use IO::File;
use IPC::Open2;
use PVE::INotify;
+use PVE::Tools;
use PVE::Cluster;
use PVE::Storage;
use PVE::QemuServer;
+use Time::HiRes qw( usleep );
+use PVE::RPCEnvironment;
use base qw(PVE::AbstractMigrate);
sub fork_tunnel {
my ($self, $nodeip, $lport, $rport) = @_;
- my $cmd = [@{$self->{rem_ssh}}, '-L', "$lport:localhost:$rport",
- 'qm', 'mtunnel' ];
+ my @localtunnelinfo = $lport ? ('-L' , "$lport:localhost:$rport" ) : ();
+
+ my $cmd = [@{$self->{rem_ssh}}, @localtunnelinfo, 'qm', 'mtunnel' ];
my $tunnel = $self->fork_command_pipe($cmd);
sub lock_vm {
my ($self, $vmid, $code, @param) = @_;
-
+
return PVE::QemuServer::lock_config($vmid, $code, @param);
}
if (my $pid = PVE::QemuServer::check_running($vmid)) {
die "cant migrate running VM without --online\n" if !$online;
$running = $pid;
+ $self->{forcemachine} = PVE::QemuServer::get_current_qemu_machine($vmid);
}
if (my $loc_res = PVE::QemuServer::check_local_resources($conf, 1)) {
my $volhash = {};
my $cdromhash = {};
+ my $sharedvm = 1;
+
my @sids = PVE::Storage::storage_ids($self->{storecfg});
foreach my $storeid (@sids) {
my $scfg = PVE::Storage::storage_config($self->{storecfg}, $storeid);
PVE::Storage::storage_check_node($self->{storecfg}, $sid, $self->{node});
$volhash->{$volid} = 1;
+ $sharedvm = 0; # there is a non-shared disk
});
}
- # and add used,owned/non-shared disks (just to be sure we have all)
+ # and add used, owned/non-shared disks (just to be sure we have all)
- my $sharedvm = 1;
- PVE::QemuServer::foreach_drive($conf, sub {
- my ($ds, $drive) = @_;
+ PVE::QemuServer::foreach_volid($conf, sub {
+ my ($volid, $is_cdrom) = @_;
- my $volid = $drive->{file};
return if !$volid;
die "cant migrate local file/device '$volid'\n" if $volid =~ m|^/|;
- if (PVE::QemuServer::drive_is_cdrom($drive)) {
+ if ($is_cdrom) {
die "cant migrate local cdrom drive\n" if $volid eq 'cdrom';
return if $volid eq 'none';
$cdromhash->{$volid} = 1;
die "can't migrate '$volid' - storagy type '$scfg->{type}' not supported\n"
if $scfg->{type} ne 'dir';
+
+ # if file, check if a backing file exist
+ if (($scfg->{type} eq 'dir') && (!$sharedvm)) {
+ my (undef, undef, undef, $parent) = PVE::Storage::volume_size_info($self->{storecfg}, $volid, 1);
+ die "can't migrate '$volid' as it's a clone of '$parent'" if $parent;
+ }
}
foreach my $volid (keys %$volhash) {
if (my $err = $@) {
$self->log('err', $err);
}
-
+
if ($self->{volumes}) {
foreach my $volid (@{$self->{volumes}}) {
$self->log('err', "found stale volume copy '$volid' on node '$self->{node}'");
$self->log('info', "starting VM $vmid on remote node '$self->{node}'");
+ my $raddr;
my $rport;
+ my $nodename = PVE::INotify::nodename();
## start on remote node
- my $cmd = [@{$self->{rem_ssh}}, 'qm', 'start',
- $vmid, '--stateuri', 'tcp', '--skiplock'];
+ my $cmd = [@{$self->{rem_ssh}}];
+
+ my $spice_ticket;
+ if (PVE::QemuServer::vga_conf_has_spice($conf->{vga})) {
