+// SPDX-License-Identifier: AGPL-3.0-or-later
/*
-
- Copyright (C) 2018 Proxmox Server Solutions GmbH
-
- Copyright: qmeventd is under GNU GPL, the GNU General Public License.
-
- This program is free software; you can redistribute it and/or modify
- it under the terms of the GNU General Public License as published by
- the Free Software Foundation; version 2 dated June, 1991.
-
- This program is distributed in the hope that it will be useful,
- but WITHOUT ANY WARRANTY; without even the implied warranty of
- MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
- GNU General Public License for more details.
-
- You should have received a copy of the GNU General Public License
- along with this program; if not, write to the Free Software
- Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
- 02111-1307, USA.
+ Copyright (C) 2018 - 2021 Proxmox Server Solutions GmbH
Author: Dominik Csapak <d.csapak@proxmox.com>
-
- qmeventd listens on a given socket, and waits for qemu processes
- to connect
-
- it then waits for shutdown events followed by the closing of the socket,
- it then calls /usr/sbin/qm cleanup with following arguments
-
- /usr/sbin/qm cleanup VMID <graceful> <guest>
-
- parameter explanation:
-
- graceful:
- 1|0 depending if it saw a shutdown event before the socket closed
-
- guest:
- 1|0 depending if the shutdown was requested from the guest
-
+ Author: Stefan Reiter <s.reiter@proxmox.com>
+
+ Description:
+
+ qmeventd listens on a given socket, and waits for qemu processes to
+ connect. After accepting a connection qmeventd waits for shutdown events
+ followed by the closing of the socket. Once that happens `qm cleanup` will
+ be executed with following three arguments:
+ VMID <graceful> <guest>
+ Where `graceful` can be `1` or `0` depending if shutdown event was observed
+ before the socket got closed. The second parameter `guest` is also boolean
+ `1` or `0` depending if the shutdown was requested from the guest OS
+ (i.e., the "inside").
*/
#ifndef _GNU_SOURCE
#include <errno.h>
#include <fcntl.h>
+#include <gmodule.h>
#include <json.h>
#include <signal.h>
#include <stdbool.h>
#include <stdio.h>
+#include <stdlib.h>
#include <string.h>
#include <sys/epoll.h>
#include <sys/socket.h>
#include <sys/un.h>
#include <sys/wait.h>
#include <unistd.h>
-#include <gmodule.h>
+#include <time.h>
#include "qmeventd.h"
+#define DEFAULT_KILL_TIMEOUT 60
+
static int verbose = 0;
+static int kill_timeout = DEFAULT_KILL_TIMEOUT;
static int epoll_fd = 0;
static const char *progname;
GHashTable *vm_clients; // key=vmid (freed on remove), value=*Client (free manually)
GSList *forced_cleanups;
-volatile sig_atomic_t alarm_triggered = 0;
+static int needs_cleanup = 0;
/*
* Helper functions
fprintf(stderr, "Usage: %s [-f] [-v] PATH\n", progname);
fprintf(stderr, " -f run in foreground (default: false)\n");
fprintf(stderr, " -v verbose (default: false)\n");
+ fprintf(stderr, " -t <s> kill timeout (default: %ds)\n", DEFAULT_KILL_TIMEOUT);
fprintf(stderr, " PATH use PATH for socket\n");
}
}
/*
- * reads the vmid from /proc/<pid>/cmdline
- * after the '-id' argument
+ * parses the vmid from the qemu.slice entry of /proc/<pid>/cgroup
*/
static unsigned long
get_vmid_from_pid(pid_t pid)
{
char filename[32] = { 0 };
- int len = snprintf(filename, sizeof(filename), "/proc/%d/cmdline", pid);
+ int len = snprintf(filename, sizeof(filename), "/proc/%d/cgroup", pid);
if (len < 0) {
