timeout (in seconds) for smarctl to run, default is set to 5"),
- Option("osd_check_max_object_name_len_on_startup", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("osd_check_max_object_name_len_on_startup", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
.set_description(""),
Option("osd_max_backfills", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1)
- .set_description(""),
+ .set_description("Maximum number of concurrent local and remote backfills or recoveries per OSD ")
+ .set_long_description("There can be osd_max_backfills local reservations AND the same remote reservations per OSD. So a value of 1 lets this OSD participate as 1 PG primary in recovery and 1 shard of another recovering PG."),
Option("osd_min_recovery_priority", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description(""),
+ .set_description("Minimum priority below which recovery is not performed")
+ .set_long_description("The purpose here is to prevent the cluster from doing *any* lower priority work (e.g., rebalancing) below this threshold and focus solely on higher priority work (e.g., replicating degraded objects)."),
Option("osd_backfill_retry_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(30.0)
- .set_description(""),
+ .set_description("how frequently to retry backfill reservations after being denied (e.g., due to a full OSD)"),
Option("osd_recovery_retry_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(30.0)
- .set_description(""),
+ .set_description("how frequently to retry recovery reservations after being denied (e.g., due to a full OSD)"),
Option("osd_agent_max_ops", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(4)
- .set_description(""),
+ .set_description("maximum concurrent tiering operations for tiering agent"),
Option("osd_agent_max_low_ops", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(2)
- .set_description(""),
+ .set_description("maximum concurrent low-priority tiering operations for tiering agent"),
Option("osd_agent_min_evict_effort", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.1)
- .set_description(""),
+ .set_min_max(0.0, .99)
+ .set_description("minimum effort to expend evicting clean objects"),
Option("osd_agent_quantize_effort", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.1)
- .set_description(""),
+ .set_description("size of quantize unit for eviction effort"),
Option("osd_agent_delay_time", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(5.0)
- .set_description(""),
+ .set_description("how long agent should sleep if it has no work to do"),
- Option("osd_find_best_info_ignore_history_les", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("osd_find_best_info_ignore_history_les", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .set_description("ignore last_epoch_started value when peering AND PROBABLY LOSE DATA")
+ .set_long_description("THIS IS AN EXTREMELY DANGEROUS OPTION THAT SHOULD ONLY BE USED AT THE DIRECTION OF A DEVELOPER. It makes peering ignore the last_epoch_started value when peering, which can allow the OSD to believe an OSD has an authoritative view of a PG's contents even when it is in fact old and stale, typically leading to data loss (by believing a stale PG is up to date)."),
Option("osd_agent_hist_halflife", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1000)
- .set_description(""),
+ .set_description("halflife of agent atime and temp histograms"),
Option("osd_agent_slop", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.02)
- .set_description(""),
+ .set_description("slop factor to avoid switching tiering flush and eviction mode"),
Option("osd_uuid", Option::TYPE_UUID, Option::LEVEL_ADVANCED)
.set_default(uuid_d())
- .set_description(""),
+ .set_flag(Option::FLAG_CREATE)
+ .set_description("uuid label for a new OSD"),
Option("osd_data", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("/var/lib/ceph/osd/$cluster-$id")
- .set_description(""),
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
+ .set_description("path to OSD data"),
Option("osd_journal", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("/var/lib/ceph/osd/$cluster-$id/journal")
- .set_description(""),
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
+ .set_description("path to OSD journal (when FileStore backend is in use)"),
- Option("osd_journal_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_journal_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(5120)
- .set_description(""),
+ .set_flag(Option::FLAG_CREATE)
+ .set_description("size of FileStore journal (in MiB)"),
Option("osd_journal_flush_on_shutdown", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("flush FileStore journal contents during clean OSD shutdown"),
- Option("osd_os_flags", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_os_flags", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(0)
- .set_description(""),
+ .set_description("flags to skip filestore omap or journal initialization"),
- Option("osd_max_write_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_max_write_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_min(4)
.set_default(90)
- .set_description(""),
+ .set_description("Maximum size of a RADOS write operation in megabytes")
+ .set_long_description("This setting prevents clients from doing "
+ "very large writes to RADOS. If you set this to a value "
+ "below what clients expect, they will receive an error "
+ "when attempting to write to the cluster."),
Option("osd_max_pgls", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1024)
- .set_description(""),
+ .set_description("maximum number of results when listing objects in a pool"),
- Option("osd_client_message_size_cap", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_client_message_size_cap", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(500_M)
- .set_description(""),
+ .set_description("maximum memory to devote to in-flight client requests")
+ .set_long_description("If this value is exceeded, the OSD will not read any new client data off of the network until memory is freed."),
Option("osd_client_message_cap", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(100)
- .set_description(""),
-
- Option("osd_pg_bits", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(6)
- .set_description(""),
-
- Option("osd_pgp_bits", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(6)
- .set_description(""),
+ .set_description("maximum number of in-flight client requests"),
Option("osd_crush_update_weight_set", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("update CRUSH weight-set weights when updating weights")
+ .set_long_description("If this setting is true, we will update the weight-set weights when adjusting an item's weight, effectively making changes take effect immediately, and discarding any previous optimization in the weight-set value. Setting this value to false will leave it to the balancer to (slowly, presumably) adjust weights to approach the new target value."),
- Option("osd_crush_chooseleaf_type", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_crush_chooseleaf_type", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(1)
- .set_description(""),
+ .set_flag(Option::FLAG_CREATE)
+ .set_description("default chooseleaf type for osdmaptool --create"),
- Option("osd_pool_use_gmt_hitset", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("osd_pool_use_gmt_hitset", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
- .set_description(""),
+ .set_description("use UTC for hitset timestamps")
+ .set_long_description("This setting only exists for compatibility with hammer (and older) clusters."),
Option("osd_crush_update_on_start", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("update OSD CRUSH location on startup"),
Option("osd_class_update_on_start", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("set OSD device class on startup"),
Option("osd_crush_initial_weight", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(-1)
- .set_description(""),
+ .set_description("if >= 0, initial CRUSH weight for newly created OSDs")
+ .set_long_description("If this value is negative, the size of the OSD in TiB is used."),
+
+ Option("osd_pool_default_ec_fast_read", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description("set ec_fast_read for new erasure-coded pools")
+ .add_service("mon"),
Option("osd_pool_default_crush_rule", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(-1)
- .set_description(""),
+ .set_description("CRUSH rule for newly created pools")
+ .add_service("mon"),
- Option("osd_pool_erasure_code_stripe_unit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_pool_erasure_code_stripe_unit", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(4_K)
- .set_description(""),
+ .set_description("the amount of data (in bytes) in a data chunk, per stripe")
+ .add_service("mon"),
- Option("osd_pool_default_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_pool_default_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(3)
- .set_description(""),
+ .set_min_max(0, 10)
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("the number of copies of an object for new replicated pools")
+ .add_service("mon"),
- Option("osd_pool_default_min_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_pool_default_min_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description(""),
+ .set_min_max(0, 255)
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("the minimal number of copies allowed to write to a degraded pool for new replicated pools")
+ .set_long_description("0 means no specific default; ceph will use size-size/2")
+ .add_see_also("osd_pool_default_size")
+ .add_service("mon"),
- Option("osd_pool_default_pg_num", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_pool_default_pg_num", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(8)
- .set_description(""),
+ .set_description("number of PGs for new pools")
+ .set_flag(Option::FLAG_RUNTIME)
+ .add_service("mon"),
- Option("osd_pool_default_pgp_num", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(8)
- .set_description(""),
+ Option("osd_pool_default_pgp_num", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("number of PGs for placement purposes (0 to match pg_num)")
+ .add_see_also("osd_pool_default_pg_num")
+ .set_flag(Option::FLAG_RUNTIME)
+ .add_service("mon"),
Option("osd_pool_default_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("replicated")
- .set_description(""),
+ .set_enum_allowed({"replicated", "erasure"})
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("default type of pool to create")
+ .add_service("mon"),
Option("osd_pool_default_erasure_code_profile", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("plugin=jerasure technique=reed_sol_van k=2 m=1")
- .set_description(""),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("default erasure code profile for new erasure-coded pools")
+ .add_service("mon"),
Option("osd_erasure_code_plugins", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("jerasure lrc"
" isa"
#endif
)
- .set_description(""),
+ .set_flag(Option::FLAG_STARTUP)
+ .set_description("erasure code plugins to load")
+ .add_service("mon")
+ .add_service("osd"),
- Option("osd_allow_recovery_below_min_size", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("osd_allow_recovery_below_min_size", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
- .set_description(""),
+ .set_description("allow replicated pools to recover with < min_size active members")
+ .add_service("osd"),
- Option("osd_pool_default_flags", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_pool_default_flags", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
- .set_description(""),
+ .set_description("(integer) flags to set on new pools")
+ .add_service("mon"),
Option("osd_pool_default_flag_hashpspool", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("set hashpspool (better hashing scheme) flag on new pools")
+ .add_service("mon"),
Option("osd_pool_default_flag_nodelete", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("set nodelete flag on new pools")
+ .add_service("mon"),
Option("osd_pool_default_flag_nopgchange", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("set nopgchange flag on new pools")
+ .add_service("mon"),
Option("osd_pool_default_flag_nosizechange", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("set nosizechange flag on new pools")
+ .add_service("mon"),
Option("osd_pool_default_hit_set_bloom_fpp", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.05)
- .set_description(""),
+ .set_description("")
+ .add_see_also("osd_tier_default_cache_hit_set_type")
+ .add_service("mon"),
Option("osd_pool_default_cache_target_dirty_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.4)
.set_default(10)
.set_description(""),
+ Option("osd_pool_default_pg_autoscale_mode", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("warn")
+ .set_enum_allowed({"off", "warn", "on"})
+ .set_description("Default PG autoscaling behavior for new pools"),
+
Option("osd_hit_set_min_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1000)
.set_description(""),
.set_default(25)
.set_description(""),
- Option("osd_tier_promote_max_bytes_sec", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_tier_promote_max_bytes_sec", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(5_M)
.set_description(""),
Option("osd_tier_default_cache_mode", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("writeback")
+ .set_enum_allowed({"none", "writeback", "forward",
+ "readonly", "readforward", "readproxy", "proxy"})
+ .set_flag(Option::FLAG_RUNTIME)
.set_description(""),
- Option("osd_tier_default_cache_hit_set_count", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_tier_default_cache_hit_set_count", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(4)
.set_description(""),
- Option("osd_tier_default_cache_hit_set_period", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_tier_default_cache_hit_set_period", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1200)
.set_description(""),
Option("osd_tier_default_cache_hit_set_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("bloom")
+ .set_enum_allowed({"bloom", "explicit_hash", "explicit_object"})
+ .set_flag(Option::FLAG_RUNTIME)
.set_description(""),
- Option("osd_tier_default_cache_min_read_recency_for_promote", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_tier_default_cache_min_read_recency_for_promote", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1)
- .set_description(""),
+ .set_description("number of recent HitSets the object must appear in to be promoted (on read)"),
- Option("osd_tier_default_cache_min_write_recency_for_promote", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_tier_default_cache_min_write_recency_for_promote", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1)
- .set_description(""),
+ .set_description("number of recent HitSets the object must appear in to be promoted (on write)"),
- Option("osd_tier_default_cache_hit_set_grade_decay_rate", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_tier_default_cache_hit_set_grade_decay_rate", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(20)
.set_description(""),
- Option("osd_tier_default_cache_hit_set_search_last_n", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_tier_default_cache_hit_set_search_last_n", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1)
.set_description(""),
- Option("osd_map_dedup", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(true)
+ Option("osd_objecter_finishers", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(1)
+ .set_flag(Option::FLAG_STARTUP)
.set_description(""),
- Option("osd_map_max_advance", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(40)
+ Option("osd_map_dedup", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
.set_description(""),
Option("osd_map_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
Option("osd_map_message_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(40)
- .set_description(""),
+ .set_description("maximum number of OSDMaps to include in a single message"),
+
+ Option("osd_map_message_max_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(10_M)
+ .set_description("maximum number of bytes worth of OSDMaps to include in a single message"),
Option("osd_map_share_max_epochs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(40)
.set_description(""),
+ Option("osd_pg_epoch_max_lag_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(2.0)
+ .set_description("Max multiple of the map cache that PGs can lag before we throttle map injest")
+ .add_see_also("osd_map_cache_size"),
+
Option("osd_inject_bad_map_crc_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
.set_default(5)
.set_description(""),
- Option("osd_peering_wq_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(2)
- .set_description(""),
-
- Option("osd_peering_wq_batch_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(20)
- .set_description(""),
-
Option("osd_op_pq_max_tokens_per_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(4194304)
.set_description(""),
- Option("osd_op_pq_min_cost", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_op_pq_min_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(65536)
.set_description(""),
- Option("osd_remove_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(1)
- .set_description(""),
-
- Option("osd_recovery_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(1)
- .set_description(""),
-
- Option("osd_disk_thread_ioprio_class", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("")
- .set_description(""),
-
- Option("osd_disk_thread_ioprio_priority", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(-1)
- .set_description(""),
-
Option("osd_recover_clone_overlap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
.set_description(""),
Option("osd_op_num_threads_per_shard", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(0)
+ .set_flag(Option::FLAG_STARTUP)
.set_description(""),
Option("osd_op_num_threads_per_shard_hdd", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1)
- .set_description(""),
+ .set_flag(Option::FLAG_STARTUP)
+ .set_description("")
+ .add_see_also("osd_op_num_threads_per_shard"),
Option("osd_op_num_threads_per_shard_ssd", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(2)
- .set_description(""),
+ .set_flag(Option::FLAG_STARTUP)
+ .set_description("")
+ .add_see_also("osd_op_num_threads_per_shard"),
Option("osd_op_num_shards", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(0)
+ .set_flag(Option::FLAG_STARTUP)
.set_description(""),
Option("osd_op_num_shards_hdd", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_flag(Option::FLAG_STARTUP)
+ .set_description("")
+ .add_see_also("osd_op_num_shards"),
Option("osd_op_num_shards_ssd", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(8)
- .set_description(""),
+ .set_flag(Option::FLAG_STARTUP)
+ .set_description("")
+ .add_see_also("osd_op_num_shards"),
Option("osd_skip_data_digest", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description("Do not store full-object checksums if the backend (bluestore) does its own checksums. Do not ever turn this off if it has ever been turned on."),
-
- Option("osd_distrust_data_digest", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
- .set_description("Do not trust stored data_digest (due to previous bug or corruption)"),
+ .set_description("Do not store full-object checksums if the backend (bluestore) does its own checksums. Only usable with all BlueStore OSDs."),
Option("osd_op_queue", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("wpq")
.set_enum_allowed( { "wpq", "prioritized", "mclock_opclass", "mclock_client", "debug_random" } )
.set_description("which operation queue algorithm to use")
.set_long_description("which operation queue algorithm to use; mclock_opclass and mclock_client are currently experimental")
+ .set_flag(Option::FLAG_STARTUP)
.add_see_also("osd_op_queue_cut_off"),
Option("osd_op_queue_cut_off", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_recov_lim")
.add_see_also("osd_op_queue_mclock_scrub_res")
.add_see_also("osd_op_queue_mclock_scrub_wgt")
- .add_see_also("osd_op_queue_mclock_scrub_lim"),
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
Option("osd_op_queue_mclock_client_op_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(500.0)
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_recov_lim")
.add_see_also("osd_op_queue_mclock_scrub_res")
.add_see_also("osd_op_queue_mclock_scrub_wgt")
- .add_see_also("osd_op_queue_mclock_scrub_lim"),
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
Option("osd_op_queue_mclock_client_op_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0.0)
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_recov_lim")
.add_see_also("osd_op_queue_mclock_scrub_res")
.add_see_also("osd_op_queue_mclock_scrub_wgt")
- .add_see_also("osd_op_queue_mclock_scrub_lim"),
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
- Option("osd_op_queue_mclock_osd_subop_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("osd_op_queue_mclock_osd_rep_op_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1000.0)
- .set_description("mclock reservation of osd sub-operation requests")
- .set_long_description("mclock reservation of osd sub-operation requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
+ .set_description("mclock reservation of osd replication operation requests and replies")
+ .set_long_description("mclock reservation of replication operation requests and replies when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_recov_lim")
.add_see_also("osd_op_queue_mclock_scrub_res")
.add_see_also("osd_op_queue_mclock_scrub_wgt")
- .add_see_also("osd_op_queue_mclock_scrub_lim"),
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
- Option("osd_op_queue_mclock_osd_subop_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("osd_op_queue_mclock_osd_rep_op_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(500.0)
- .set_description("mclock weight of osd sub-operation requests")
- .set_long_description("mclock weight of osd sub-operation requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