+ my $res = PVE::QemuServer::vm_mon_cmd($vmid, 'query-spice');
+ $spice_ticket = $res->{ticket};
+ }
- PVE::Tools::run_command($cmd, outfunc => sub {
+ push @$cmd , 'qm', 'start', $vmid, '--stateuri', 'tcp', '--skiplock', '--migratedfrom', $nodename;
+
+ if ($self->{forcemachine}) {
+ push @$cmd, '--machine', $self->{forcemachine};
+ }
+
+ my $spice_port;
+
+ # Note: We try to keep $spice_ticket secret (do not pass via command line parameter)
+ # instead we pipe it through STDIN
+ PVE::Tools::run_command($cmd, input => $spice_ticket, outfunc => sub {
my $line = shift;
- if ($line =~ m/^migration listens on port (\d+)$/) {
- $rport = $1;
+ if ($line =~ m/^migration listens on tcp:([\d\.]+|localhost):(\d+)$/) {
+ $raddr = $1;
+ $rport = int($2);
}
- }, errfunc => sub {});
-
- die "unable to detect remote migration port\n" if !$rport;
+ elsif ($line =~ m/^migration listens on port (\d+)$/) {
+ $raddr = "localhost";
+ $rport = int($1);
+ }
+ elsif ($line =~ m/^spice listens on port (\d+)$/) {
+ $spice_port = int($1);
+ }
+ }, errfunc => sub {
+ my $line = shift;
+ $self->log('info', $line);
+ });
- $self->log('info', "starting migration tunnel");
+ die "unable to detect remote migration address\n" if !$raddr;
## create tunnel to remote port
- my $lport = PVE::QemuServer::next_migrate_port();
+ $self->log('info', "starting ssh migration tunnel");
+ my $lport = ($raddr eq "localhost") ? PVE::Tools::next_migrate_port() : undef;
$self->{tunnel} = $self->fork_tunnel($self->{nodeip}, $lport, $rport);
- $self->log('info', "starting online/live migration on port $lport");
- # start migration
-
my $start = time();
+ $self->log('info', "starting online/live migration on $raddr:$rport");
+ $self->{livemigration} = 1;
+
+ # load_defaults
+ my $defaults = PVE::QemuServer::load_defaults();
+
+ # always set migrate speed (overwrite kvm default of 32m)
+ # we set a very hight default of 8192m which is basically unlimited
+ my $migrate_speed = $defaults->{migrate_speed} || 8192;
+ $migrate_speed = $conf->{migrate_speed} || $migrate_speed;
+ $migrate_speed = $migrate_speed * 1048576;
+ $self->log('info', "migrate_set_speed: $migrate_speed");
eval {
- PVE::QemuServer::vm_mon_cmd_nocheck($vmid, "migrate", uri => "tcp:localhost:$lport");
+ PVE::QemuServer::vm_mon_cmd_nocheck($vmid, "migrate_set_speed", value => int($migrate_speed));
+ };
+ $self->log('info', "migrate_set_speed error: $@") if $@;
+
+ my $migrate_downtime = $defaults->{migrate_downtime};
+ $migrate_downtime = $conf->{migrate_downtime} if defined($conf->{migrate_downtime});
+ if (defined($migrate_downtime)) {
+ $self->log('info', "migrate_set_downtime: $migrate_downtime");
+ eval {
+ PVE::QemuServer::vm_mon_cmd_nocheck($vmid, "migrate_set_downtime", value => int($migrate_downtime*100)/100);
+ };
+ $self->log('info', "migrate_set_downtime error: $@") if $@;
+ }
+
+ my $capabilities = {};
+ $capabilities->{capability} = "xbzrle";
+ $capabilities->{state} = JSON::false;
+
+ eval {