fprintf(stderr, "error during snprintf for %d: %s\n", pid,
strerror(errno));
}
unsigned long vmid = 0;
- ssize_t rc = 0;
char *buf = NULL;
size_t buflen = 0;
- while ((rc = getdelim(&buf, &buflen, '\0', fp)) >= 0) {
- if (!strcmp(buf, "-id")) {
- break;
+
+ while (getline(&buf, &buflen, fp) >= 0) {
+ char *cgroup_path = strrchr(buf, ':');
+ if (!cgroup_path) {
+ fprintf(stderr, "unexpected cgroup entry %s\n", buf);
+ goto ret;
}
- }
+ cgroup_path++;
- if (rc < 0) {
- goto err;
- }
+ if (strncmp(cgroup_path, "/qemu.slice/", 12)) {
+ continue;
+ }
+
+ char *vmid_start = strrchr(buf, '/');
+ if (!vmid_start) {
+ fprintf(stderr, "unexpected cgroup entry %s\n", buf);
+ goto ret;
+ }
+ vmid_start++;
- if (getdelim(&buf, &buflen, '\0', fp) >= 0) {
- if (buf[0] == '-' || buf[0] == '\0') {
- fprintf(stderr, "invalid vmid %s\n", buf);
+ if (vmid_start[0] == '-' || vmid_start[0] == '\0') {
+ fprintf(stderr, "invalid vmid in cgroup entry %s\n", buf);
goto ret;
}
errno = 0;
char *endptr = NULL;
- vmid = strtoul(buf, &endptr, 10);
- if (errno != 0) {
+ vmid = strtoul(vmid_start, &endptr, 10);
+ if (!endptr || strncmp(endptr, ".scope", 6)) {
+ fprintf(stderr, "unexpected cgroup entry %s\n", buf);
vmid = 0;
- goto err;
- } else if (*endptr != '\0') {
- fprintf(stderr, "invalid vmid %s\n", buf);
+ continue;
+ }
+ if (errno != 0) {
+ fprintf(stderr, "error parsing vmid for %d: %s\n", pid, strerror(errno));
vmid = 0;
}
goto ret;
}
-err:
- fprintf(stderr, "error parsing vmid for %d: %s\n", pid, strerror(errno));
+ if (errno) {
+ fprintf(stderr, "error parsing vmid for %d: %s\n", pid, strerror(errno));
+ } else {
+ fprintf(stderr, "error parsing vmid for %d: no matching qemu.slice cgroup entry\n", pid);
+ }
ret:
free(buf);
if (!json_object_object_get_ex(obj, "event", &event)) {
return;
}
- VERBOSE_PRINT("%s: got QMP event: %s\n", client->qemu.vmid,
- json_object_get_string(event));
+ VERBOSE_PRINT("%s: got QMP event: %s\n", client->qemu.vmid, json_object_get_string(event));
if (client->state == STATE_TERMINATING) {
// QEMU sometimes sends a second SHUTDOWN after SIGTERM, ignore
- VERBOSE_PRINT("%s: event was after termination, ignoring\n",
- client->qemu.vmid);
+ VERBOSE_PRINT("%s: event was after termination, ignoring\n", client->qemu.vmid);
return;
}
VERBOSE_PRINT("%s: QMP handshake complete\n", client->qemu.vmid);
break;
- case STATE_IDLE:
+ // we expect an empty return object after sending quit
case STATE_TERMINATING:
+ break;
+ case STATE_IDLE:
VERBOSE_PRINT("%s: spurious return value received\n",
client->qemu.vmid);
break;
json_bool has_vmid = data && json_object_object_get_ex(data, "vmid", &vmid_obj);
if (!has_vmid) {
- VERBOSE_PRINT("pid%d: invalid vzdump handshake: no vmid\n",
- client->pid);
+ VERBOSE_PRINT("pid%d: invalid vzdump handshake: no vmid\n", client->pid);
return;
}
const char *vmid_str = json_object_get_string(vmid_obj);
if (!vmid_str) {
- VERBOSE_PRINT("pid%d: invalid vzdump handshake: vmid is not a string\n",
- client->pid);
+ VERBOSE_PRINT("pid%d: invalid vzdump handshake: vmid is not a string\n", client->pid);
return;
}
int res = snprintf(client->vzdump.vmid, sizeof(client->vzdump.vmid), "%s", vmid_str);
if (res < 0 || res >= (int)sizeof(client->vzdump.vmid)) {
- VERBOSE_PRINT("pid%d: invalid vzdump handshake: vmid too long or invalid\n",
- client->pid);
+ VERBOSE_PRINT("pid%d: invalid vzdump handshake: vmid too long or invalid\n", client->pid);
return;
}