+ .set_description("mclock weight of osd replication operation requests and replies")
+ .set_long_description("mclock weight of osd replication operation requests and replies when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_recov_lim")
.add_see_also("osd_op_queue_mclock_scrub_res")
.add_see_also("osd_op_queue_mclock_scrub_wgt")
- .add_see_also("osd_op_queue_mclock_scrub_lim"),
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
- Option("osd_op_queue_mclock_osd_subop_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("osd_op_queue_mclock_osd_rep_op_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0.0)
- .set_description("mclock limit of osd sub-operation requests")
+ .set_description("mclock limit of osd replication operation requests and replies")
.set_long_description("mclock limit of osd sub-operation requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_recov_lim")
.add_see_also("osd_op_queue_mclock_scrub_res")
.add_see_also("osd_op_queue_mclock_scrub_wgt")
- .add_see_also("osd_op_queue_mclock_scrub_lim"),
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
Option("osd_op_queue_mclock_snap_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0.0)
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_recov_res")
.add_see_also("osd_op_queue_mclock_recov_lim")
.add_see_also("osd_op_queue_mclock_scrub_res")
.add_see_also("osd_op_queue_mclock_scrub_wgt")
- .add_see_also("osd_op_queue_mclock_scrub_lim"),
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
Option("osd_op_queue_mclock_snap_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1.0)
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
+ .add_see_also("osd_op_queue_mclock_snap_res")
+ .add_see_also("osd_op_queue_mclock_snap_lim")
+ .add_see_also("osd_op_queue_mclock_recov_res")
+ .add_see_also("osd_op_queue_mclock_recov_wgt")
+ .add_see_also("osd_op_queue_mclock_recov_lim")
+ .add_see_also("osd_op_queue_mclock_scrub_res")
+ .add_see_also("osd_op_queue_mclock_scrub_wgt")
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
+
+ Option("osd_op_queue_mclock_snap_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.001)
+ .set_description("")
+ .set_description("mclock limit of snaptrim requests")
+ .set_long_description("mclock limit of snaptrim requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
+ .add_see_also("osd_op_queue_mclock_client_op_res")
+ .add_see_also("osd_op_queue_mclock_client_op_wgt")
+ .add_see_also("osd_op_queue_mclock_client_op_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
+ .add_see_also("osd_op_queue_mclock_snap_res")
+ .add_see_also("osd_op_queue_mclock_snap_wgt")
+ .add_see_also("osd_op_queue_mclock_recov_res")
+ .add_see_also("osd_op_queue_mclock_recov_wgt")
+ .add_see_also("osd_op_queue_mclock_recov_lim")
+ .add_see_also("osd_op_queue_mclock_scrub_res")
+ .add_see_also("osd_op_queue_mclock_scrub_wgt")
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
+
+ Option("osd_op_queue_mclock_recov_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.0)
+ .set_description("mclock reservation of recovery requests")
+ .set_long_description("mclock reservation of recovery requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
+ .add_see_also("osd_op_queue")
+ .add_see_also("osd_op_queue_mclock_client_op_res")
+ .add_see_also("osd_op_queue_mclock_client_op_wgt")
+ .add_see_also("osd_op_queue_mclock_client_op_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
+ .add_see_also("osd_op_queue_mclock_snap_res")
+ .add_see_also("osd_op_queue_mclock_snap_wgt")
+ .add_see_also("osd_op_queue_mclock_snap_lim")
+ .add_see_also("osd_op_queue_mclock_recov_wgt")
+ .add_see_also("osd_op_queue_mclock_recov_lim")
+ .add_see_also("osd_op_queue_mclock_scrub_res")
+ .add_see_also("osd_op_queue_mclock_scrub_wgt")
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
+
+ Option("osd_op_queue_mclock_recov_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(1.0)
+ .set_description("mclock weight of recovery requests")
+ .set_long_description("mclock weight of recovery requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
+ .add_see_also("osd_op_queue")
+ .add_see_also("osd_op_queue_mclock_client_op_res")
+ .add_see_also("osd_op_queue_mclock_client_op_wgt")
+ .add_see_also("osd_op_queue_mclock_client_op_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
+ .add_see_also("osd_op_queue_mclock_snap_res")
+ .add_see_also("osd_op_queue_mclock_snap_wgt")
+ .add_see_also("osd_op_queue_mclock_snap_lim")
+ .add_see_also("osd_op_queue_mclock_recov_res")
+ .add_see_also("osd_op_queue_mclock_recov_lim")
+ .add_see_also("osd_op_queue_mclock_scrub_res")
+ .add_see_also("osd_op_queue_mclock_scrub_wgt")
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
+
+ Option("osd_op_queue_mclock_recov_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.001)
+ .set_description("mclock limit of recovery requests")
+ .set_long_description("mclock limit of recovery requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
+ .add_see_also("osd_op_queue")
+ .add_see_also("osd_op_queue_mclock_client_op_res")
+ .add_see_also("osd_op_queue_mclock_client_op_wgt")
+ .add_see_also("osd_op_queue_mclock_client_op_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
+ .add_see_also("osd_op_queue_mclock_snap_res")
+ .add_see_also("osd_op_queue_mclock_snap_wgt")
+ .add_see_also("osd_op_queue_mclock_snap_lim")
+ .add_see_also("osd_op_queue_mclock_recov_res")
+ .add_see_also("osd_op_queue_mclock_recov_wgt")
+ .add_see_also("osd_op_queue_mclock_scrub_res")
+ .add_see_also("osd_op_queue_mclock_scrub_wgt")
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
+
+ Option("osd_op_queue_mclock_scrub_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.0)
+ .set_description("mclock reservation of scrub requests")
+ .set_long_description("mclock reservation of scrub requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
+ .add_see_also("osd_op_queue")
+ .add_see_also("osd_op_queue_mclock_client_op_res")
+ .add_see_also("osd_op_queue_mclock_client_op_wgt")
+ .add_see_also("osd_op_queue_mclock_client_op_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
+ .add_see_also("osd_op_queue_mclock_snap_res")
+ .add_see_also("osd_op_queue_mclock_snap_wgt")
+ .add_see_also("osd_op_queue_mclock_snap_lim")
+ .add_see_also("osd_op_queue_mclock_recov_res")
+ .add_see_also("osd_op_queue_mclock_recov_wgt")
+ .add_see_also("osd_op_queue_mclock_recov_lim")
+ .add_see_also("osd_op_queue_mclock_scrub_wgt")
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
+
+ Option("osd_op_queue_mclock_scrub_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(1.0)
+ .set_description("mclock weight of scrub requests")
+ .set_long_description("mclock weight of scrub requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
+ .add_see_also("osd_op_queue")
+ .add_see_also("osd_op_queue_mclock_client_op_res")
+ .add_see_also("osd_op_queue_mclock_client_op_wgt")
+ .add_see_also("osd_op_queue_mclock_client_op_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
+ .add_see_also("osd_op_queue_mclock_snap_res")
+ .add_see_also("osd_op_queue_mclock_snap_wgt")
+ .add_see_also("osd_op_queue_mclock_snap_lim")
+ .add_see_also("osd_op_queue_mclock_recov_res")
+ .add_see_also("osd_op_queue_mclock_recov_wgt")
+ .add_see_also("osd_op_queue_mclock_recov_lim")
+ .add_see_also("osd_op_queue_mclock_scrub_res")
+ .add_see_also("osd_op_queue_mclock_scrub_lim")
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
+
+ Option("osd_op_queue_mclock_scrub_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.001)
+ .set_description("mclock weight of limit requests")
+ .set_long_description("mclock weight of limit requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
+ .add_see_also("osd_op_queue")
+ .add_see_also("osd_op_queue_mclock_client_op_res")
+ .add_see_also("osd_op_queue_mclock_client_op_wgt")
+ .add_see_also("osd_op_queue_mclock_client_op_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
+ .add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_recov_res")
.add_see_also("osd_op_queue_mclock_recov_wgt")
.add_see_also("osd_op_queue_mclock_recov_lim")
.add_see_also("osd_op_queue_mclock_scrub_res")
.add_see_also("osd_op_queue_mclock_scrub_wgt")
- .add_see_also("osd_op_queue_mclock_scrub_lim"),
+ .add_see_also("osd_op_queue_mclock_anticipation_timeout"),
- Option("osd_op_queue_mclock_snap_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
- .set_default(0.001)
- .set_description("")
- .set_description("mclock limit of snaptrim requests")
- .set_long_description("mclock limit of snaptrim requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
+ Option("osd_op_queue_mclock_anticipation_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.0)
+ .set_description("mclock anticipation timeout in seconds")
+ .set_long_description("the amount of time that mclock waits until the unused resource is forfeited")
+ .add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
+ .add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_recov_res")
.add_see_also("osd_op_queue_mclock_recov_wgt")
.add_see_also("osd_op_queue_mclock_recov_lim")
.add_see_also("osd_op_queue_mclock_scrub_wgt")
.add_see_also("osd_op_queue_mclock_scrub_lim"),
- Option("osd_op_queue_mclock_recov_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("osd_op_queue_mclock_pg_delete_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0.0)
- .set_description("mclock reservation of recovery requests")
- .set_long_description("mclock reservation of recovery requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
+ .set_description("mclock reservation of pg delete work")
+ .set_long_description("mclock reservation of pg delete work when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
+ .add_see_also("osd_op_queue_mclock_recov_res")
.add_see_also("osd_op_queue_mclock_recov_wgt")
.add_see_also("osd_op_queue_mclock_recov_lim")
- .add_see_also("osd_op_queue_mclock_scrub_res")
.add_see_also("osd_op_queue_mclock_scrub_wgt")
.add_see_also("osd_op_queue_mclock_scrub_lim"),
- Option("osd_op_queue_mclock_recov_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("osd_op_queue_mclock_pg_delete_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1.0)
- .set_description("mclock weight of recovery requests")
- .set_long_description("mclock weight of recovery requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
+ .set_description("mclock weight of pg delete work")
+ .set_long_description("mclock weight of pg delete work when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_recov_res")
+ .add_see_also("osd_op_queue_mclock_recov_wgt")
.add_see_also("osd_op_queue_mclock_recov_lim")
.add_see_also("osd_op_queue_mclock_scrub_res")
- .add_see_also("osd_op_queue_mclock_scrub_wgt")
.add_see_also("osd_op_queue_mclock_scrub_lim"),
- Option("osd_op_queue_mclock_recov_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("osd_op_queue_mclock_pg_delete_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0.001)
- .set_description("mclock limit of recovery requests")
- .set_long_description("mclock limit of recovery requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
+ .set_description("mclock weight of pg delete work limit requests")
+ .set_long_description("mclock weight of limit pg delete work when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_recov_res")
.add_see_also("osd_op_queue_mclock_recov_wgt")
+ .add_see_also("osd_op_queue_mclock_recov_lim")
.add_see_also("osd_op_queue_mclock_scrub_res")
- .add_see_also("osd_op_queue_mclock_scrub_wgt")
- .add_see_also("osd_op_queue_mclock_scrub_lim"),
+ .add_see_also("osd_op_queue_mclock_scrub_wgt"),
- Option("osd_op_queue_mclock_scrub_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("osd_op_queue_mclock_peering_event_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0.0)
- .set_description("mclock reservation of scrub requests")
+ .set_description("mclock reservation of peering events")
.set_long_description("mclock reservation of scrub requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the reservation")
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_scrub_wgt")
.add_see_also("osd_op_queue_mclock_scrub_lim"),
- Option("osd_op_queue_mclock_scrub_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("osd_op_queue_mclock_peering_event_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1.0)
- .set_description("mclock weight of scrub requests")
+ .set_description("mclock weight of peering events")
.set_long_description("mclock weight of scrub requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the weight")
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.add_see_also("osd_op_queue_mclock_scrub_res")
.add_see_also("osd_op_queue_mclock_scrub_lim"),
- Option("osd_op_queue_mclock_scrub_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("osd_op_queue_mclock_peering_event_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0.001)
- .set_description("mclock weight of limit requests")
+ .set_description("mclock weight of limit peering events")
.set_long_description("mclock weight of limit requests when osd_op_queue is either 'mclock_opclass' or 'mclock_client'; higher values increase the limit")
.add_see_also("osd_op_queue")
.add_see_also("osd_op_queue_mclock_client_op_res")
.add_see_also("osd_op_queue_mclock_client_op_wgt")
.add_see_also("osd_op_queue_mclock_client_op_lim")
- .add_see_also("osd_op_queue_mclock_osd_subop_res")
- .add_see_also("osd_op_queue_mclock_osd_subop_wgt")
- .add_see_also("osd_op_queue_mclock_osd_subop_lim")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_res")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_wgt")
+ .add_see_also("osd_op_queue_mclock_osd_rep_op_lim")
.add_see_also("osd_op_queue_mclock_snap_res")
.add_see_also("osd_op_queue_mclock_snap_wgt")
.add_see_also("osd_op_queue_mclock_snap_lim")
.set_default(150)
.set_description(""),
- Option("osd_recovery_thread_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(30)
- .set_description(""),
-
- Option("osd_recovery_thread_suicide_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(300)
- .set_description(""),
-
Option("osd_recovery_sleep", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("Time in seconds to sleep before next recovery or backfill op"),
Option("osd_recovery_sleep_ssd", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description("Time in seconds to sleep before next recovery or backfill op for SSDs"),
+ .set_description("Time in seconds to sleep before next recovery or backfill op for SSDs")
+ .add_see_also("osd_recovery_sleep"),
Option("osd_recovery_sleep_hybrid", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0.025)
- .set_description("Time in seconds to sleep before next recovery or backfill op when data is on HDD and journal is on SSD"),
+ .set_description("Time in seconds to sleep before next recovery or backfill op when data is on HDD and journal is on SSD")
+ .add_see_also("osd_recovery_sleep"),
Option("osd_snap_trim_sleep", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0)
.set_default(true)
.set_description(""),
- Option("osd_remove_thread_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(1_hr)
- .set_description(""),
-
- Option("osd_remove_thread_suicide_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(10_hr)
- .set_description(""),
-
Option("osd_command_thread_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(10_min)
.set_description(""),
.set_default(15_min)
.set_description(""),
- Option("osd_heartbeat_addr", Option::TYPE_ADDR, Option::LEVEL_ADVANCED)
- .set_default(entity_addr_t())
- .set_description(""),
-
Option("osd_heartbeat_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(6)
- .set_description(""),
+ .set_min_max(1, 86400)
+ .set_description("Interval (in seconds) between peer pings"),
Option("osd_heartbeat_grace", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(20)
.set_default(false)
.set_description(""),
- Option("osd_heartbeat_min_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_heartbeat_min_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(2000)
- .set_description(""),
+ .set_description("Minimum heartbeat packet size in bytes. Will add dummy payload if heartbeat packet is smaller than this."),
Option("osd_pg_max_concurrent_snap_trims", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(2)
.set_default(30)
.set_description(""),
- Option("osd_mon_report_interval_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(600)
- .set_description(""),
-
- Option("osd_mon_report_interval_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("osd_mon_report_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("Frequency of OSD reports to mon for peer failures, fullness status changes"),
Option("osd_mon_report_max_in_flight", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(2)
.set_default(.9)
.set_description(""),
+ Option("osd_max_snap_prune_intervals_per_epoch", Option::TYPE_UINT, Option::LEVEL_DEV)
+ .set_default(512)
+ .set_description("Max number of snap intervals to report to mgr in pg_stat_t"),
+
Option("osd_default_data_pool_replay_window", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(45)
.set_description(""),
.set_default(1)
.set_description(""),
- Option("osd_recovery_max_chunk", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(8<<20)
+ Option("osd_recovery_max_chunk", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(8_M)
.set_description(""),
Option("osd_recovery_max_omap_entries_per_chunk", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(8096)
.set_description(""),
- Option("osd_copyfrom_max_chunk", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(8<<20)
+ Option("osd_copyfrom_max_chunk", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(8_M)
.set_description(""),
- Option("osd_push_per_object_cost", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_push_per_object_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1000)
.set_description(""),
- Option("osd_max_push_cost", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_max_push_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(8<<20)
.set_description(""),
.set_default(10)
.set_description(""),
- Option("osd_recovery_forget_lost_objects", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
- .set_description(""),
-
Option("osd_max_scrubs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1)
.set_description("Maximum concurrent scrubs on a single OSD"),
Option("osd_scrub_chunk_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(25)
- .set_description("Maximum number of object to scrub in a single chunk")
+ .set_description("Maximum number of objects to scrub in a single chunk")
.add_see_also("osd_scrub_chunk_min"),
Option("osd_scrub_sleep", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_description("Scrubs will randomly become deep scrubs at this rate (0.15 -> 15% of scrubs are deep)")
.set_long_description("This prevents a deep scrub 'stampede' by spreading deep scrubs so they are uniformly distributed over the week"),
- Option("osd_deep_scrub_stride", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(524288)
+ Option("osd_deep_scrub_stride", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(512_K)
.set_description("Number of bytes to read from an object at a time during deep scrub"),
Option("osd_deep_scrub_keys", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.add_service("osd")
.add_see_also("osd_deep_scrub_large_omap_object_value_sum_threshold"),
- Option("osd_deep_scrub_large_omap_object_value_sum_threshold", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_deep_scrub_large_omap_object_value_sum_threshold", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1_G)
.set_description("Warn when we encounter an object with more omap key bytes than this")
.add_service("osd")
.set_description(""),
Option("osd_class_load_list", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("cephfs hello journal lock log numops " "rbd refcount replica_log rgw statelog timeindex user version")
+ .set_default("cephfs hello journal lock log numops " "otp rbd refcount rgw timeindex user version cas")
.set_description(""),
Option("osd_class_default_list", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("cephfs hello journal lock log numops " "rbd refcount replica_log rgw statelog timeindex user version")
+ .set_default("cephfs hello journal lock log numops " "otp rbd refcount rgw timeindex user version cas")
.set_description(""),
Option("osd_check_for_log_corruption", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_description(""),
Option("osd_min_pg_log_entries", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(1500)
+ .set_default(3000)
.set_description("minimum number of entries to maintain in the PG log")
.add_service("osd")
.add_see_also("osd_max_pg_log_entries")
.add_see_also("osd_pg_log_dups_tracked"),
Option("osd_max_pg_log_entries", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(10000)
+ .set_default(3000)
.set_description("maximum number of entries to maintain in the PG log when degraded before we trim")
.add_service("osd")
.add_see_also("osd_min_pg_log_entries")
.set_default(100)
.set_description(""),
+ Option("osd_force_auth_primary_missing_objects", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(100)
+ .set_description("Approximate missing objects above which to force auth_log_shard to be primary temporarily"),
+
+ Option("osd_async_recovery_min_cost", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(100)
+ .set_description("A mixture measure of number of current log entries difference "
+ "and historical missing objects, above which we switch to use "
+ "asynchronous recovery when appropriate"),
+
Option("osd_max_pg_per_osd_hard_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(3)
.set_min(1)
.set_default(false)
.set_description(""),
- Option("osd_backoff_on_down", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(true)
- .set_description(""),
-
Option("osd_backoff_on_peering", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
.set_description(""),
.set_default(0)
.set_description("Inject an expensive sleep during deep scrub IO to make it easier to induce preemption"),
+ Option("osd_debug_no_acting_change", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ .set_default(false),