+ PVE::QemuServer::vm_mon_cmd_nocheck($vmid, "migrate-set-capabilities", capabilities => [$capabilities]);
+ };
+
+ #set cachesize 10% of the total memory
+ my $cachesize = int($conf->{memory}*1048576/10);
+ eval {
+ PVE::QemuServer::vm_mon_cmd_nocheck($vmid, "migrate-set-cache-size", value => $cachesize);
+ };
+
+ if (PVE::QemuServer::vga_conf_has_spice($conf->{vga})) {
+ my $rpcenv = PVE::RPCEnvironment::get();
+ my $authuser = $rpcenv->get_user();
+
+ my (undef, $proxyticket) = PVE::AccessControl::assemble_spice_ticket($authuser, $vmid, $self->{node});
+
+ my $filename = "/etc/pve/nodes/$self->{node}/pve-ssl.pem";
+ my $subject = PVE::QemuServer::read_x509_subject_spice($filename);
+
+ $self->log('info', "spice client_migrate_info");
+
+ eval {
+ PVE::QemuServer::vm_mon_cmd_nocheck($vmid, "client_migrate_info", protocol => 'spice',
+ hostname => $proxyticket, 'tls-port' => $spice_port,
+ 'cert-subject' => $subject);
+ };
+ $self->log('info', "client_migrate_info error: $@") if $@;
+
+ }
+
+ eval {
+ PVE::QemuServer::vm_mon_cmd_nocheck($vmid, "migrate", uri => "tcp:$raddr:$rport");
};
my $merr = $@;
+ $self->log('info', "migrate uri => tcp:$raddr:$rport failed: $merr") if $merr;
my $lstat = 0;
+ my $usleep = 2000000;
+ my $i = 0;
+ my $err_count = 0;
+ my $lastrem = undef;
+ my $downtimecounter = 0;
while (1) {
- sleep (2);
- my $stat = PVE::QemuServer::vm_mon_cmd_nocheck($vmid, "query-migrate");
+ $i++;
+ my $avglstat = $lstat/$i if $lstat;
+
+ usleep($usleep);
+ my $stat;
+ eval {
+ $stat = PVE::QemuServer::vm_mon_cmd_nocheck($vmid, "query-migrate");
+ };
+ if (my $err = $@) {
+ $err_count++;
+ warn "query migrate failed: $err\n";
+ if ($err_count <= 5) {
+ usleep(1000000);
+ next;
+ }
+ die "too many query migrate failures - aborting\n";
+ }
if ($stat->{status} =~ m/^(active|completed|failed|cancelled)$/im) {
$merr = undef;
-
+ $err_count = 0;
if ($stat->{status} eq 'completed') {
my $delay = time() - $start;
if ($delay > 0) {
my $mbps = sprintf "%.2f", $conf->{memory}/$delay;
- $self->log('info', "migration speed: $mbps MB/s");
+ my $downtime = $stat->{downtime} || 0;
+ $self->log('info', "migration speed: $mbps MB/s - downtime $downtime ms");
}
}
-
+
if ($stat->{status} eq 'failed' || $stat->{status} eq 'cancelled') {
die "aborting\n"
}
my $trans = $stat->{ram}->{transferred} || 0;
my $rem = $stat->{ram}->{remaining} || 0;
my $total = $stat->{ram}->{total} || 0;
+ my $xbzrlecachesize = $stat->{"xbzrle-cache"}->{"cache-size"} || 0;
+ my $xbzrlebytes = $stat->{"xbzrle-cache"}->{"bytes"} || 0;
+ my $xbzrlepages = $stat->{"xbzrle-cache"}->{"pages"} || 0;
+ my $xbzrlecachemiss = $stat->{"xbzrle-cache"}->{"cache-miss"} || 0;
+ my $xbzrleoverflow = $stat->{"xbzrle-cache"}->{"overflow"} || 0;
+ #reduce sleep if remainig memory if lower than the everage transfert
+ $usleep = 300000 if $avglstat && $rem < $avglstat;
$self->log('info', "migration status: $stat->{status} (transferred ${trans}, " .