- struct Client *vmc =
- (struct Client*) g_hash_table_lookup(vm_clients, client->vzdump.vmid);
+ struct Client *vmc = (struct Client*) g_hash_table_lookup(vm_clients, client->vzdump.vmid);
if (vmc) {
vmc->qemu.backup = true;
// only mark as VZDUMP once we have set everything up, otherwise 'cleanup'
// might try to access an invalid value
client->type = CLIENT_VZDUMP;
- VERBOSE_PRINT("%s: vzdump backup started\n",
- client->vzdump.vmid);
+ VERBOSE_PRINT("%s: vzdump backup started\n", client->vzdump.vmid);
} else {
- VERBOSE_PRINT("%s: vzdump requested backup start for unregistered VM\n",
- client->vzdump.vmid);
+ VERBOSE_PRINT("%s: vzdump requested backup start for unregistered VM\n", client->vzdump.vmid);
}
}
add_new_client(int client_fd)
{
struct Client *client = calloc(sizeof(struct Client), 1);
+ if (client == NULL) {
+ fprintf(stderr, "could not add new client - allocation failed!\n");
+ fflush(stderr);
+ return;
+ }
client->state = STATE_HANDSHAKE;
client->type = CLIENT_NONE;
client->fd = client_fd;
break;
}
+ if (client->pidfd > 0) {
+ (void)close(client->pidfd);
+ }
+ VERBOSE_PRINT("removing %s from forced cleanups\n", client->qemu.vmid);
+ forced_cleanups = g_slist_remove(forced_cleanups, client);
free(client);
}
void
terminate_client(struct Client *client)
{
- VERBOSE_PRINT("%s: terminating client (pid %d)\n",
- client->qemu.vmid, client->pid);
+ VERBOSE_PRINT("%s: terminating client (pid %d)\n", client->qemu.vmid, client->pid);
client->state = STATE_TERMINATING;
}
}
- int err = kill(client->pid, SIGTERM);
- log_neg(err, "kill");
+ // try to send a 'quit' command first, fallback to SIGTERM of the pid
+ static const char qmp_quit_command[] = "{\"execute\":\"quit\"}\n";
+ VERBOSE_PRINT("%s: sending 'quit' via QMP\n", client->qemu.vmid);
+ if (!must_write(client->fd, qmp_quit_command, sizeof(qmp_quit_command) - 1)) {
+ VERBOSE_PRINT("%s: sending 'SIGTERM' to pid %d\n", client->qemu.vmid, client->pid);
+ int err = kill(client->pid, SIGTERM);
+ log_neg(err, "kill");
+ }
+
+ time_t timeout = time(NULL) + kill_timeout;
- struct CleanupData *data_ptr = malloc(sizeof(struct CleanupData));
- struct CleanupData data = {
- .pid = client->pid,
- .pidfd = pidfd
- };
- *data_ptr = data;
- forced_cleanups = g_slist_prepend(forced_cleanups, (void *)data_ptr);
+ client->pidfd = pidfd;
+ client->timeout = timeout;
- // resets any other alarms, but will fire eventually and cleanup all
- alarm(5);
+ forced_cleanups = g_slist_prepend(forced_cleanups, (void *)client);
+ needs_cleanup = 1;
}
void
break;
case json_tokener_continue:
if (client->buflen >= sizeof(client->buf)) {
- VERBOSE_PRINT("pid%d: msg too large, discarding buffer\n",
- client->pid);
+ VERBOSE_PRINT("pid%d: msg too large, discarding buffer\n", client->pid);
memset(client->buf, 0, sizeof(client->buf));
client->buflen = 0;
} // else we have enough space try again after next read
break;
default:
- VERBOSE_PRINT("pid%d: parse error: %d, discarding buffer\n",
- client->pid, jerr);
+ VERBOSE_PRINT("pid%d: parse error: %d, discarding buffer\n", client->pid, jerr);
memset(client->buf, 0, client->buflen);
client->buflen = 0;
break;
json_tokener_free(tok);
}
-
-/*
- * SIGALRM and cleanup handling
- *
- * terminate_client will set an alarm for 5 seconds and add its client's PID to
- * the forced_cleanups list - when the timer expires, we iterate the list and
- * attempt to issue SIGKILL to all processes which haven't yet stopped.