+ Option("osd_debug_no_purge_strays", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ .set_default(false),
+
Option("osd_enable_op_tracker", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
.set_description(""),
Option("osd_delete_sleep", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description("Time in seconds to sleep before next removal transaction"),
+ .set_description("Time in seconds to sleep before next removal transaction (overrides values below)"),
+
+ Option("osd_delete_sleep_hdd", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(5)
+ .set_description("Time in seconds to sleep before next removal transaction for HDDs"),
+
+ Option("osd_delete_sleep_ssd", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("Time in seconds to sleep before next removal transaction for SSDs"),
+
+ Option("osd_delete_sleep_hybrid", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(2)
+ .set_description("Time in seconds to sleep before next removal transaction when data is on HDD and journal is on SSD"),
Option("osd_failsafe_full_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.97)
.set_default(true)
.set_description(""),
- Option("leveldb_write_buffer_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("leveldb_write_buffer_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(8_M)
.set_description(""),
- Option("leveldb_cache_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("leveldb_cache_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(128_M)
.set_description(""),
- Option("leveldb_block_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("leveldb_block_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description(""),
.set_default(false)
.set_description(""),
- Option("rocksdb_separate_wal_dir", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
- .set_description(""),
-
- Option("rocksdb_db_paths", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("")
- .set_description("")
- .set_safe(),
-
Option("rocksdb_log_to_ceph_log", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
.set_description(""),
- Option("rocksdb_cache_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("rocksdb_cache_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(512_M)
.set_description(""),
.set_default("binned_lru")
.set_description(""),
- Option("rocksdb_block_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rocksdb_block_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(4_K)
.set_description(""),
.set_description("(experimental) partition SST index/filters into smaller blocks")
.set_long_description("This is an experimental option for rocksdb that works in conjunction with two_level indices to avoid having to keep the entire filter/index in cache when cache_index_and_filter_blocks is true. The idea is to keep a much smaller top-level index in heap/cache and then opportunistically cache the lower level indices. See: https://github.com/facebook/rocksdb/wiki/Partitioned-Index-Filters"),
- Option("rocksdb_metadata_block_size", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("rocksdb_metadata_block_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(4_K)
.set_description("The block size for index partitions. (0 = rocksdb default)"),
Option("osd_recovery_op_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(3)
+ .set_description("Priority to use for recovery operations if not specified for the pool"),
+
+ Option("osd_peering_op_priority", Option::TYPE_UINT, Option::LEVEL_DEV)
+ .set_default(255)
.set_description(""),
Option("osd_snap_trim_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(5)
.set_description(""),
- Option("osd_snap_trim_cost", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_snap_trim_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(1<<20)
+ .set_description(""),
+
+ Option("osd_pg_delete_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(5)
+ .set_description(""),
+
+ Option("osd_pg_delete_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1<<20)
.set_description(""),
.set_default(5)
.set_description("Priority for scrub operations in work queue"),
- Option("osd_scrub_cost", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_scrub_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(50<<20)
.set_description("Cost for scrub operations in work queue"),
Option("osd_recovery_priority", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("Priority of recovery in the work queue")
+ .set_long_description("Not related to a pool's recovery_priority"),
- Option("osd_recovery_cost", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_recovery_cost", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(20<<20)
.set_description(""),
.set_default(false)
.set_description(""),
- Option("osd_max_object_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("osd_max_object_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(128_M)
.set_description(""),
.set_default(131072)
.set_description(""),
- Option("osd_max_omap_bytes_per_request", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(1<<30)
+ Option("osd_max_omap_bytes_per_request", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(1_G)
.set_description(""),
Option("osd_objectstore", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("filestore")
- .set_description(""),
+ .set_default("bluestore")
+ .set_enum_allowed({"bluestore", "filestore", "memstore", "kstore"})
+ .set_flag(Option::FLAG_CREATE)
+ .set_description("backend type for an OSD (like filestore or bluestore)"),
Option("osd_objectstore_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
.set_default(100)
.set_description(""),
- Option("osd_bench_large_size_max_throughput", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(100 << 20)
+ Option("osd_bench_large_size_max_throughput", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(100_M)
.set_description(""),
- Option("osd_bench_max_block_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(64 << 20)
+ Option("osd_bench_max_block_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(64_M)
.set_description(""),
Option("osd_bench_duration", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(true)
.set_description(""),
-
- Option("osd_memory_target", Option::TYPE_UINT, Option::LEVEL_BASIC)
+ Option("osd_memory_target", Option::TYPE_SIZE, Option::LEVEL_BASIC)
.set_default(4_G)
.add_see_also("bluestore_cache_autotune")
.set_description("When tcmalloc and cache autotuning is enabled, try to keep this many bytes mapped in memory."),
- Option("osd_memory_base", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("osd_memory_target_cgroup_limit_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.8)
+ .set_min_max(0.0, 1.0)
+ .add_see_also("osd_memory_target")
+ .set_description("Set the default value for osd_memory_target to the cgroup memory limit (if set) times this value")
+ .set_long_description("A value of 0 disables this feature."),
+
+ Option("osd_memory_base", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(768_M)
.add_see_also("bluestore_cache_autotune")
.set_description("When tcmalloc and cache autotuning is enabled, estimate the minimum amount of memory in bytes the OSD will need."),
.add_see_also("bluestore_cache_autotune")
.set_description("When tcmalloc and cache autotuning is enabled, estimate the percent of memory fragmentation."),
- Option("osd_memory_cache_min", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("osd_memory_cache_min", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(128_M)
.add_see_also("bluestore_cache_autotune")
.set_description("When tcmalloc and cache autotuning is enabled, set the minimum amount of memory used for caches."),
.add_see_also("bluestore_cache_autotune")
.set_description("When tcmalloc and cache autotuning is enabled, wait this many seconds between resizing caches."),
- Option("memstore_device_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("memstore_device_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1_G)
.set_description(""),
.set_default(false)
.set_description(""),
- Option("memstore_page_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("memstore_page_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(64_K)
.set_description(""),
// --------------------------
// bluestore
- Option("bdev_inject_bad_size", Option::TYPE_BOOL, Option::LEVEL_DEV)
- .set_default(false)
- .set_description(""),
-
Option("bdev_debug_inflight_ios", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
.set_default(16)
.set_description(""),
- Option("bdev_block_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("bdev_block_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(4_K)
.set_description(""),
.set_default(60.0)
.set_description(""),
+ Option("bdev_debug_aio_log_age", Option::TYPE_FLOAT, Option::LEVEL_DEV)
+ .set_default(5.0)
+ .set_description(""),
+
Option("bdev_nvme_unbind_from_kernel", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
.set_description(""),
.set_default(-1)
.set_description(""),
- Option("bluefs_alloc_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bdev_enable_discard", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description(""),
+
+ Option("bdev_async_discard", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description(""),
+
+ Option("bluefs_alloc_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1_M)
.set_description(""),
- Option("bluefs_max_prefetch", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluefs_max_prefetch", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1_M)
.set_description(""),
- Option("bluefs_min_log_runway", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluefs_min_log_runway", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1_M)
.set_description(""),
- Option("bluefs_max_log_runway", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluefs_max_log_runway", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(4194304)
.set_description(""),
.set_default(5.0)
.set_description(""),
- Option("bluefs_log_compact_min_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluefs_log_compact_min_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(16_M)
.set_description(""),
- Option("bluefs_min_flush_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluefs_min_flush_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(512_K)
.set_description(""),
.set_description(""),
Option("bluefs_buffered_io", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
+ .set_default(true)
.set_description(""),
Option("bluefs_sync_write", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_description(""),
Option("bluefs_allocator", Option::TYPE_STR, Option::LEVEL_DEV)
- .set_default("stupid")
+ .set_default("bitmap")
.set_description(""),
Option("bluefs_preextend_wal_files", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
Option("bluestore_bluefs", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Use BlueFS to back rocksdb")
.set_long_description("BlueFS allows rocksdb to share the same physical device(s) as the rest of BlueStore. It should be used in all cases unless testing/developing an alternative metadata database for BlueStore."),
Option("bluestore_bluefs_env_mirror", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Mirror bluefs data to file system for testing/validation"),
- Option("bluestore_bluefs_min", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_bluefs_min", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1_G)
.set_description("minimum disk space allocated to BlueFS (e.g., at mkfs)"),
- Option("bluestore_bluefs_min_free", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(1*1024*1024*1024)
+ Option("bluestore_bluefs_min_free", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(1_G)
.set_description("minimum free space allocated to BlueFS"),
Option("bluestore_bluefs_min_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1)
.set_description("How frequently (in seconds) to balance free space between BlueFS and BlueStore"),
- Option("bluestore_bluefs_balance_failure_dump_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("bluestore_bluefs_alloc_failure_dump_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description("How frequently (in seconds) to dump information on "
- "allocation failure occurred during BlueFS space rebalance"),
+ .set_description("How frequently (in seconds) to dump allocator on"
+ "BlueFS space allocation failure"),
+
+ Option("bluestore_bluefs_db_compatibility", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ .set_default(true)
+ .set_description("Sync db with legacy bluefs extents info")
+ .set_long_description("Enforces db sync with legacy bluefs extents information on close."
+ " Enables downgrades to pre-nautilus releases"),
- Option("bluestore_spdk_mem", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_spdk_mem", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(512)
- .set_description(""),
+ .set_description("Amount of dpdk memory size in MB")
+ .set_long_description("If running multiple SPDK instances per node, you must specify the amount of dpdk memory size in MB each instance will use, to make sure each instance uses its own dpdk memory"),
Option("bluestore_spdk_coremask", Option::TYPE_STR, Option::LEVEL_DEV)
- .set_default("0x3")
- .set_description(""),
+ .set_default("0x1")
+ .set_description("A hexadecimal bit mask of the cores to run on. Note the core numbering can change between platforms and should be determined beforehand"),
Option("bluestore_spdk_max_io_completion", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(0)
- .set_description(""),
+ .set_description("Maximal I/Os to be batched completed while checking queue pair completions, 0 means let spdk library determine it"),
+
+ Option("bluestore_spdk_io_sleep", Option::TYPE_UINT, Option::LEVEL_DEV)
+ .set_default(5)
+ .set_description("Time period to wait if there is no completed I/O from polling"),
Option("bluestore_block_path", Option::TYPE_STR, Option::LEVEL_DEV)
.set_default("")
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Path to block device/file"),
- Option("bluestore_block_size", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_block_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(10_G)
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Size of file to create for backing bluestore"),
Option("bluestore_block_create", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Create bluestore_block_path if it doesn't exist")
.add_see_also("bluestore_block_path").add_see_also("bluestore_block_size"),
Option("bluestore_block_db_path", Option::TYPE_STR, Option::LEVEL_DEV)
.set_default("")
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Path for db block device"),
Option("bluestore_block_db_size", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(0)
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Size of file to create for bluestore_block_db_path"),
Option("bluestore_block_db_create", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Create bluestore_block_db_path if it doesn't exist")
.add_see_also("bluestore_block_db_path")
.add_see_also("bluestore_block_db_size"),
Option("bluestore_block_wal_path", Option::TYPE_STR, Option::LEVEL_DEV)
.set_default("")
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Path to block device/file backing bluefs wal"),
- Option("bluestore_block_wal_size", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_block_wal_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(96_M)
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Size of file to create for bluestore_block_wal_path"),
Option("bluestore_block_wal_create", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Create bluestore_block_wal_path if it doesn't exist")
.add_see_also("bluestore_block_wal_path")
.add_see_also("bluestore_block_wal_size"),
Option("bluestore_block_preallocate_file", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Preallocate file created via bluestore_block*_create"),
+ Option("bluestore_ignore_data_csum", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ .set_default(false)
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Ignore checksum errors on read and do not generate an EIO error"),
+
Option("bluestore_csum_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("crc32c")
.set_enum_allowed({"none", "crc32c", "crc32c_16", "crc32c_8", "xxhash32", "xxhash64"})
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Default checksum algorithm to use")
.set_long_description("crc32c, xxhash32, and xxhash64 are available. The _16 and _8 variants use only a subset of the bits for more compact (but less reliable) checksumming."),
- Option("bluestore_csum_min_block", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(4096)
- .set_safe()
- .set_description("Minimum block size to checksum")
- .set_long_description("A larger checksum block means less checksum metadata to store, but results in read amplification when doing a read smaller than this size (because the entire block must be read to verify the checksum).")
- .add_see_also("bluestore_csum_max_block"),
-
- Option("bluestore_csum_max_block", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(64_K)
- .set_safe()
- .set_description("Maximum block size to checksum")
- .add_see_also("bluestore_csum_min_block"),
-
Option("bluestore_retry_disk_reads", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(3)
.set_min_max(0, 255)
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Number of read retries on checksum validation error")
.set_long_description("Retries to read data from the disk this many times when checksum validation fails to handle spurious read errors gracefully."),
Option("bluestore_min_alloc_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(0)
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Minimum allocation size to allocate for an object")
.set_long_description("A smaller allocation size generally means less data is read and then rewritten when a copy-on-write operation is triggered (e.g., when writing to something that was recently snapshotted). Similarly, less data is journaled before performing an overwrite (writes smaller than min_alloc_size must first pass through the BlueStore journal). Larger values of min_alloc_size reduce the amount of metadata required to describe the on-disk layout and reduce overall fragmentation."),
- Option("bluestore_min_alloc_size_hdd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_min_alloc_size_hdd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(64_K)
- .add_tag("mkfs")
- .set_description("Default min_alloc_size value for rotational media"),
+ .set_flag(Option::FLAG_CREATE)
+ .set_description("Default min_alloc_size value for rotational media")
+ .add_see_also("bluestore_min_alloc_size"),
- Option("bluestore_min_alloc_size_ssd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_min_alloc_size_ssd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(16_K)
- .add_tag("mkfs")
- .set_description("Default min_alloc_size value for non-rotational (solid state) media"),
+ .set_flag(Option::FLAG_CREATE)
+ .set_description("Default min_alloc_size value for non-rotational (solid state) media")
+ .add_see_also("bluestore_min_alloc_size"),
- Option("bluestore_max_alloc_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_max_alloc_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Maximum size of a single allocation (0 for no max)"),
- Option("bluestore_prefer_deferred_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_prefer_deferred_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Writes smaller than this size will be written to the journal and then asynchronously written to the device. This can be beneficial when using rotational media where seeks are expensive, and is helpful both with and without solid state journal/wal devices."),
- Option("bluestore_prefer_deferred_size_hdd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_prefer_deferred_size_hdd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(32768)
- .set_safe()
- .set_description("Default bluestore_prefer_deferred_size for rotational media"),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Default bluestore_prefer_deferred_size for rotational media")
+ .add_see_also("bluestore_prefer_deferred_size"),
- Option("bluestore_prefer_deferred_size_ssd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_prefer_deferred_size_ssd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_safe()
- .set_description("Default bluestore_prefer_deferred_size for non-rotational (solid state) media"),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Default bluestore_prefer_deferred_size for non-rotational (solid state) media")
+ .add_see_also("bluestore_prefer_deferred_size"),
Option("bluestore_compression_mode", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("none")
.set_enum_allowed({"none", "passive", "aggressive", "force"})
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Default policy for using compression when pool does not specify")
.set_long_description("'none' means never use compression. 'passive' means use compression when clients hint that data is compressible. 'aggressive' means use compression unless clients hint that data is not compressible. This option is used when the per-pool property for the compression mode is not present."),
Option("bluestore_compression_algorithm", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("snappy")
.set_enum_allowed({"", "snappy", "zlib", "zstd", "lz4"})
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Default compression algorithm to use when writing object data")
.set_long_description("This controls the default compressor to use (if any) if the per-pool property is not set. Note that zstd is *not* recommended for bluestore due to high CPU overhead when compressing small amounts of data."),
- Option("bluestore_compression_min_blob_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_compression_min_blob_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_safe()
- .set_description("Chunks smaller than this are never compressed"),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Maximum chunk size to apply compression to when random access is expected for an object.")