"remaining ${rem}), total ${total})");
+
+ #$self->log('info', "migration xbzrle cachesize: ${xbzrlecachesize} transferred ${xbzrlebytes} pages ${xbzrlepages} cachemiss ${xbzrlecachemiss} overflow ${xbzrleoverflow}");
+ if (($lastrem && $rem > $lastrem ) || ($rem == 0)) {
+ $downtimecounter++;
+ }
+ $lastrem = $rem;
+
+ if ($downtimecounter > 5) {
+ $downtimecounter = 0;
+ $migrate_downtime *= 2;
+ $self->log('info', "migrate_set_downtime: $migrate_downtime");
+ eval {
+ PVE::QemuServer::vm_mon_cmd_nocheck($vmid, "migrate_set_downtime", value => int($migrate_downtime*100)/100);
+ };
+ $self->log('info', "migrate_set_downtime error: $@") if $@;
+ }
+
}
- $lstat = $stat->{ram}->{transferred};
+ $lstat = $stat->{ram}->{transferred};
+
} else {
die $merr if $merr;
die "unable to parse migration status '$stat->{status}' - aborting\n";
}
}
+sub phase2_cleanup {
+ my ($self, $vmid, $err) = @_;
+
+ return if !$self->{errors};
+ $self->{phase2errors} = 1;
+
+ $self->log('info', "aborting phase 2 - cleanup resources");
+
+ $self->log('info', "migrate_cancel");
+ eval {
+ PVE::QemuServer::vm_mon_cmd_nocheck($vmid, "migrate_cancel");
+ };
+ $self->log('info', "migrate_cancel error: $@") if $@;
+
+ my $conf = $self->{vmconf};
+ delete $conf->{lock};
+ eval { PVE::QemuServer::update_config_nolock($vmid, $conf, 1) };
+ if (my $err = $@) {
+ $self->log('err', $err);
+ }
+
+ # cleanup ressources on target host
+ my $nodename = PVE::INotify::nodename();
+
+ my $cmd = [@{$self->{rem_ssh}}, 'qm', 'stop', $vmid, '--skiplock', '--migratedfrom', $nodename];
+ eval{ PVE::Tools::run_command($cmd, outfunc => sub {}, errfunc => sub {}) };
+ if (my $err = $@) {
+ $self->log('err', $err);
+ $self->{errors} = 1;
+ }
+}
+
sub phase3 {
my ($self, $vmid) = @_;
-
+
my $volids = $self->{volumes};
+ return if $self->{phase2errors};
# destroy local copies
foreach my $volid (@$volids) {
my ($self, $vmid, $err) = @_;
my $conf = $self->{vmconf};
+ return if $self->{phase2errors};
# move config to remote node
my $conffile = PVE::QemuServer::config_file($vmid);
die "Failed to move config to node '$self->{node}' - rename failed: $!\n"
if !rename($conffile, $newconffile);
+ if ($self->{livemigration}) {
+ # now that config file is move, we can resume vm on target if livemigrate
+ my $cmd = [@{$self->{rem_ssh}}, 'qm', 'resume', $vmid, '--skiplock'];
+ eval{ PVE::Tools::run_command($cmd, outfunc => sub {},
+ errfunc => sub {
+ my $line = shift;
+ $self->log('err', $line);
+ });
+ };
+ if (my $err = $@) {
+ $self->log('err', $err);
+ $self->{errors} = 1;
+ }
+ }
+
+ eval {
+
+ my $timer = 0;
+ if (PVE::QemuServer::vga_conf_has_spice($conf->{vga}) && $self->{running}) {
+ $self->log('info', "Waiting for spice server migration");
+ while (1) {
+ my $res = PVE::QemuServer::vm_mon_cmd_nocheck($vmid, 'query-spice');
+ last if int($res->{'migrated'}) == 1;
+ last if $timer > 50;
+ $timer ++;
+ usleep(200000);
+ }
+ }
+ };
+
# always stop local VM
eval { PVE::QemuServer::vm_stop($self->{storecfg}, $vmid, 1, 1); };
if (my $err = $@) {
projects / qemu-server.git / blobdiff