- */
-
static void
-alarm_handler(__attribute__((unused)) int signum)
+sigkill(void *ptr, void *time_ptr)
{
- alarm_triggered = 1;
-}
-
-static void
-sigkill(void *ptr, __attribute__((unused)) void *unused)
-{
- struct CleanupData data = *((struct CleanupData *)ptr);
+ struct Client *data = ptr;
int err;
- if (data.pidfd > 0) {
- err = pidfd_send_signal(data.pidfd, SIGKILL, NULL, 0);
- (void)close(data.pidfd);
+ if (data->timeout != 0 && data->timeout > *(time_t *)time_ptr) {
+ return;
+ }
+
+ if (data->pidfd > 0) {
+ err = pidfd_send_signal(data->pidfd, SIGKILL, NULL, 0);
+ (void)close(data->pidfd);
+ data->pidfd = -1;
} else {
- err = kill(data.pid, SIGKILL);
+ err = kill(data->pid, SIGKILL);
}
if (err < 0) {
if (errno != ESRCH) {
fprintf(stderr, "SIGKILL cleanup of pid '%d' failed - %s\n",
- data.pid, strerror(errno));
+ data->pid, strerror(errno));
}
} else {
fprintf(stderr, "cleanup failed, terminating pid '%d' with SIGKILL\n",
- data.pid);
+ data->pid);
}
+
+ data->timeout = 0;
+
+ // remove ourselves from the list
+ forced_cleanups = g_slist_remove(forced_cleanups, ptr);
}
static void
handle_forced_cleanup()
{
- if (alarm_triggered) {
+ if (g_slist_length(forced_cleanups) > 0) {
VERBOSE_PRINT("clearing forced cleanup backlog\n");
- alarm_triggered = 0;
- g_slist_foreach(forced_cleanups, sigkill, NULL);
- g_slist_free_full(forced_cleanups, free);
- forced_cleanups = NULL;
+ time_t cur_time = time(NULL);
+ g_slist_foreach(forced_cleanups, sigkill, &cur_time);
}
+ needs_cleanup = g_slist_length(forced_cleanups) > 0;
}
-
int
main(int argc, char *argv[])
{
char *socket_path = NULL;
progname = argv[0];
- while ((opt = getopt(argc, argv, "hfv")) != -1) {
+ while ((opt = getopt(argc, argv, "hfvt:")) != -1) {
switch (opt) {
case 'f':
daemonize = 0;
case 'v':
verbose = 1;
break;
+ case 't':
+ errno = 0;
+ char *endptr = NULL;
+ kill_timeout = strtoul(optarg, &endptr, 10);
+ if (errno != 0 || *endptr != '\0' || kill_timeout == 0) {
+ usage();
+ exit(EXIT_FAILURE);
+ }
+ break;
case 'h':
usage();
exit(EXIT_SUCCESS);
}
signal(SIGCHLD, SIG_IGN);
- signal(SIGALRM, alarm_handler);
socket_path = argv[optind];
int nevents;
for(;;) {
- nevents = epoll_wait(epoll_fd, events, 1, -1);
+ nevents = epoll_wait(epoll_fd, events, 1, needs_cleanup ? 10*1000 : -1);
if (nevents < 0 && errno == EINTR) {
- handle_forced_cleanup();
continue;
}
bail_neg(nevents, "epoll_wait");
for (int n = 0; n < nevents; n++) {
if (events[n].data.fd == sock) {
- int conn_sock = accept4(sock, NULL, NULL,
- SOCK_NONBLOCK | SOCK_CLOEXEC);
+ int conn_sock = accept4(sock, NULL, NULL, SOCK_NONBLOCK | SOCK_CLOEXEC);
log_neg(conn_sock, "accept");
if (conn_sock > -1) {
add_new_client(conn_sock);
handle_client((struct Client *)events[n].data.ptr);
}
}
-
handle_forced_cleanup();
}
}
projects / qemu-server.git / blobdiff