+ .set_long_description("Chunks larger than this are broken into smaller chunks before being compressed"),
- Option("bluestore_compression_min_blob_size_hdd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_compression_min_blob_size_hdd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(128_K)
- .set_safe()
- .set_description("Default value of bluestore_compression_min_blob_size for rotational media"),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Default value of bluestore_compression_min_blob_size for rotational media")
+ .add_see_also("bluestore_compression_min_blob_size"),
- Option("bluestore_compression_min_blob_size_ssd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_compression_min_blob_size_ssd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(8_K)
- .set_safe()
- .set_description("Default value of bluestore_compression_min_blob_size for non-rotational (solid state) media"),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Default value of bluestore_compression_min_blob_size for non-rotational (solid state) media")
+ .add_see_also("bluestore_compression_min_blob_size"),
- Option("bluestore_compression_max_blob_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_compression_max_blob_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_safe()
- .set_description("Chunks larger than this are broken into smaller chunks before being compressed"),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Maximum chunk size to apply compression to when non-random access is expected for an object.")
+ .set_long_description("Chunks larger than this are broken into smaller chunks before being compressed"),
- Option("bluestore_compression_max_blob_size_hdd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_compression_max_blob_size_hdd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(512_K)
- .set_safe()
- .set_description("Default value of bluestore_compression_max_blob_size for rotational media"),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Default value of bluestore_compression_max_blob_size for rotational media")
+ .add_see_also("bluestore_compression_max_blob_size"),
- Option("bluestore_compression_max_blob_size_ssd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_compression_max_blob_size_ssd", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(64_K)
- .set_safe()
- .set_description("Default value of bluestore_compression_max_blob_size for non-rotational (solid state) media"),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Default value of bluestore_compression_max_blob_size for non-rotational (solid state) media")
+ .add_see_also("bluestore_compression_max_blob_size"),
Option("bluestore_gc_enable_blob_threshold", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description(""),
Option("bluestore_gc_enable_total_threshold", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description(""),
- Option("bluestore_max_blob_size", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_max_blob_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(0)
- .set_safe()
- .set_description(""),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("")
+ .set_long_description("Bluestore blobs are collections of extents (ie on-disk data) originating from one or more objects. Blobs can be compressed, typically have checksum data, may be overwritten, may be shared (with an extent ref map), or split. This setting controls the maximum size a blob is allowed to be."),
- Option("bluestore_max_blob_size_hdd", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_max_blob_size_hdd", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(512_K)
- .set_safe()
- .set_description(""),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("")
+ .add_see_also("bluestore_max_blob_size"),
- Option("bluestore_max_blob_size_ssd", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_max_blob_size_ssd", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(64_K)
- .set_safe()
- .set_description(""),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("")
+ .add_see_also("bluestore_max_blob_size"),
Option("bluestore_compression_required_ratio", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.875)
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Compression ratio required to store compressed data")
.set_long_description("If we compress data and get less than this we discard the result and store the original uncompressed data."),
- Option("bluestore_extent_map_shard_max_size", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_extent_map_shard_max_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(1200)
.set_description("Max size (bytes) for a single extent map shard before splitting"),
- Option("bluestore_extent_map_shard_target_size", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_extent_map_shard_target_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(500)
.set_description("Target size (bytes) for a single extent map shard"),
- Option("bluestore_extent_map_shard_min_size", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_extent_map_shard_min_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(150)
.set_description("Min size (bytes) for a single extent map shard before merging"),
.set_default(.2)
.set_description("Ratio above/below target for a shard when trying to align to an existing extent or blob boundary"),
- Option("bluestore_extent_map_inline_shard_prealloc_size", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_extent_map_inline_shard_prealloc_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(256)
.set_description("Preallocated buffer for inline shards"),
.set_default(.5)
.set_description("2Q paper suggests .5"),
- Option("bluestore_cache_size", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_cache_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(0)
.set_description("Cache size (in bytes) for BlueStore")
.set_long_description("This includes data and metadata cached by BlueStore as well as memory devoted to rocksdb's cache(s)."),
- Option("bluestore_cache_size_hdd", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_cache_size_hdd", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(1_G)
- .set_description("Default bluestore_cache_size for rotational media"),
+ .set_description("Default bluestore_cache_size for rotational media")
+ .add_see_also("bluestore_cache_size"),
- Option("bluestore_cache_size_ssd", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("bluestore_cache_size_ssd", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(3_G)
- .set_description("Default bluestore_cache_size for non-rotational (solid state) media"),
+ .set_description("Default bluestore_cache_size for non-rotational (solid state) media")
+ .add_see_also("bluestore_cache_size"),
Option("bluestore_cache_meta_ratio", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(.4)
.add_see_also("bluestore_cache_meta_ratio")
.set_description("Automatically tune the ratio of caches while respecting min values."),
- Option("bluestore_cache_autotune_chunk_size", Option::TYPE_UINT, Option::LEVEL_DEV)
- .set_default(33554432)
- .add_see_also("bluestore_cache_autotune")
- .set_description("The chunk size in bytes to allocate to caches when cache autotune is enabled."),
-
Option("bluestore_cache_autotune_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(5)
.add_see_also("bluestore_cache_autotune")
Option("bluestore_kvbackend", Option::TYPE_STR, Option::LEVEL_DEV)
.set_default("rocksdb")
- .add_tag("mkfs")
+ .set_flag(Option::FLAG_CREATE)
.set_description("Key value database to use for bluestore"),
- Option("bluestore_allocator", Option::TYPE_STR, Option::LEVEL_DEV)
- .set_default("stupid")
+ Option("bluestore_allocator", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("bitmap")
.set_enum_allowed({"bitmap", "stupid"})
- .set_description("Allocator policy"),
+ .set_description("Allocator policy")
+ .set_long_description("Allocator to use for bluestore. Stupid should only be used for testing."),
- Option("bluestore_freelist_blocks_per_key", Option::TYPE_INT, Option::LEVEL_DEV)
+ Option("bluestore_freelist_blocks_per_key", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(128)
.set_description("Block (and bits) per database key"),
- Option("bluestore_bitmapallocator_blocks_per_zone", Option::TYPE_INT, Option::LEVEL_DEV)
+ Option("bluestore_bitmapallocator_blocks_per_zone", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(1024)
.set_description(""),
- Option("bluestore_bitmapallocator_span_size", Option::TYPE_INT, Option::LEVEL_DEV)
+ Option("bluestore_bitmapallocator_span_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(1024)
.set_description(""),
.set_default("compression=kNoCompression,max_write_buffer_number=4,min_write_buffer_number_to_merge=1,recycle_log_file_num=4,write_buffer_size=268435456,writable_file_max_buffer_size=0,compaction_readahead_size=2097152")
.set_description("Rocksdb options"),
+ Option("bluestore_rocksdb_cf", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description("Enable use of rocksdb column families for bluestore metadata"),
+
+ Option("bluestore_rocksdb_cfs", Option::TYPE_STR, Option::LEVEL_DEV)
+ .set_default("M= P= L=")
+ .set_description("List of whitespace-separate key/value pairs where key is CF name and value is CF options"),
+
Option("bluestore_fsck_on_mount", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description("Run fsck at mount"),
.set_default(false)
.set_description("Try to submit metadata transaction to rocksdb in queuing thread context"),
- Option("bluestore_fsck_read_bytes_cap", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_fsck_read_bytes_cap", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(64_M)
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Maximum bytes read at once by deep fsck"),
- Option("bluestore_throttle_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_throttle_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(64_M)
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Maximum bytes in flight before we throttle IO submission"),
- Option("bluestore_throttle_deferred_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_throttle_deferred_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(128_M)
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Maximum bytes for deferred writes before we throttle IO submission"),
- Option("bluestore_throttle_cost_per_io", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("bluestore_throttle_cost_per_io", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Overhead added to transaction cost (in bytes) for each IO"),
Option("bluestore_throttle_cost_per_io_hdd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(670000)
- .set_safe()
- .set_description("Default bluestore_throttle_cost_per_io for rotational media"),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Default bluestore_throttle_cost_per_io for rotational media")
+ .add_see_also("bluestore_throttle_cost_per_io"),
Option("bluestore_throttle_cost_per_io_ssd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(4000)
- .set_safe()
- .set_description("Default bluestore_throttle_cost_per_io for non-rotation (solid state) media"),
-
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Default bluestore_throttle_cost_per_io for non-rotation (solid state) media")
+ .add_see_also("bluestore_throttle_cost_per_io"),
Option("bluestore_deferred_batch_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Max number of deferred writes before we flush the deferred write queue"),
Option("bluestore_deferred_batch_ops_hdd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(64)
- .set_safe()
- .set_description("Default bluestore_deferred_batch_ops for rotational media"),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Default bluestore_deferred_batch_ops for rotational media")
+ .add_see_also("bluestore_deferred_batch_ops"),
Option("bluestore_deferred_batch_ops_ssd", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(16)
- .set_safe()
- .set_description("Default bluestore_deferred_batch_ops for non-rotational (solid state) media"),
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_description("Default bluestore_deferred_batch_ops for non-rotational (solid state) media")
+ .add_see_also("bluestore_deferred_batch_ops"),
Option("bluestore_nid_prealloc", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(1024)
Option("bluestore_clone_cow", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Use copy-on-write when cloning objects (versus reading and rewriting them at clone time)"),
Option("bluestore_default_buffered_read", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Cache read results by default (unless hinted NOCACHE or WONTNEED)"),
Option("bluestore_default_buffered_write", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_safe()
+ .set_flag(Option::FLAG_RUNTIME)
.set_description("Cache writes by default (unless hinted NOCACHE or WONTNEED)"),
Option("bluestore_debug_misc", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(0)
.set_description("simulate fragmentation"),
- Option("bluestore_debug_prefragment_max", Option::TYPE_INT, Option::LEVEL_DEV)
+ Option("bluestore_debug_prefragment_max", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(1_M)
.set_description(""),
.set_default(false)
.set_description(""),
- Option("bluestore_shard_finishers", Option::TYPE_BOOL, Option::LEVEL_DEV)
- .set_default(false)
- .set_description(""),
-
Option("bluestore_debug_random_read_err", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
+ Option("bluestore_debug_inject_bug21040", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ .set_default(false)
+ .set_description(""),
+
Option("bluestore_debug_inject_csum_err_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0.0)
.set_description("inject crc verification errors into bluestore device reads"),
+ Option("bluestore_no_per_pool_stats_tolerance", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("until_repair")
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_enum_allowed({"enforce", "until_repair", "until_fsck"})
+ .set_description("Specified how to treat lack of per-pool stats, e.g. caused by an upgrade")
+ .set_long_description(
+ "'until_fsck' will tolerate the case for regular ops and fail on fsck or repair, the latter will fix the issue, "
+ "'until_repair' will tolerate for regular ops and fsck. Repair indicates and fixes the issue, "
+ "'enforce' will unconditionally use global stats mode."),
+
+ Option("bluestore_warn_on_bluefs_spillover", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .set_description("Enable health indication on bluefs slow device usage"),
+
+ Option("bluestore_log_op_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(5)
+ .set_description("log operation if it's slower than this age (seconds)"),
+
+ Option("bluestore_log_omap_iterator_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(1)
+ .set_description("log omap iteration operation if it's slower than this age (seconds)"),
+
// -----------------------------------------
// kstore
.set_default(512)
.set_description(""),
- Option("kstore_max_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("kstore_max_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(64_M)
.set_description(""),
Option("kstore_rocksdb_options", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("compression=kNoCompression")
- .set_description(""),
+ .set_description("Options to pass through when RocksDB is used as the KeyValueDB for kstore."),
Option("kstore_fsck_on_mount", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("Whether or not to run fsck on mount for kstore."),
Option("kstore_fsck_on_mount_deep", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("Whether or not to run deep fsck on mount for kstore"),
Option("kstore_nid_prealloc", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1024)
.set_default(1024)
.set_description(""),
- Option("kstore_default_stripe_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("kstore_default_stripe_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(65536)
.set_description(""),
// ---------------------
// filestore
- Option("filestore_rocksdb_options", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("max_background_compactions=8,compaction_readahead_size=2097152,compression=kNoCompression")
- .set_description(""),
+ Option("filestore_rocksdb_options", Option::TYPE_STR, Option::LEVEL_DEV)
+ .set_default("max_background_jobs=10,compaction_readahead_size=2097152,compression=kNoCompression")
+ .set_description("Options to pass through when RocksDB is used as the KeyValueDB for filestore."),
- Option("filestore_omap_backend", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ Option("filestore_omap_backend", Option::TYPE_STR, Option::LEVEL_DEV)
.set_default("rocksdb")
- .set_description(""),
+ .set_enum_allowed({"leveldb", "rocksdb"})
+ .set_description("The KeyValueDB to use for filestore metadata (ie omap)."),
- Option("filestore_omap_backend_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ Option("filestore_omap_backend_path", Option::TYPE_STR, Option::LEVEL_DEV)
.set_default("")
- .set_description(""),
+ .set_description("The path where the filestore KeyValueDB should store it's database(s)."),
Option("filestore_wbthrottle_enable", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("Enabling throttling of operations to backing file system"),
- Option("filestore_wbthrottle_btrfs_bytes_start_flusher", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_wbthrottle_btrfs_bytes_start_flusher", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(41943040)
- .set_description(""),
+ .set_description("Start flushing (fsyncing) when this many bytes are written(btrfs)"),
- Option("filestore_wbthrottle_btrfs_bytes_hard_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_wbthrottle_btrfs_bytes_hard_limit", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(419430400)
- .set_description(""),
+ .set_description("Block writes when this many bytes haven't been flushed (fsynced) (btrfs)"),
Option("filestore_wbthrottle_btrfs_ios_start_flusher", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(500)
- .set_description(""),
+ .set_description("Start flushing (fsyncing) when this many IOs are written (brtrfs)"),
Option("filestore_wbthrottle_btrfs_ios_hard_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(5000)
- .set_description(""),
+ .set_description("Block writes when this many IOs haven't been flushed (fsynced) (btrfs)"),
Option("filestore_wbthrottle_btrfs_inodes_start_flusher", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(500)
- .set_description(""),
+ .set_description("Start flushing (fsyncing) when this many distinct inodes have been modified (btrfs)"),
- Option("filestore_wbthrottle_xfs_bytes_start_flusher", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_wbthrottle_xfs_bytes_start_flusher", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(41943040)
- .set_description(""),
+ .set_description("Start flushing (fsyncing) when this many bytes are written(xfs)"),
- Option("filestore_wbthrottle_xfs_bytes_hard_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_wbthrottle_xfs_bytes_hard_limit", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(419430400)
- .set_description(""),
+ .set_description("Block writes when this many bytes haven't been flushed (fsynced) (xfs)"),
Option("filestore_wbthrottle_xfs_ios_start_flusher", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(500)
- .set_description(""),
+ .set_description("Start flushing (fsyncing) when this many IOs are written (xfs)"),
Option("filestore_wbthrottle_xfs_ios_hard_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(5000)
- .set_description(""),
+ .set_description("Block writes when this many IOs haven't been flushed (fsynced) (xfs)"),
Option("filestore_wbthrottle_xfs_inodes_start_flusher", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(500)
- .set_description(""),
+ .set_description("Start flushing (fsyncing) when this many distinct inodes have been modified (xfs)"),
Option("filestore_wbthrottle_btrfs_inodes_hard_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(5000)
- .set_description(""),
+ .set_description("Block writing when this many inodes have outstanding writes (btrfs)"),
Option("filestore_wbthrottle_xfs_inodes_hard_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(5000)
- .set_description(""),
+ .set_description("Block writing when this many inodes have outstanding writes (xfs)"),
- Option("filestore_odsync_write", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("filestore_odsync_write", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
- .set_description(""),
+ .set_description("Write with O_DSYNC"),
- Option("filestore_index_retry_probability", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("filestore_index_retry_probability", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
.set_default(false)
.set_description(""),
- Option("filestore_omap_header_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("filestore_omap_header_cache_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(1024)
.set_description(""),
- Option("filestore_max_inline_xattr_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_max_inline_xattr_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("filestore_max_inline_xattr_size_xfs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_max_inline_xattr_size_xfs", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(65536)
.set_description(""),
- Option("filestore_max_inline_xattr_size_btrfs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_max_inline_xattr_size_btrfs", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(2048)
.set_description(""),
- Option("filestore_max_inline_xattr_size_other", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_max_inline_xattr_size_other", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(512)
.set_description(""),
- Option("filestore_max_inline_xattrs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_max_inline_xattrs", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("filestore_max_inline_xattrs_xfs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_max_inline_xattrs_xfs", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(10)
.set_description(""),
- Option("filestore_max_inline_xattrs_btrfs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_max_inline_xattrs_btrfs", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(10)
.set_description(""),
- Option("filestore_max_inline_xattrs_other", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_max_inline_xattrs_other", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(2)
.set_description(""),
- Option("filestore_max_xattr_value_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_max_xattr_value_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("filestore_max_xattr_value_size_xfs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(64<<10)
+ Option("filestore_max_xattr_value_size_xfs", Option::TYPE_SIZE, Option::LEVEL_DEV)
+ .set_default(64_K)
.set_description(""),
- Option("filestore_max_xattr_value_size_btrfs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(64<<10)
+ Option("filestore_max_xattr_value_size_btrfs", Option::TYPE_SIZE, Option::LEVEL_DEV)
+ .set_default(64_K)
.set_description(""),
- Option("filestore_max_xattr_value_size_other", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(1<<10)
+ Option("filestore_max_xattr_value_size_other", Option::TYPE_SIZE, Option::LEVEL_DEV)
+ .set_default(1_K)
.set_description(""),
- Option("filestore_sloppy_crc", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("filestore_sloppy_crc", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
- Option("filestore_sloppy_crc_block_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("filestore_sloppy_crc_block_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(65536)
.set_description(""),
- Option("filestore_max_alloc_hint_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_max_alloc_hint_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(1ULL << 20)
.set_description(""),
Option("filestore_max_sync_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("Period between calls to syncfs(2) and journal trims (seconds)"),
- Option("filestore_min_sync_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("filestore_min_sync_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(.01)
- .set_description(""),
+ .set_description("Minimum period between calls to syncfs(2)"),
- Option("filestore_btrfs_snap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("filestore_btrfs_snap", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
.set_description(""),
Option("filestore_btrfs_clone_range", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("Use btrfs clone_range ioctl to efficiently duplicate objects"),
- Option("filestore_zfs_snap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("filestore_zfs_snap", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
- Option("filestore_fsync_flushes_journal_data", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("filestore_fsync_flushes_journal_data", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
Option("filestore_fiemap", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("Use fiemap ioctl(2) to determine which parts of objects are sparse"),
Option("filestore_punch_hole", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("Use fallocate(2) FALLOC_FL_PUNCH_HOLE to efficiently zero ranges of objects"),
Option("filestore_seek_data_hole", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("Use lseek(2) SEEK_HOLE and SEEK_DATA to determine which parts of objects are sparse"),
Option("filestore_splice", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("Use splice(2) to more efficiently copy data between files"),
Option("filestore_fadvise", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("Use posix_fadvise(2) to pass hints to file system"),
Option("filestore_collect_device_partition_information", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("Collect metadata about the backing file system on OSD startup"),
Option("filestore_xfs_extsize", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("Use XFS extsize ioctl(2) to hint allocator about expected write sizes"),
- Option("filestore_journal_parallel", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("filestore_journal_parallel", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
- Option("filestore_journal_writeahead", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("filestore_journal_writeahead", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
- Option("filestore_journal_trailing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("filestore_journal_trailing", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
Option("filestore_queue_max_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(50)
- .set_description(""),
+ .set_description("Max IO operations in flight"),
- Option("filestore_queue_max_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(100 << 20)
- .set_description(""),
+ Option("filestore_queue_max_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(100_M)
+ .set_description("Max (written) bytes in flight"),
- Option("filestore_caller_concurrency", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("filestore_caller_concurrency", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(10)
.set_description(""),
Option("filestore_expected_throughput_bytes", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
- .set_default(200 << 20)
- .set_description(""),
+ .set_default(200_M)
+ .set_description("Expected throughput of backend device (aids throttling calculations)"),
Option("filestore_expected_throughput_ops", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(200)
+ .set_description("Expected through of backend device in IOPS (aids throttling calculations)"),
+
+ Option("filestore_queue_max_delay_multiple", Option::TYPE_FLOAT, Option::LEVEL_DEV)
+ .set_default(0)
+ .set_description(""),
+
+ Option("filestore_queue_high_delay_multiple", Option::TYPE_FLOAT, Option::LEVEL_DEV)
+ .set_default(0)
+ .set_description(""),
+
+ Option("filestore_queue_max_delay_multiple_bytes", Option::TYPE_FLOAT, Option::LEVEL_DEV)
+ .set_default(0)
+ .set_description(""),
+
+ Option("filestore_queue_high_delay_multiple_bytes", Option::TYPE_FLOAT, Option::LEVEL_DEV)
+ .set_default(0)
.set_description(""),
- Option("filestore_queue_max_delay_multiple", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("filestore_queue_max_delay_multiple_ops", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("filestore_queue_high_delay_multiple", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("filestore_queue_high_delay_multiple_ops", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("filestore_queue_low_threshhold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("filestore_queue_low_threshhold", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0.3)
.set_description(""),
- Option("filestore_queue_high_threshhold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("filestore_queue_high_threshhold", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0.9)
.set_description(""),
Option("filestore_op_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(2)
- .set_description(""),
+ .set_description("Threads used to apply changes to backing file system"),
Option("filestore_op_thread_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(60)
- .set_description(""),
+ .set_description("Seconds before a worker thread is considered stalled"),
Option("filestore_op_thread_suicide_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(180)
- .set_description(""),
+ .set_description("Seconds before a worker thread is considered dead"),
Option("filestore_commit_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(600)
- .set_description(""),
+ .set_description("Seconds before backing file system is considered hung"),
- Option("filestore_fiemap_threshold", Option::TYPE_INT, Option::LEVEL_DEV)
+ Option("filestore_fiemap_threshold", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(4_K)
.set_description(""),
- Option("filestore_merge_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("filestore_merge_threshold", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(-10)
.set_description(""),
- Option("filestore_split_multiple", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("filestore_split_multiple", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(2)
.set_description(""),
- Option("filestore_split_rand_factor", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("filestore_split_rand_factor", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(20)
.set_description(""),
- Option("filestore_update_to", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("filestore_update_to", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(1000)
.set_description(""),
- Option("filestore_blackhole", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("filestore_blackhole", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
- Option("filestore_fd_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("filestore_fd_cache_size", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(128)
.set_description(""),
- Option("filestore_fd_cache_shards", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("filestore_fd_cache_shards", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(16)
.set_description(""),
- Option("filestore_ondisk_finisher_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("filestore_ondisk_finisher_threads", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(1)
.set_description(""),
- Option("filestore_apply_finisher_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("filestore_apply_finisher_threads", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(1)
.set_description(""),
- Option("filestore_dump_file", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ Option("filestore_dump_file", Option::TYPE_STR, Option::LEVEL_DEV)
.set_default("")
.set_description(""),
.set_default(0)
.set_description(""),
- Option("filestore_fail_eio", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("filestore_fail_eio", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
.set_description(""),
.set_default(false)
.set_description(""),
- Option("journal_dio", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("journal_dio", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
.set_description(""),
- Option("journal_aio", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("journal_aio", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
.set_description(""),
- Option("journal_force_aio", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("journal_force_aio", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
- Option("journal_block_size", Option::TYPE_INT, Option::LEVEL_DEV)
+ Option("journal_block_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(4_K)
.set_description(""),
- Option("journal_max_corrupt_search", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(10<<20)
- .set_description(""),
-
- Option("journal_block_align", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("journal_block_align", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(true)
.set_description(""),
- Option("journal_write_header_frequency", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("journal_write_header_frequency", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("journal_max_write_bytes", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(10 << 20)
- .set_description(""),
+ Option("journal_max_write_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(10_M)
+ .set_description("Max bytes in flight to journal"),
Option("journal_max_write_entries", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(100)
- .set_description(""),
+ .set_description("Max IOs in flight to journal"),
- Option("journal_throttle_low_threshhold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("journal_throttle_low_threshhold", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0.6)
.set_description(""),
- Option("journal_throttle_high_threshhold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("journal_throttle_high_threshhold", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0.9)
.set_description(""),
- Option("journal_throttle_high_multiple", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("journal_throttle_high_multiple", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("journal_throttle_max_multiple", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("journal_throttle_max_multiple", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("journal_align_min_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(64 << 10)
+ Option("journal_align_min_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
+ .set_default(64_K)
.set_description(""),
- Option("journal_replay_from", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("journal_replay_from", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
.set_min_max((int64_t)PerfCountersBuilder::PRIO_DEBUGONLY,
(int64_t)PerfCountersBuilder::PRIO_CRITICAL + 1),
- Option("journal_zero_on_create", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("journal_zero_on_create", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
- Option("journal_ignore_corruption", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("journal_ignore_corruption", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
- Option("journal_discard", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("journal_discard", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
.set_description(""),
Option("mgr_module_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default(CEPH_PKGLIBDIR "/mgr")
+ .set_default(CEPH_DATADIR "/mgr")
.add_service("mgr")
.set_description("Filesystem path to manager modules."),
Option("mgr_initial_modules", Option::TYPE_STR, Option::LEVEL_BASIC)
- .set_default("restful status balancer")
+ .set_default("restful iostat")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
+ .set_flag(Option::FLAG_CLUSTER_CREATE)
.add_service("mon")
.set_description("List of manager modules to enable when the cluster is "
"first started")
Option("mgr_data", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("/var/lib/ceph/mgr/$cluster-$id")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.add_service("mgr")
.set_description("Filesystem path to the ceph-mgr data directory, used to "
"contain keyring."),
- Option("mgr_tick_period", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mgr_tick_period", Option::TYPE_SECS, Option::LEVEL_ADVANCED)
.set_default(2)
.add_service("mgr")
.set_description("Period in seconds of beacon messages to monitor"),
"if you simply do not require the most up to date "
"performance counter data."),
- Option("mgr_client_bytes", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("mgr_client_bytes", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(128_M)
.add_service("mgr"),
.set_default(512)
.add_service("mgr"),
- Option("mgr_osd_bytes", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("mgr_osd_bytes", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(512_M)
.add_service("mgr"),
.set_default(8192)
.add_service("mgr"),
- Option("mgr_mds_bytes", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("mgr_mds_bytes", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(128_M)
.add_service("mgr"),
.set_default(128)
.add_service("mgr"),
- Option("mgr_mon_bytes", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("mgr_mon_bytes", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(128_M)
.add_service("mgr"),
.set_description("Period in seconds from last beacon to manager dropping "
"state about a monitored service (RGW, rbd-mirror etc)"),
+ Option("mgr_client_service_daemon_unregister_timeout", Option::TYPE_FLOAT, Option::LEVEL_DEV)
+ .set_default(1.0)
+ .set_description("Time to wait during shutdown to deregister service with mgr"),
+
+ Option("mgr_debug_aggressive_pg_num_changes", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ .set_default(false)
+ .set_description("Bypass most throttling and safety checks in pg[p]_num controller")
+ .add_service("mgr"),
+
Option("mon_mgr_digest_period", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(5)
.add_service("mon")
.set_description("Period in seconds between monitor-to-manager "
"health/status updates"),
- Option("mon_mgr_beacon_grace", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mon_mgr_beacon_grace", Option::TYPE_SECS, Option::LEVEL_ADVANCED)
.set_default(30)
.add_service("mon")
.set_description("Period in seconds from last beacon to monitor marking "
"failures about managers that aren't online yet"),
Option("mon_mgr_mkfs_grace", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(60)
+ .set_default(120)
.add_service("mon")
.set_description("Period in seconds that the cluster may have no active "
"manager before this is reported as an ERR rather than "
"a WARN"),
- Option("mutex_perf_counter", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
- .set_description(""),
-
Option("throttler_perf_counter", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
.set_description(""),
.set_default(false)
.set_description(""),
- Option("internal_safe_to_start_threads", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
- .set_description(""),
-
Option("debug_deliberately_leak_memory", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
-
+
Option("debug_asserts_on_shutdown", Option::TYPE_BOOL,Option::LEVEL_DEV)
.set_default(false)
.set_description("Enable certain asserts to check for refcounting bugs on shutdown; see http://tracker.ceph.com/issues/21738"),
+
+ Option("debug_asok_assert_abort", Option::TYPE_BOOL, Option::LEVEL_DEV)
+ .set_default(false)
+ .set_description("allow commands 'assert' and 'abort' via asok for testing crash dumps etc"),
+
+ Option("target_max_misplaced_ratio", Option::TYPE_FLOAT, Option::LEVEL_BASIC)
+ .set_default(.05)
+ .set_description("Max ratio of misplaced objects to target when throttling data rebalancing activity"),
+
+ Option("device_failure_prediction_mode", Option::TYPE_STR, Option::LEVEL_BASIC)
+ .set_default("none")
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_enum_allowed({"none", "local", "cloud"})
+ .set_description("Method used to predict device failures")
+ .set_long_description("To disable prediction, use 'none', 'local' uses a prediction model that runs inside the mgr daemon. 'cloud' will share metrics with a cloud service and query the service for devicelife expectancy."),
+
+ /* KRB Authentication. */
+ Option("gss_ktab_client_file", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("/var/lib/ceph/$name/gss_client_$name.ktab")
+ .set_description("GSS/KRB5 Keytab file for client authentication")
+ .add_service({"mon", "osd"})
+ .set_long_description("This sets the full path for the GSS/Kerberos client keytab file location."),
+
+ Option("gss_target_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("ceph")
+ .set_description("")
+ .add_service({"mon", "osd"})
+ .set_long_description("This sets the gss target service name."),
});
}
.set_default(100)
.set_description("Max number of ACL grants in a single request"),
- Option("rgw_max_chunk_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rgw_cors_rules_max_num", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(100)
+ .set_description("Max number of cors rules in a single request"),
+
+ Option("rgw_delete_multi_obj_max_num", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(1000)
+ .set_description("Max number of objects in a single multi-object delete request"),
+
+ Option("rgw_website_routing_rules_max_num", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(50)
+ .set_description("Max number of website routing rules in a single request"),
+
+ Option("rgw_rados_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description("true if LTTng-UST tracepoints should be enabled"),
+
+ Option("rgw_op_tracing", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description("true if LTTng-UST tracepoints should be enabled"),
+
+ Option("rgw_max_chunk_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(4_M)
.set_description("Set RGW max chunk size")
.set_long_description(
"need to be atomic, and anything larger than this would require more than a single "
"operation."),
- Option("rgw_put_obj_min_window_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rgw_put_obj_min_window_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(16_M)
.set_description("The minimum RADOS write window size (in bytes).")
.set_long_description(
"in order to better utilize the pipe.")
.add_see_also({"rgw_put_obj_max_window_size", "rgw_max_chunk_size"}),
- Option("rgw_put_obj_max_window_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rgw_put_obj_max_window_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(64_M)
.set_description("The maximum RADOS write window size (in bytes).")
.set_long_description("The window size may be dynamically adjusted, but will not surpass this value.")
.add_see_also({"rgw_put_obj_min_window_size", "rgw_max_chunk_size"}),
- Option("rgw_max_put_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("rgw_max_put_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(5_G)
.set_description("Max size (in bytes) of regular (non multi-part) object upload.")
.set_long_description(
"objects, a special upload mechanism is required. The S3 API provides the "
"multi-part upload, and Swift provides DLO and SLO."),
- Option("rgw_max_put_param_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("rgw_max_put_param_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1_M)
.set_description("The maximum size (in bytes) of data input of certain RESTful requests."),
- Option("rgw_max_attr_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("rgw_max_attr_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("The maximum length of metadata value. 0 skips the check"),
- Option("rgw_max_attr_name_len", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("rgw_max_attr_name_len", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("The maximum length of metadata name. 0 skips the check"),
Option("rgw_enable_lc_threads", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description("Enables the lifecycle maintenance thread. This is required on at least on rgw for each zone.")
+ .set_description("Enables the lifecycle maintenance thread. This is required on at least one rgw for each zone.")
.set_long_description(
"The lifecycle maintenance thread is responsible for lifecycle related maintenance "
"work. The thread itself can be disabled, but in order for lifecycle to work "
Option("rgw_data", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("/var/lib/ceph/radosgw/$cluster-$id")
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
.set_description("Alternative location for RGW configuration.")
.set_long_description(
"If this is set, the different Ceph system configurables (such as the keyring file "
"will be located in the path that is specified here. "),
Option("rgw_enable_apis", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("s3, s3website, swift, swift_auth, admin")
+ .set_default("s3, s3website, swift, swift_auth, admin, sts")
.set_description("A list of set of RESTful APIs that rgw handles."),
Option("rgw_cache_enabled", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
"This is needed for virtual hosting of buckets, unless configured via zonegroup "
"configuration."),
+ Option("rgw_service_provider_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("")
+ .set_description("Service provider name which is contained in http response headers")
+ .set_long_description(
+ "As S3 or other cloud storage providers do, http response headers should contain the name of the provider. "
+ "This name will be placed in http header 'Server'."),
+
Option("rgw_content_length_compat", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
.set_description("Multiple content length headers compatibility")
"Try to handle requests with abiguous multiple content length headers "
"(Content-Length, Http-Content-Length)."),
+ Option("rgw_relaxed_region_enforcement", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description("Disable region constraint enforcement")
+ .set_long_description(
+ "Enable requests such as bucket creation to succeed irrespective of region restrictions (Jewel compat)."),
+
Option("rgw_lifecycle_work_time", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("00:00-06:00")
.set_description("Lifecycle allowed work time")
.set_default(60)
.set_description(""),
+ Option("rgw_lc_thread_delay", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("Delay after processing of bucket listing chunks (i.e., per 1000 entries) in milliseconds"),
+
Option("rgw_lc_max_objs", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(32)
.set_description("Number of lifecycle data shards")
Option("rgw_keystone_admin_token", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description("The admin token (shared secret) that is used for the Keystone requests."),
+ .set_description("DEPRECATED: The admin token (shared secret) that is used for the Keystone requests."),
+
+ Option("rgw_keystone_admin_token_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("")
+ .set_description("Path to a file containing the admin token (shared secret) that is used for the Keystone requests."),
Option("rgw_keystone_admin_user", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
Option("rgw_keystone_admin_password", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description("Keystone admin password."),
+ .set_description("DEPRECATED: Keystone admin password."),
+
+ Option("rgw_keystone_admin_password_path", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("")
+ .set_description("Path to a file containing the Keystone admin password."),
Option("rgw_keystone_admin_tenant", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
.set_default(true)
.set_description("Should RGW verify the Keystone server SSL certificate."),
- Option("rgw_keystone_implicit_tenants", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("false")
- .set_enum_allowed( { "false", "true", "swift", "s3", "both", "0", "1", "none" } )
+ Option("rgw_keystone_implicit_tenants", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
.set_description("RGW Keystone implicit tenants creation")
.set_long_description(
"Implicitly create new users in their own tenant with the same name when "
- "authenticating via Keystone. Can be limited to s3 or swift only."),
+ "authenticating via Keystone."),
Option("rgw_cross_domain_policy", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("<allow-access-from domain=\"*\" secure=\"false\" />")
.set_description("Should S3 authentication use Keystone."),
Option("rgw_s3_auth_order", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("external, local")
+ .set_default("sts, external, local")
.set_description("Authentication strategy order to use for s3 authentication")
.set_long_description(
"Order of authentication strategies to try for s3 authentication, the allowed "
.set_default("")
.set_description("LDAP search filter."),
+ Option("rgw_opa_url", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("")
+ .set_description("URL to OPA server."),
+
+ Option("rgw_opa_token", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("")
+ .set_description("The Bearer token OPA uses to authenticate client requests."),
+
+ Option("rgw_opa_verify_ssl", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .set_description("Should RGW verify the OPA server SSL certificate."),
+
+ Option("rgw_use_opa_authz", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description("Should OPA be used to authorize client requests."),
+
Option("rgw_admin_entry", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("admin")
.set_description("Path prefix to be used for accessing RGW RESTful admin API."),
.set_default(".rgw.root")
.set_description("Period root pool")
.set_long_description(
- "The realm root pool, is the pool where the RGW realm configuration located."
+ "The period root pool, is the pool where the RGW period configuration located."
)
.add_see_also({"rgw_zonegroup_root_pool", "rgw_zone_root_pool", "rgw_realm_root_pool"}),
"RGW will send ops log data through it.")
.add_see_also({"rgw_enable_ops_log", "rgw_ops_log_data_backlog"}),
- Option("rgw_ops_log_data_backlog", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rgw_ops_log_data_backlog", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(5 << 20)
.set_description("Ops log socket backlog")
.set_long_description(
.set_long_description(
"The number of garbage collector data shards, is the number of RADOS objects that "
"RGW will use to store the garbage collection information on.")
- .add_see_also({"rgw_gc_obj_min_wait", "rgw_gc_processor_max_time", "rgw_gc_processor_period"}),
+ .add_see_also({"rgw_gc_obj_min_wait", "rgw_gc_processor_max_time", "rgw_gc_processor_period", "rgw_gc_max_concurrent_io"}),
Option("rgw_gc_obj_min_wait", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(2_hr)
- .set_description("Garabge collection object expiration time")
+ .set_description("Garbage collection object expiration time")
.set_long_description(
"The length of time (in seconds) that the RGW collector will wait before purging "
"a deleted object's data. RGW will not remove object immediately, as object could "
"still have readers. A mechanism exists to increase the object's expiration time "
"when it's being read.")
- .add_see_also({"rgw_gc_max_objs", "rgw_gc_processor_max_time", "rgw_gc_processor_period"}),
+ .add_see_also({"rgw_gc_max_objs", "rgw_gc_processor_max_time", "rgw_gc_processor_period", "rgw_gc_max_concurrent_io"}),
Option("rgw_gc_processor_max_time", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1_hr)
"Garbage collection thread in RGW process holds a lease on its data shards. These "
"objects contain the information about the objects that need to be removed. RGW "
"takes a lease in order to prevent multiple RGW processes from handling the same "
- "objects concurrently. This time signifies that maximum amount of time that RGW "
+ "objects concurrently. This time signifies that maximum amount of time (in seconds) that RGW "
"is allowed to hold that lease. In the case where RGW goes down uncleanly, this "
"is the amount of time where processing of that data shard will be blocked.")
- .add_see_also({"rgw_gc_max_objs", "rgw_gc_obj_min_wait", "rgw_gc_processor_period"}),
+ .add_see_also({"rgw_gc_max_objs", "rgw_gc_obj_min_wait", "rgw_gc_processor_period", "rgw_gc_max_concurrent_io"}),
Option("rgw_gc_processor_period", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1_hr)
"The amount of time between the start of consecutive runs of the garbage collector "
"threads. If garbage collector runs takes more than this period, it will not wait "
"before running again.")
- .add_see_also({"rgw_gc_max_objs", "rgw_gc_obj_min_wait", "rgw_gc_processor_max_time"}),
+ .add_see_also({"rgw_gc_max_objs", "rgw_gc_obj_min_wait", "rgw_gc_processor_max_time", "rgw_gc_max_concurrent_io", "rgw_gc_max_trim_chunk"}),
+
+ Option("rgw_gc_max_concurrent_io", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(10)
+ .set_description("Max concurrent RADOS IO operations for garbage collection")
+ .set_long_description(
+ "The maximum number of concurrent IO operations that the RGW garbage collection "
+ "thread will use when purging old data.")
+ .add_see_also({"rgw_gc_max_objs", "rgw_gc_obj_min_wait", "rgw_gc_processor_max_time", "rgw_gc_max_trim_chunk"}),
+
+ Option("rgw_gc_max_trim_chunk", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(16)
+ .set_description("Max number of keys to remove from garbage collector log in a single operation")
+ .add_see_also({"rgw_gc_max_objs", "rgw_gc_obj_min_wait", "rgw_gc_processor_max_time", "rgw_gc_max_concurrent_io"}),
Option("rgw_s3_success_create_obj_status", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("HTTP return code override for object creation")
.set_long_description(
- "If not zero, this is the HTTP return code that will be returned on a succesful S3 "
+ "If not zero, this is the HTTP return code that will be returned on a successful S3 "
"object creation."),
Option("rgw_resolve_cname", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
"it will use it instead. This gives user to have the ability of creating a unique "
"domain of their own to point at data in their bucket."),
- Option("rgw_obj_stripe_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rgw_obj_stripe_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(4_M)
.set_description("RGW object stripe size")
.set_long_description(
.set_description("RGW shutdown timeout")
.set_long_description("Number of seconds to wait for a process before exiting unconditionally."),
- Option("rgw_get_obj_window_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rgw_get_obj_window_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(16_M)
.set_description("RGW object read window size")
.set_long_description("The window size in bytes for a single object read request"),
- Option("rgw_get_obj_max_req_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rgw_get_obj_max_req_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(4_M)
.set_description("RGW object read chunk size")
.set_long_description(
"the metadata sync parallelism as a shard can only be processed by a single "
"RGW at a time"),
- Option("rgw_num_zone_opstate_shards", Option::TYPE_INT, Option::LEVEL_DEV)
- .set_default(128)
- .set_description(""),
-
- Option("rgw_opstate_ratelimit_sec", Option::TYPE_INT, Option::LEVEL_DEV)
- .set_default(30)
- .set_description(""),
-
Option("rgw_curl_wait_timeout_ms", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(1000)
.set_description(""),
.set_long_description(
"If true, RGW will send progress information when copy operation is executed. "),
- Option("rgw_copy_obj_progress_every_bytes", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rgw_copy_obj_progress_every_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(1_M)
.set_description("Send copy-object progress info after these many bytes"),
.set_default("data_log")
.set_description(""),
- Option("rgw_replica_log_obj_prefix", Option::TYPE_STR, Option::LEVEL_DEV)
- .set_default("replica_log")
- .set_description(""),
-
Option("rgw_bucket_quota_ttl", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(600)
.set_description("Bucket quota stats cache TTL")
"contain the name of the bucket the operation happened on."),
Option("rgw_frontends", Option::TYPE_STR, Option::LEVEL_BASIC)
- .set_default("civetweb port=7480")
+ .set_default("beast port=7480")
.set_description("RGW frontends configuration")
.set_long_description(
"A comma delimited list of frontends configuration. Each configuration contains "
"The default quota configuration for total size of objects for a single user. A "
"negative number means 'unlimited'."),
- Option("rgw_multipart_min_part_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rgw_multipart_min_part_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(5_M)
.set_description("Minimum S3 multipart-upload part size")
.set_long_description(
.set_description("Length of time to aggregate metadata changes")
.set_long_description(
"Length of time (in milliseconds) in which the master zone aggregates all the "
- "metadata changes that occured, before sending notifications to all the other "
+ "metadata changes that occurred, before sending notifications to all the other "
"zones."),
Option("rgw_run_sync_thread", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description(""),
+ Option("rgw_sync_trace_history_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(4096)
+ .set_description("Sync trace history size")
+ .set_long_description(
+ "Maximum number of complete sync trace entries to keep."),
+
+ Option("rgw_sync_trace_per_node_log_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(32)
+ .set_description("Sync trace per-node log size")
+ .set_long_description(
+ "The number of log entries to keep per sync-trace node."),
+
+ Option("rgw_sync_trace_servicemap_update_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default(10)
+ .set_description("Sync-trace service-map update interval")
+ .set_long_description(
+ "Number of seconds between service-map updates of sync-trace events."),
+
Option("rgw_period_push_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(2)
.set_description("Period push interval")
.set_default(true)
.set_description("Enable stats on bucket listing in Swift"),
- Option("rgw_reshard_num_logs", Option::TYPE_INT, Option::LEVEL_DEV)
+ Option("rgw_reshard_num_logs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(16)
- .set_description(""),
+ .set_min(1)
+ .set_description("")
+ .add_service("rgw"),
- Option("rgw_reshard_bucket_lock_duration", Option::TYPE_INT, Option::LEVEL_DEV)
- .set_default(120)
- .set_description(""),
+ Option("rgw_reshard_bucket_lock_duration", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(360)
+ .set_min(30)
+ .set_description("Number of seconds the timeout on the reshard locks (bucket reshard lock and reshard log lock) are set to. As a reshard proceeds these locks can be renewed/extended. If too short, reshards cannot complete and will fail, causing a future reshard attempt. If too long a hung or crashed reshard attempt will keep the bucket locked for an extended period, not allowing RGW to detect the failed reshard attempt and recover.")
+ .add_tag("performance")
+ .add_service("rgw"),
+
+ Option("rgw_reshard_batch_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(64)
+ .set_min(8)
+ .set_description("Number of reshard entries to batch together before sending the operations to the CLS back-end")
+ .add_tag("performance")
+ .add_service("rgw"),
+
+ Option("rgw_reshard_max_aio", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(128)
+ .set_min(16)
+ .set_description("Maximum number of outstanding asynchronous I/O operations to allow at a time during resharding")
+ .add_tag("performance")
+ .add_service("rgw"),
Option("rgw_trust_forwarded_https", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
Option("rgw_torrent_tracker", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description("Torrent field annouce and annouce list"),
+ .set_description("Torrent field announce and announce list"),
Option("rgw_torrent_createby", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
.set_default("")
.set_description("Torrent origin"),
- Option("rgw_torrent_sha_unit", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rgw_torrent_sha_unit", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(512*1024)
.set_description(""),
"a high number of objects per shard.")
.add_see_also("rgw_max_objs_per_shard"),
- Option("rgw_max_objs_per_shard", Option::TYPE_INT, Option::LEVEL_BASIC)
+ Option("rgw_max_objs_per_shard", Option::TYPE_UINT, Option::LEVEL_BASIC)
.set_default(100000)
.set_description("Max objects per shard for dynamic resharding")
.set_long_description(
Option("rgw_reshard_thread_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(10_min)
- .set_description(""),
+ .set_min(10_min)
+ .set_description("Number of seconds between processing of reshard log entries"),
Option("rgw_cache_expiry_interval", Option::TYPE_UINT,
Option::LEVEL_ADVANCED)
- .set_default(900)
- .set_description("Number of seconds before entries in the bucket info "
- "cache are assumed stale and re-fetched. Zero is never.")
+ .set_default(15_min)
+ .set_description("Number of seconds before entries in the cache are "
+ "assumed stale and re-fetched. Zero is never.")
.add_tag("performance")
.add_service("rgw")
- .set_long_description("The Rados Gateway stores metadata about buckets in "
+ .set_long_description("The Rados Gateway stores metadata and objects in "
"an internal cache. This should be kept consistent "
"by the OSD's relaying notify events between "
"multiple watching RGW processes. In the event "
"of RGW instances under heavy use. If you would like "
"to turn off cache expiry, set this value to zero."),
+ Option("rgw_inject_notify_timeout_probability", Option::TYPE_FLOAT,
+ Option::LEVEL_DEV)
+ .set_default(0)
+ .add_tag("fault injection")
+ .add_tag("testing")
+ .add_service("rgw")
+ .set_min_max(0.0, 1.0)
+ .set_description("Likelihood of ignoring a notify")
+ .set_long_description("This is the probability that the RGW cache will "
+ "ignore a cache notify message. It exists to help "
+ "with the development and testing of cache "
+ "consistency and recovery improvements. Please "
+ "do not set it in a production cluster, as it "
+ "actively causes failures. Set this to a floating "
+ "point value between 0 and 1."),
+ Option("rgw_max_notify_retries", Option::TYPE_UINT,
+ Option::LEVEL_ADVANCED)
+ .set_default(3)
+ .add_tag("error recovery")
+ .add_service("rgw")
+ .set_description("Number of attempts to notify peers before giving up.")
+ .set_long_description("The number of times we will attempt to update "
+ "a peer's cache in the event of error before giving "
+ "up. This is unlikely to be an issue unless your "
+ "cluster is very heavily loaded. Beware that "
+ "increasing this value may cause some operations to "
+ "take longer in exceptional cases and thus may, "
+ "rarely, cause clients to time out."),
+ Option("rgw_sts_entry", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("sts")
+ .set_description("STS URL prefix")
+ .set_long_description("URL path prefix for internal STS requests."),
+
+ Option("rgw_sts_key", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("sts")
+ .set_description("STS Key")
+ .set_long_description("Key used for encrypting/ decrypting session token."),
+
+ Option("rgw_s3_auth_use_sts", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description("Should S3 authentication use STS."),
+
+ Option("rgw_sts_max_session_duration", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(43200)
+ .set_description("Session token max duration")
+ .set_long_description("Max duration in seconds for which the session token is valid."),
+
Option("rgw_max_listing_results", Option::TYPE_UINT,
Option::LEVEL_ADVANCED)
.set_default(1000)
"ListBucketVersions(max-keys), "
"ListBucketMultiPartUploads(max-uploads), "
"ListMultipartUploadParts(max-parts)"),
+
+ Option("rgw_sts_token_introspection_url", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("")
+ .set_description("STS Web Token introspection URL")
+ .set_long_description("URL for introspecting an STS Web Token."),
+
+ Option("rgw_sts_client_id", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("")
+ .set_description("Client Id")
+ .set_long_description("Client Id needed for introspecting a Web Token."),
+
+ Option("rgw_sts_client_secret", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("")
+ .set_description("Client Secret")
+ .set_long_description("Client Secret needed for introspecting a Web Token."),
+
+ Option("rgw_max_concurrent_requests", Option::TYPE_INT, Option::LEVEL_BASIC)
+ .set_default(1024)
+ .set_description("Maximum number of concurrent HTTP requests.")
+ .set_long_description(
+ "Maximum number of concurrent HTTP requests that the beast frontend "
+ "will process. Tuning this can help to limit memory usage under heavy "
+ "load.")
+ .add_tag("performance")
+ .add_see_also("rgw_frontends"),
+
+ Option("rgw_scheduler_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("throttler")
+ .set_description("Set the type of dmclock scheduler, defaults to throttler "
+ "Other valid values are dmclock which is experimental"),
+
+ Option("rgw_dmclock_admin_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(100.0)
+ .set_description("mclock reservation for admin requests")
+ .add_see_also("rgw_dmclock_admin_wgt")
+ .add_see_also("rgw_dmclock_admin_lim"),
+
+ Option("rgw_dmclock_admin_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(100.0)
+ .set_description("mclock weight for admin requests")
+ .add_see_also("rgw_dmclock_admin_res")
+ .add_see_also("rgw_dmclock_admin_lim"),
+
+ Option("rgw_dmclock_admin_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.0)
+ .set_description("mclock limit for admin requests")
+ .add_see_also("rgw_dmclock_admin_res")
+ .add_see_also("rgw_dmclock_admin_wgt"),
+
+ Option("rgw_dmclock_auth_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(200.0)
+ .set_description("mclock reservation for object data requests")
+ .add_see_also("rgw_dmclock_auth_wgt")
+ .add_see_also("rgw_dmclock_auth_lim"),
+
+ Option("rgw_dmclock_auth_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(100.0)
+ .set_description("mclock weight for object data requests")
+ .add_see_also("rgw_dmclock_auth_res")
+ .add_see_also("rgw_dmclock_auth_lim"),
+
+ Option("rgw_dmclock_auth_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.0)
+ .set_description("mclock limit for object data requests")
+ .add_see_also("rgw_dmclock_auth_res")
+ .add_see_also("rgw_dmclock_auth_wgt"),
+
+ Option("rgw_dmclock_data_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(500.0)
+ .set_description("mclock reservation for object data requests")
+ .add_see_also("rgw_dmclock_data_wgt")
+ .add_see_also("rgw_dmclock_data_lim"),
+
+ Option("rgw_dmclock_data_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(500.0)
+ .set_description("mclock weight for object data requests")
+ .add_see_also("rgw_dmclock_data_res")
+ .add_see_also("rgw_dmclock_data_lim"),
+
+ Option("rgw_dmclock_data_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.0)
+ .set_description("mclock limit for object data requests")
+ .add_see_also("rgw_dmclock_data_res")
+ .add_see_also("rgw_dmclock_data_wgt"),
+
+ Option("rgw_dmclock_metadata_res", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(500.0)
+ .set_description("mclock reservation for metadata requests")
+ .add_see_also("rgw_dmclock_metadata_wgt")
+ .add_see_also("rgw_dmclock_metadata_lim"),
+
+ Option("rgw_dmclock_metadata_wgt", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(500.0)
+ .set_description("mclock weight for metadata requests")
+ .add_see_also("rgw_dmclock_metadata_res")
+ .add_see_also("rgw_dmclock_metadata_lim"),
+
+ Option("rgw_dmclock_metadata_lim", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0.0)
+ .set_description("mclock limit for metadata requests")
+ .add_see_also("rgw_dmclock_metadata_res")
+ .add_see_also("rgw_dmclock_metadata_wgt"),
+
});
}
.set_default("rbd")
.set_description("default pool for storing new images")
.set_validator([](std::string *value, std::string *error_message){
- boost::regex pattern("^[^@/]+$");
- if (!boost::regex_match (*value, pattern)) {
+ std::regex pattern("^[^@/]+$");
+ if (!std::regex_match (*value, pattern)) {
*value = "rbd";
*error_message = "invalid RBD default pool, resetting to 'rbd'";
}
.set_default("")
.set_description("default pool for storing data blocks for new images")
.set_validator([](std::string *value, std::string *error_message){
- boost::regex pattern("^[^@/]*$");
- if (!boost::regex_match (*value, pattern)) {
+ std::regex pattern("^[^@/]*$");
+ if (!std::regex_match (*value, pattern)) {
*value = "";
*error_message = "ignoring invalid RBD data pool";
}
"name is as follows: +1 -> layering, +2 -> striping, "
"+4 -> exclusive-lock, +8 -> object-map, +16 -> fast-diff, "
"+32 -> deep-flatten, +64 -> journaling, +128 -> data-pool")
- .set_safe()
- .set_validator([](std::string *value, std::string *error_message){
- static const std::map<std::string, uint64_t> FEATURE_MAP = {
- {RBD_FEATURE_NAME_LAYERING, RBD_FEATURE_LAYERING},
- {RBD_FEATURE_NAME_STRIPINGV2, RBD_FEATURE_STRIPINGV2},
- {RBD_FEATURE_NAME_EXCLUSIVE_LOCK, RBD_FEATURE_EXCLUSIVE_LOCK},
- {RBD_FEATURE_NAME_OBJECT_MAP, RBD_FEATURE_OBJECT_MAP},
- {RBD_FEATURE_NAME_FAST_DIFF, RBD_FEATURE_FAST_DIFF},
- {RBD_FEATURE_NAME_DEEP_FLATTEN, RBD_FEATURE_DEEP_FLATTEN},
- {RBD_FEATURE_NAME_JOURNALING, RBD_FEATURE_JOURNALING},
- {RBD_FEATURE_NAME_DATA_POOL, RBD_FEATURE_DATA_POOL},
- };
- static_assert((RBD_FEATURE_DATA_POOL << 1) > RBD_FEATURES_ALL,
- "new RBD feature added");
-
- // convert user-friendly comma delimited feature name list to a bitmask
- // that is used by the librbd API
- uint64_t features = 0;
- error_message->clear();
-
- try {
- features = boost::lexical_cast<decltype(features)>(*value);
-
- uint64_t unsupported_features = (features & ~RBD_FEATURES_ALL);
- if (unsupported_features != 0ull) {
- features &= RBD_FEATURES_ALL;
-
- std::stringstream ss;
- ss << "ignoring unknown feature mask 0x"
- << std::hex << unsupported_features;
- *error_message = ss.str();
- }
- } catch (const boost::bad_lexical_cast& ) {
- int r = 0;
- std::vector<std::string> feature_names;
- boost::split(feature_names, *value, boost::is_any_of(","));
- for (auto feature_name: feature_names) {
- boost::trim(feature_name);
- auto feature_it = FEATURE_MAP.find(feature_name);
- if (feature_it != FEATURE_MAP.end()) {
- features += feature_it->second;
- } else {
- if (!error_message->empty()) {
- *error_message += ", ";
- }
- *error_message += "ignoring unknown feature " + feature_name;
- r = -EINVAL;
- }
- }
-
- if (features == 0 && r == -EINVAL) {
- features = RBD_FEATURES_DEFAULT;
- }
- }
- *value = stringify(features);
- return 0;
- }),
-
- Option("rbd_op_threads", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_flag(Option::FLAG_RUNTIME)
+ .set_validator([](std::string *value, std::string *error_message) {
+ ostringstream ss;
+ uint64_t features = librbd::rbd_features_from_string(*value, &ss);
+ // Leave this in integer form to avoid breaking Cinder. Someday
+ // we would like to present this in string form instead...
+ *value = stringify(features);
+ if (ss.str().size()) {
+ return -EINVAL;
+ }
+ return 0;
+ }),
+
+ Option("rbd_op_threads", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1)
.set_description("number of threads to utilize for internal processing"),
- Option("rbd_op_thread_timeout", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_op_thread_timeout", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(60)
.set_description("time in seconds for detecting a hung thread"),
"flush is called, to be sure the user of librbd will send "
"flushes so that writeback is safe"),
- Option("rbd_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(32<<20)
+ Option("rbd_cache_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(32_M)
.set_description("cache size in bytes"),
- Option("rbd_cache_max_dirty", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(24<<20)
+ Option("rbd_cache_max_dirty", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(24_M)
.set_description("dirty limit in bytes - set to 0 for write-through caching"),
- Option("rbd_cache_target_dirty", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(16<<20)
+ Option("rbd_cache_target_dirty", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(16_M)
.set_description("target dirty limit in bytes"),
Option("rbd_cache_max_dirty_age", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1.0)
.set_description("seconds in cache before writeback starts"),
- Option("rbd_cache_max_dirty_object", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_cache_max_dirty_object", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("dirty limit for objects - set to 0 for auto calculate from rbd_cache_size"),
.set_default(false)
.set_description("whether to block writes to the cache before the aio_write call completes"),
- Option("rbd_concurrent_management_ops", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_concurrent_management_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(10)
.set_min(1)
.set_description("how many operations can be in flight for a management operation like deleting or resizing an image"),
.set_default(false)
.set_description("localize parent requests to closest OSD"),
- Option("rbd_sparse_read_threshold_bytes", Option::TYPE_UINT,
+ Option("rbd_sparse_read_threshold_bytes", Option::TYPE_SIZE,
Option::LEVEL_ADVANCED)
.set_default(64_K)
.set_description("threshold for issuing a sparse-read")
.set_long_description("minimum number of sequential bytes to read against "
"an object before issuing a sparse-read request to "
- "the cluster. 0 implies it must be a full object read"
+ "the cluster. 0 implies it must be a full object read "
"to issue a sparse-read, 1 implies always use "
"sparse-read, and any value larger than the maximum "
"object size will disable sparse-read for all "
"requests"),
- Option("rbd_readahead_trigger_requests", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_readahead_trigger_requests", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(10)
.set_description("number of sequential requests necessary to trigger readahead"),
- Option("rbd_readahead_max_bytes", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_readahead_max_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(512_K)
.set_description("set to 0 to disable readahead"),
- Option("rbd_readahead_disable_after_bytes", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_readahead_disable_after_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(50_M)
.set_description("how many bytes are read in total before readahead is disabled"),
.set_default(true)
.set_description("whether to blacklist clients whose lock was broken"),
- Option("rbd_blacklist_expire_seconds", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_blacklist_expire_seconds", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("number of seconds to blacklist - set to 0 for OSD default"),
- Option("rbd_request_timed_out_seconds", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_request_timed_out_seconds", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(30)
.set_description("number of seconds before maintenance request times out"),
Option("rbd_skip_partial_discard", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
- .set_description("when trying to discard a range inside an object, set to true to skip zeroing the range"),
+ .set_default(true)
+ .set_description("skip discard (zero) of unaligned extents within an object"),
+
+ Option("rbd_discard_granularity_bytes", Option::TYPE_UINT,
+ Option::LEVEL_ADVANCED)
+ .set_default(64_K)
+ .set_min_max(4_K, 32_M)
+ .set_validator([](std::string *value, std::string *error_message){
+ uint64_t f = strict_si_cast<uint64_t>(value->c_str(), error_message);
+ if (!error_message->empty()) {
+ return -EINVAL;
+ } else if (!isp2(f)) {
+ *error_message = "value must be a power of two";
+ return -EINVAL;
+ }
+ return 0;
+ })
+ .set_description("minimum aligned size of discard operations"),
Option("rbd_enable_alloc_hint", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
.set_default(true)
.set_description("automatically acquire/release exclusive lock until it is explicitly requested"),
+ Option("rbd_move_to_trash_on_remove", Option::TYPE_BOOL, Option::LEVEL_BASIC)
+ .set_default(false)
+ .set_description("automatically move images to the trash when deleted"),
+
+ Option("rbd_move_to_trash_on_remove_expire_seconds", Option::TYPE_UINT, Option::LEVEL_BASIC)
+ .set_default(0)
+ .set_description("default number of seconds to protect deleted images in the trash"),
+
Option("rbd_mirroring_resync_after_disconnect", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
.set_description("automatically start image resync after mirroring is disconnected due to being laggy"),
- Option("rbd_mirroring_replay_delay", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_mirroring_delete_delay", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("time-delay in seconds for rbd-mirror delete propagation"),
+
+ Option("rbd_mirroring_replay_delay", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("time-delay in seconds for rbd-mirror asynchronous replication"),
- Option("rbd_default_format", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_default_format", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(2)
.set_description("default image format for new images"),
- Option("rbd_default_order", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_default_order", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(22)
.set_description("default order (data block object size) for new images"),
.set_default(0)
.set_description("default stripe count for new images"),
- Option("rbd_default_stripe_unit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("rbd_default_stripe_unit", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("default stripe width for new images"),
.set_default("")
.set_description("default krbd map options"),
+ Option("rbd_default_clone_format", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_enum_allowed({"1", "2", "auto"})
+ .set_default("auto")
+ .set_description("default internal format for handling clones")
+ .set_long_description("This sets the internal format for tracking cloned "
+ "images. The setting of '1' requires attaching to "
+ "protected snapshots that cannot be removed until "
+ "the clone is removed/flattened. The setting of '2' "
+ "will allow clones to be attached to any snapshot "
+ "and permits removing in-use parent snapshots but "
+ "requires Mimic or later clients. The default "
+ "setting of 'auto' will use the v2 format if the "
+ "cluster is configured to require mimic or later "
+ "clients.")
+ .set_flag(Option::FLAG_RUNTIME),
+
Option("rbd_journal_order", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_min_max(12, 26)
.set_default(24)
.set_default(5)
.set_description("commit time interval, seconds"),
- Option("rbd_journal_object_flush_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_journal_object_flush_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("maximum number of pending commits per journal object"),
- Option("rbd_journal_object_flush_bytes", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_journal_object_flush_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("maximum number of pending bytes per journal object"),
.set_default(0)
.set_description("maximum age (in seconds) for pending commits"),
+ Option("rbd_journal_object_max_in_flight_appends", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("maximum number of in-flight appends per journal object"),
+
Option("rbd_journal_pool", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
.set_description("pool for journal objects"),
- Option("rbd_journal_max_payload_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("rbd_journal_max_payload_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(16384)
.set_description("maximum journal payload size before splitting"),
- Option("rbd_journal_max_concurrent_object_sets", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_journal_max_concurrent_object_sets", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("maximum number of object sets a journal client can be behind before it is automatically unregistered"),
+
+ Option("rbd_qos_iops_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired limit of IO operations per second"),
+
+ Option("rbd_qos_bps_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired limit of IO bytes per second"),
+
+ Option("rbd_qos_read_iops_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired limit of read operations per second"),
+
+ Option("rbd_qos_write_iops_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired limit of write operations per second"),
+
+ Option("rbd_qos_read_bps_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired limit of read bytes per second"),
+
+ Option("rbd_qos_write_bps_limit", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired limit of write bytes per second"),
+
+ Option("rbd_qos_iops_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired burst limit of IO operations"),
+
+ Option("rbd_qos_bps_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired burst limit of IO bytes"),
+
+ Option("rbd_qos_read_iops_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired burst limit of read operations"),
+
+ Option("rbd_qos_write_iops_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired burst limit of write operations"),
+
+ Option("rbd_qos_read_bps_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired burst limit of read bytes"),
+
+ Option("rbd_qos_write_bps_burst", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("the desired burst limit of write bytes"),
+
+ Option("rbd_qos_schedule_tick_min", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(50)
+ .set_min(1)
+ .set_description("minimum schedule tick (in milliseconds) for QoS"),
+
+ Option("rbd_discard_on_zeroed_write_same", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .set_description("discard data on zeroed write same instead of writing zero"),
+
+ Option("rbd_mtime_update_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(60)
+ .set_min(0)
+ .set_description("RBD Image modify timestamp refresh interval. Set to 0 to disable modify timestamp update."),
+
+ Option("rbd_atime_update_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(60)
+ .set_min(0)
+ .set_description("RBD Image access timestamp refresh interval. Set to 0 to disable access timestamp update."),
});
}
.set_default(5)
.set_description("maximum age (in seconds) between successive journal polls"),
- Option("rbd_mirror_journal_max_fetch_bytes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("rbd_mirror_journal_max_fetch_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(32768)
.set_description("maximum bytes to read from each journal data object per fetch"),
.set_default(5)
.set_description("maximum number of image syncs in parallel"),
- Option("rbd_mirror_pool_replayers_refresh_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_mirror_pool_replayers_refresh_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(30)
.set_description("interval to refresh peers in rbd-mirror daemon"),
+ Option("rbd_mirror_concurrent_image_deletions", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(1)
+ .set_min(1)
+ .set_description("maximum number of image deletions in parallel"),
+
Option("rbd_mirror_delete_retry_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(30)
- .set_description("interval to check and retry the failed requests in deleter"),
+ .set_description("interval to check and retry the failed deletion requests"),
- Option("rbd_mirror_image_state_check_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_mirror_image_state_check_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(30)
.set_min(1)
.set_description("interval to get images from pool watcher and set sources in replayer"),
- Option("rbd_mirror_leader_heartbeat_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_mirror_leader_heartbeat_interval", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(5)
.set_min(1)
.set_description("interval (in seconds) between mirror leader heartbeats"),
- Option("rbd_mirror_leader_max_missed_heartbeats", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_mirror_leader_max_missed_heartbeats", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(2)
.set_description("number of missed heartbeats for non-lock owner to attempt to acquire lock"),
- Option("rbd_mirror_leader_max_acquire_attempts_before_break", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("rbd_mirror_leader_max_acquire_attempts_before_break", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(3)
.set_description("number of failed attempts to acquire lock after missing heartbeats before breaking lock"),
+
+ Option("rbd_mirror_image_policy_type", Option::TYPE_STR, Option::LEVEL_ADVANCED)
+ .set_default("simple")
+ .set_enum_allowed({"none", "simple"})
+ .set_description("active/active policy type for mapping images to instances"),
+
+ Option("rbd_mirror_image_policy_migration_throttle", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(300)
+ .set_description("number of seconds after which an image can be reshuffled (migrated) again"),
+
+ Option("rbd_mirror_image_policy_update_throttle_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(1)
+ .set_min(1)
+ .set_description("interval (in seconds) to throttle images for mirror daemon peer updates"),
+
+ Option("rbd_mirror_image_policy_rebalance_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("number of seconds policy should be idle before trigerring reshuffle (rebalance) of images"),
+
+ Option("rbd_mirror_perf_stats_prio", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ .set_default((int64_t)PerfCountersBuilder::PRIO_USEFUL)
+ .set_description("Priority level for mirror daemon replication perf counters")
+ .set_long_description("The daemon will send perf counter data to the "
+ "manager daemon if the priority is not lower than "
+ "mgr_stats_threshold.")
+ .set_min_max((int64_t)PerfCountersBuilder::PRIO_DEBUGONLY,
+ (int64_t)PerfCountersBuilder::PRIO_CRITICAL + 1),
});
}
return std::vector<Option>({
Option("mds_data", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("/var/lib/ceph/mds/$cluster-$id")
- .set_description(""),
+ .set_flag(Option::FLAG_NO_MON_UPDATE)
+ .set_description("path to MDS data and keyring"),
- Option("mds_max_xattr_pairs_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(64 << 10)
- .set_description(""),
+ Option("mds_max_xattr_pairs_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(64_K)
+ .set_description("maximum aggregate size of extended attributes on a file"),
Option("mds_cache_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("maximum number of inodes in MDS cache (<=0 is unlimited)")
.set_long_description("This tunable is no longer recommended. Use mds_cache_memory_limit."),
- Option("mds_cache_memory_limit", Option::TYPE_UINT, Option::LEVEL_BASIC)
+ Option("mds_cache_memory_limit", Option::TYPE_SIZE, Option::LEVEL_BASIC)
.set_default(1*(1LL<<30))
.set_description("target maximum memory usage of MDS cache")
.set_long_description("This sets a target maximum memory usage of the MDS cache and is the primary tunable to limit the MDS memory usage. The MDS will try to stay under a reservation of this limit (by default 95%; 1 - mds_cache_reservation) by trimming unused metadata in its cache and recalling cached items in the client caches. It is possible for the MDS to exceed this limit due to slow recall from clients. The mds_health_cache_threshold (150%) sets a cache full threshold for when the MDS signals a cluster health warning."),
Option("mds_cache_reservation", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.05)
- .set_description("amount of memory to reserve"),
+ .set_description("amount of memory to reserve for future cached objects"),
Option("mds_health_cache_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1.5)
Option("mds_cache_mid", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(.7)
- .set_description(""),
+ .set_description("midpoint for MDS cache LRU"),
Option("mds_cache_trim_decay_rate", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1)
.set_description("decay rate for trimming MDS cache throttle"),
- Option("mds_cache_trim_threshold", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mds_cache_trim_threshold", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(64_K)
.set_description("threshold for number of dentries that can be trimmed"),
Option("mds_max_file_recover", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(32)
- .set_description(""),
+ .set_description("maximum number of files to recover file sizes in parallel"),
Option("mds_dir_max_commit_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(10)
- .set_description(""),
+ .set_description("maximum size in megabytes for a RADOS write to a directory"),
Option("mds_dir_keys_per_op", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(16384)
- .set_description(""),
+ .set_description("number of directory entries to read in one RADOS operation"),
Option("mds_decay_halflife", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("rate of decay for temperature counters on each directory for balancing"),
Option("mds_beacon_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(4)
- .set_description(""),
+ .set_description("interval in seconds between MDS beacons to monitors"),
Option("mds_beacon_grace", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(15)
- .set_description(""),
+ .set_description("tolerance in seconds for missed MDS beacons to monitors"),
Option("mds_heartbeat_grace", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(15)
Option("mds_enforce_unique_name", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("require MDS name is unique in the cluster"),
Option("mds_session_blacklist_on_timeout", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("blacklist clients whose sessions have become stale"),
Option("mds_session_blacklist_on_evict", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("blacklist clients that have been evicted"),
Option("mds_sessionmap_keys_per_op", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1024)
- .set_description(""),
+ .set_description("number of omap keys to read from the SessionMap in one operation"),
- Option("mds_recall_max_caps", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mds_recall_max_caps", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(5000)
.set_description("maximum number of caps to recall from client session in single recall"),
.set_default(2.5)
.set_description("decay rate for throttle on recalled caps on a session"),
- Option("mds_recall_max_decay_threshold", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mds_recall_max_decay_threshold", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(16_K)
.set_description("decay threshold for throttle on recalled caps on a session"),
- Option("mds_recall_global_max_decay_threshold", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mds_recall_global_max_decay_threshold", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(64_K)
.set_description("decay threshold for throttle on recalled caps globally"),
- Option("mds_recall_warning_threshold", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mds_recall_warning_threshold", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(32_K)
.set_description("decay threshold for warning on slow session cap recall"),
.set_default(60.0)
.set_description("decay rate for warning on slow session cap recall"),
- Option("mds_freeze_tree_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mds_freeze_tree_timeout", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(30)
.set_description(""),
Option("mds_health_summarize_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(10)
- .set_description(""),
+ .set_description("threshold of number of clients to summarize late client recall"),
Option("mds_reconnect_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(45)
- .set_description(""),
+ .set_description("timeout in seconds to wait for clients to reconnect during MDS reconnect recovery state"),
Option("mds_tick_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("time in seconds between upkeep tasks"),
- Option("mds_dirstat_min_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mds_dirstat_min_interval", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(1)
.set_description(""),
Option("mds_scatter_nudge_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("minimum interval between scatter lock updates"),
Option("mds_client_prealloc_inos", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1000)
- .set_description(""),
+ .set_description("number of unused inodes to pre-allocate to clients for file creation"),
Option("mds_early_reply", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("additional reply to clients that metadata requests are complete but not yet durable"),
Option("mds_default_dir_hash", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(CEPH_STR_HASH_RJENKINS)
- .set_description(""),
+ .set_description("hash function to select directory fragment for dentry name"),
- Option("mds_log_pause", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("mds_log_pause", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
- Option("mds_log_skip_corrupt_events", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("mds_log_skip_corrupt_events", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
Option("mds_log_max_events", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(-1)
- .set_description(""),
+ .set_description("maximum number of events in the MDS journal (-1 is unlimited)"),
Option("mds_log_events_per_segment", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(1024)
- .set_description(""),
+ .set_description("maximum number of events in an MDS journal segment"),
- Option("mds_log_segment_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mds_log_segment_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description(""),
+ .set_description("size in bytes of each MDS log segment"),
Option("mds_log_max_segments", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(128)
- .set_description(""),
+ .set_description("maximum number of segments which may be untrimmed"),
Option("mds_bal_export_pin", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("allow setting directory export pins to particular ranks"),
Option("mds_bal_sample_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(3.0)
- .set_description(""),
+ .set_description("interval in seconds between balancer ticks"),
Option("mds_bal_replicate_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(8000)
- .set_description(""),
+ .set_description("hot popularity threshold to replicate a subtree"),
Option("mds_bal_unreplicate_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description(""),
+ .set_description("cold popularity threshold to merge subtrees"),
Option("mds_bal_split_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(10000)
- .set_description(""),
+ .set_description("minimum size of directory fragment before splitting"),
Option("mds_bal_split_rd", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(25000)
- .set_description(""),
+ .set_description("hot read popularity threshold for splitting a directory fragment"),
Option("mds_bal_split_wr", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(10000)
- .set_description(""),
+ .set_description("hot write popularity threshold for splitting a directory fragment"),
Option("mds_bal_split_bits", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(3)
- .set_description(""),
+ .set_min_max(1, 24)
+ .set_description("power of two child fragments for a fragment on split"),
Option("mds_bal_merge_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(50)
- .set_description(""),
+ .set_description("size of fragments where merging should occur"),
Option("mds_bal_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(10)
- .set_description(""),
+ .set_description("interval between MDS balancer cycles"),
Option("mds_bal_fragment_interval", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("delay in seconds before interrupting client IO to perform splits"),
Option("mds_bal_fragment_size_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(10000*10)
- .set_description(""),
+ .set_description("maximum size of a directory fragment before new creat/links fail"),
Option("mds_bal_fragment_fast_factor", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1.5)
- .set_description(""),
+ .set_description("ratio of mds_bal_split_size at which fast fragment splitting occurs"),
+
+ Option("mds_bal_fragment_dirs", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(true)
+ .set_description("enable directory fragmentation")
+ .set_long_description("Directory fragmentation is a standard feature of CephFS that allows sharding directories across multiple objects for performance and stability. Additionally, this allows fragments to be distributed across multiple active MDSs to increase throughput. Disabling (new) fragmentation should only be done in exceptional circumstances and may lead to performance issues."),
Option("mds_bal_idle_threshold", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description(""),
+ .set_description("idle metadata popularity threshold before rebalancing"),
- Option("mds_bal_max", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mds_bal_max", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(-1)
.set_description(""),
- Option("mds_bal_max_until", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mds_bal_max_until", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(-1)
.set_description(""),
- Option("mds_bal_mode", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mds_bal_mode", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("mds_bal_min_rebalance", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mds_bal_min_rebalance", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(.1)
- .set_description(""),
+ .set_description("amount overloaded over internal target before balancer begins offloading"),
- Option("mds_bal_min_start", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mds_bal_min_start", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(.2)
.set_description(""),
- Option("mds_bal_need_min", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mds_bal_need_min", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(.8)
.set_description(""),
- Option("mds_bal_need_max", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mds_bal_need_max", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(1.2)
.set_description(""),
- Option("mds_bal_midchunk", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mds_bal_midchunk", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(.3)
.set_description(""),
- Option("mds_bal_minchunk", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mds_bal_minchunk", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(.001)
.set_description(""),
Option("mds_bal_target_decay", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(10.0)
- .set_description(""),
+ .set_description("rate of decay for export targets communicated to clients"),
Option("mds_replay_interval", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(1.0)
- .set_description(""),
+ .set_description("time in seconds between replay of updates to journal by standby replay MDS"),
- Option("mds_shutdown_check", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mds_shutdown_check", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("mds_thrash_exports", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mds_thrash_exports", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("mds_thrash_fragments", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mds_thrash_fragments", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("mds_dump_cache_on_map", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("mds_dump_cache_on_map", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
- Option("mds_dump_cache_after_rejoin", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("mds_dump_cache_after_rejoin", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
- Option("mds_verify_scatter", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("mds_verify_scatter", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false)
.set_description(""),
.set_default(0)
.set_description(""),
- Option("mds_max_export_size", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("mds_max_export_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(20_M)
.set_description(""),
.set_default(0)
.set_description(""),
- Option("mds_journal_format", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mds_journal_format", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(1)
.set_description(""),
.set_default(0)
.set_description(""),
- Option("mds_wipe_sessions", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("mds_wipe_sessions", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("mds_wipe_ino_prealloc", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ Option("mds_wipe_ino_prealloc", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("mds_skip_ino", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mds_skip_ino", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(0)
.set_description(""),
- Option("mds_standby_for_name", Option::TYPE_STR, Option::LEVEL_ADVANCED)
- .set_default("")
- .set_description(""),
-
- Option("mds_standby_for_rank", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(-1)
- .set_description(""),
-
- Option("mds_standby_for_fscid", Option::TYPE_INT, Option::LEVEL_ADVANCED)
- .set_default(-1)
- .set_description(""),
-
- Option("mds_standby_replay", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(false)
- .set_description(""),
-
Option("mds_enable_op_tracker", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("track remote operation progression and statistics"),
Option("mds_op_history_size", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(20)
- .set_description(""),
+ .set_description("maximum size for list of historical operations"),
Option("mds_op_history_duration", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(600)
- .set_description(""),
+ .set_description("expiration time in seconds of historical operations"),
Option("mds_op_complaint_time", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(30)
- .set_description(""),
+ .set_description("time in seconds to consider an operation blocked after no updates"),
- Option("mds_op_log_threshold", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("mds_op_log_threshold", Option::TYPE_INT, Option::LEVEL_DEV)
.set_default(5)
.set_description(""),
Option("mds_snap_min_uid", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description(""),
+ .set_description("minimum uid of client to perform snapshots"),
Option("mds_snap_max_uid", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(4294967294)
- .set_description(""),
+ .set_description("maximum uid of client to perform snapshots"),
Option("mds_snap_rstat", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
- .set_description(""),
+ .set_description("enabled nested rstat for snapshots"),
- Option("mds_verify_backtrace", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mds_verify_backtrace", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(1)
.set_description(""),
- Option("mds_max_completed_flushes", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mds_max_completed_flushes", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(100000)
.set_description(""),
- Option("mds_max_completed_requests", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ Option("mds_max_completed_requests", Option::TYPE_UINT, Option::LEVEL_DEV)
.set_default(100000)
.set_description(""),
Option("mds_action_on_write_error", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1)
- .set_description(""),
+ .set_description("action to take when MDS cannot write to RADOS (0:ignore, 1:read-only, 2:suicide)"),
Option("mds_mon_shutdown_timeout", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("time to wait for mon to receive damaged MDS rank notification"),
Option("mds_max_purge_files", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(64)
- .set_description(""),
+ .set_description("maximum number of deleted files to purge in parallel"),
Option("mds_max_purge_ops", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(8192)
- .set_description(""),
+ .set_description("maximum number of purge operations performed in parallel"),
Option("mds_max_purge_ops_per_pg", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(0.5)
- .set_description(""),
+ .set_description("number of parallel purge operations performed per PG"),
- Option("mds_purge_queue_busy_flush_period", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
+ Option("mds_purge_queue_busy_flush_period", Option::TYPE_FLOAT, Option::LEVEL_DEV)
.set_default(1.0)
.set_description(""),
Option("mds_root_ino_uid", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description(""),
+ .set_description("default uid for new root directory"),
Option("mds_root_ino_gid", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(0)
- .set_description(""),
+ .set_description("default gid for new root directory"),
Option("mds_max_scrub_ops_in_progress", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(5)
- .set_description(""),
+ .set_description("maximum number of scrub operations performed in parallel"),
Option("mds_damage_table_max_entries", Option::TYPE_INT, Option::LEVEL_ADVANCED)
.set_default(10000)
- .set_description(""),
+ .set_description("maximum number of damage table entries"),
Option("mds_client_writeable_range_max_inc_objs", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(1024)
- .set_description(""),
-
+ .set_description("maximum number of objects in writeable range of a file for a client"),
+
Option("mds_min_caps_per_client", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
.set_default(100)
.set_description("minimum number of capabilities a client may hold"),
Option("mds_inject_migrator_session_race", Option::TYPE_BOOL, Option::LEVEL_DEV)
.set_default(false),
- Option("mds_inject_migrator_message_loss", Option::TYPE_INT, Option::LEVEL_DEV)
- .set_default(0)
- .set_description(""),
-
- Option("mds_max_retries_on_remount_failure", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
- .set_default(5)
- .set_description("number of consecutive failed remount attempts for invalidating kernel dcache after which client would abort."),
-
Option("mds_request_load_average_decay_rate", Option::TYPE_FLOAT, Option::LEVEL_ADVANCED)
.set_default(60)
.set_description("rate of decay in seconds for calculating request load average"),
.set_default(0)
.set_description("number of seconds after which clients which have not responded to cap revoke messages by the MDS are evicted."),
- Option("mds_dump_cache_threshold_formatter", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("mds_max_retries_on_remount_failure", Option::TYPE_UINT, Option::LEVEL_ADVANCED)
+ .set_default(5)
+ .set_description("number of consecutive failed remount attempts for invalidating kernel dcache after which client would abort."),
+
+ Option("mds_dump_cache_threshold_formatter", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(1_G)
.set_description("threshold for cache usage to disallow \"dump cache\" operation to formatter")
.set_long_description("Disallow MDS from dumping caches to formatter via \"dump cache\" command if cache usage exceeds this threshold."),
- Option("mds_dump_cache_threshold_file", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("mds_dump_cache_threshold_file", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(0)
.set_description("threshold for cache usage to disallow \"dump cache\" operation to file")
.set_long_description("Disallow MDS from dumping caches to file via \"dump cache\" command if cache usage exceeds this threshold."),
std::vector<Option> get_mds_client_options() {
return std::vector<Option>({
- Option("client_cache_size", Option::TYPE_INT, Option::LEVEL_BASIC)
+ Option("client_cache_size", Option::TYPE_SIZE, Option::LEVEL_BASIC)
.set_default(16384)
.set_description("soft maximum number of directory entries in client cache"),
.set_default("")
.set_description("file containing trace of client operations"),
- Option("client_readahead_min", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("client_readahead_min", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(128*1024)
.set_description("minimum bytes to readahead in a file"),
- Option("client_readahead_max_bytes", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("client_readahead_max_bytes", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(0)
.set_description("maximum bytes to readahead in a file (zero is unlimited)"),
.set_default(true)
.set_description("enable object caching"),
- Option("client_oc_size", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("client_oc_size", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(200_M)
.set_description("maximum size of object cache"),
- Option("client_oc_max_dirty", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("client_oc_max_dirty", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(100_M)
.set_description("maximum size of dirty pages in object cache"),
- Option("client_oc_target_dirty", Option::TYPE_INT, Option::LEVEL_ADVANCED)
+ Option("client_oc_target_dirty", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
.set_default(8_M)
.set_description("target size of dirty pages object cache"),
.set_default(0)
.set_description(""),
- Option("client_max_inline_size", Option::TYPE_UINT, Option::LEVEL_DEV)
+ Option("client_max_inline_size", Option::TYPE_SIZE, Option::LEVEL_DEV)
.set_default(4_K)
.set_description(""),
.set_description("set the directory size as the number of file bytes recursively used")
.set_long_description("This option enables a CephFS feature that stores the recursive directory size (the bytes used by files in the directory and its descendents) in the st_size field of the stat structure."),
+ Option("client_force_lazyio", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
+ .set_default(false)
+ .set_description(""),
+
+ // note: the max amount of "in flight" dirty data is roughly (max - target)
Option("fuse_use_invalidate_cb", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
- .set_description(""),
+ .set_description("use fuse 2.8+ invalidate callback to keep page cache consistent"),
Option("fuse_disable_pagecache", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(false)
.set_description("pass default_permisions to FUSE on mount"),
Option("fuse_big_writes", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
- .set_default(true)
- .set_description(""),
+ .set_default(false)
+ .set_description("big_writes is deprecated in libfuse 3.0.0"),
+
+ Option("fuse_max_write", Option::TYPE_SIZE, Option::LEVEL_ADVANCED)
+ .set_default(0)
+ .set_description("set the maximum number of bytes in a single write operation")
+ .set_long_description("Set the maximum number of bytes in a single write operation that may pass atomically through FUSE. The FUSE default is 128kB and may be indicated by setting this option to 0."),
Option("fuse_atomic_o_trunc", Option::TYPE_BOOL, Option::LEVEL_ADVANCED)
.set_default(true)
Option("client_mds_namespace", Option::TYPE_STR, Option::LEVEL_ADVANCED)
.set_default("")
- .set_description("CephFS file system name to mount"),
+
+ .set_description("CephFS file system name to mount")
+ .set_long_description("Use this with ceph-fuse, or with any process "
+ "that uses libcephfs. Programs using libcephfs may also pass "
+ "the filesystem name into mount(), which will override this setting. "
+ "If no filesystem name is given in mount() or this setting, the default "
+ "filesystem will be mounted (usually the first created)."),
+
+ Option("fake_statfs_for_testing", Option::TYPE_INT, Option::LEVEL_DEV)
+ .set_default(0)
+ .set_description("Set a value for kb and compute kb_used from total of num_bytes"),
});
}
std::vector<Option> result = get_global_options();
auto ingest = [&result](std::vector<Option>&& options, const char* svc) {
- for (const auto &o_in : options) {
- Option o(o_in);
+ for (auto &o : options) {
o.add_service(svc);
- result.push_back(o);
+ result.push_back(std::move(o));
}
};
projects / ceph.git / blobdiff