X-Git-Url: https://git.proxmox.com/?p=pve-container.git;a=blobdiff_plain;f=src%2FPVE%2FLXC%2FMigrate.pm;h=2ef1cce42adf738050c6cf37043c2bbb54bc01a8;hp=61e0806f5869d1369f50c04e2e43ed9a92d802fe;hb=b9d0dcb46b94dcdcecdad1ca03facdf74dd5faf2;hpb=d250604fbed41dcaad96193559f029dc2aaff974 diff --git a/src/PVE/LXC/Migrate.pm b/src/PVE/LXC/Migrate.pm index 61e0806..2ef1cce 100644 --- a/src/PVE/LXC/Migrate.pm +++ b/src/PVE/LXC/Migrate.pm @@ -9,7 +9,11 @@ use PVE::Tools; use PVE::INotify; use PVE::Cluster; use PVE::Storage; +use PVE::LXC::Config; use PVE::LXC; +use PVE::ReplicationConfig; +use PVE::ReplicationState; +use PVE::Replication; use base qw(PVE::AbstractMigrate); @@ -23,40 +27,67 @@ sub prepare { my ($self, $vmid) = @_; my $online = $self->{opts}->{online}; + my $restart= $self->{opts}->{restart}; $self->{storecfg} = PVE::Storage::config(); - # test is VM exist + # test if CT exists my $conf = $self->{vmconf} = PVE::LXC::Config->load_config($vmid); PVE::LXC::Config->check_lock($conf); my $running = 0; if (PVE::LXC::check_running($vmid)) { - die "lxc live migration is currently not implemented\n"; - - die "cant migrate running container without --online\n" if !$online; + die "lxc live migration is currently not implemented\n" if $online; + die "running container can only be migrated in restart mode" if !$restart; $running = 1; } + $self->{was_running} = $running; - PVE::LXC::Config->foreach_mountpoint($conf, sub { + PVE::LXC::Config->foreach_volume_full($conf, { include_unused => 1 }, sub { my ($ms, $mountpoint) = @_; my $volid = $mountpoint->{volume}; + my $type = $mountpoint->{type}; + + # skip dev/bind mps when shared + if ($type ne 'volume') { + if ($mountpoint->{shared}) { + return; + } else { + die "cannot migrate local $type mount point '$ms'\n"; + } + } + my ($storage, $volname) = PVE::Storage::parse_volume_id($volid, 1) if $volid; - die "can't determine assigned storage for mountpoint '$ms'\n" if !$storage; + die "can't determine assigned storage for mount point '$ms'\n" if !$storage; # check if storage is available on both nodes - my $scfg = PVE::Storage::storage_check_node($self->{storecfg}, $storage); - PVE::Storage::storage_check_node($self->{storecfg}, $storage, $self->{node}); + my $scfg = PVE::Storage::storage_check_enabled($self->{storecfg}, $storage); - die "unable to migrate local mountpoint '$volid' while CT is running" - if !$scfg->{shared} && $running; + my $targetsid = $storage; - }); + die "content type 'rootdir' is not available on storage '$storage'\n" + if !$scfg->{content}->{rootdir}; + + if ($scfg->{shared}) { + # PVE::Storage::activate_storage checks this for non-shared storages + my $plugin = PVE::Storage::Plugin->lookup($scfg->{type}); + warn "Used shared storage '$storage' is not online on source node!\n" + if !$plugin->check_connection($storage, $scfg); + } else { + # unless in restart mode because we shut the container down + die "unable to migrate local mount point '$volid' while CT is running" + if $running && !$restart; + + $targetsid = PVE::JSONSchema::map_id($self->{opts}->{storagemap}, $storage); + } - my $volid_list = PVE::LXC::Config->get_vm_volumes($conf); - PVE::Storage::activate_volumes($self->{storecfg}, $volid_list); + my $target_scfg = PVE::Storage::storage_check_enabled($self->{storecfg}, $targetsid, $self->{node}); + + die "$volid: content type 'rootdir' is not available on storage '$targetsid'\n" + if !$target_scfg->{content}->{rootdir}; + }); # todo: test if VM uses local resources @@ -65,6 +96,17 @@ sub prepare { eval { $self->cmd_quiet($cmd); }; die "Can't connect to destination address using public key\n" if $@; + # in restart mode, we shutdown the container before migrating + if ($restart && $running) { + my $timeout = $self->{opts}->{timeout} // 180; + + $self->log('info', "shutdown CT $vmid\n"); + + PVE::LXC::vm_stop($vmid, 0, $timeout); + + $running = 0; + } + return $running; } @@ -81,27 +123,214 @@ sub phase1 { $self->log('info', "container is running - using online migration"); } - $self->{volumes} = []; + $self->{volumes} = []; # list of already migrated volumes + my $volhash = {}; # 'config', 'snapshot' or 'storage' for local volumes + my $volhash_errors = {}; + my $abort = 0; - PVE::LXC::Config->foreach_mountpoint($conf, sub { - my ($ms, $mountpoint) = @_; + my $log_error = sub { + my ($msg, $volid) = @_; + + $volhash_errors->{$volid} = $msg if !defined($volhash_errors->{$volid}); + $abort = 1; + }; + + my $test_volid = sub { + my ($volid, $snapname) = @_; + + return if !$volid; + + my ($sid, $volname) = PVE::Storage::parse_volume_id($volid); + + # check if storage is available on source node + my $scfg = PVE::Storage::storage_check_enabled($self->{storecfg}, $sid); + + my $targetsid = $sid; + + if ($scfg->{shared}) { + $self->log('info', "volume '$volid' is on shared storage '$sid'") + if !$snapname; + return; + } else { + $targetsid = PVE::JSONSchema::map_id($self->{opts}->{storagemap}, $sid); + } + + PVE::Storage::storage_check_enabled($self->{storecfg}, $targetsid, $self->{node}); + + my $bwlimit = $self->get_bwlimit($sid, $targetsid); + + $volhash->{$volid}->{ref} = defined($snapname) ? 'snapshot' : 'config'; + $volhash->{$volid}->{snapshots} = 1 if defined($snapname); + $volhash->{$volid}->{targetsid} = $targetsid; + $volhash->{$volid}->{bwlimit} = $bwlimit; + + my ($path, $owner) = PVE::Storage::path($self->{storecfg}, $volid); + + die "owned by other guest (owner = $owner)\n" + if !$owner || ($owner != $self->{vmid}); + + if (defined($snapname)) { + # we cannot migrate shapshots on local storage + # exceptions: 'zfspool', 'btrfs' + if ($scfg->{type} eq 'zfspool' || $scfg->{type} eq 'btrfs') { + return; + } + die "non-migratable snapshot exists\n"; + } + }; + + my $test_mp = sub { + my ($ms, $mountpoint, $snapname) = @_; my $volid = $mountpoint->{volume}; - my ($storage, $volname) = PVE::Storage::parse_volume_id($volid); - my $scfg = PVE::Storage::storage_check_node($self->{storecfg}, $storage); + # already checked in prepare + if ($mountpoint->{type} ne 'volume') { + $self->log('info', "ignoring shared '$mountpoint->{type}' mount point '$ms' ('$volid')") + if !$snapname; + return; + } + + eval { + &$test_volid($volid, $snapname); + }; + + &$log_error($@, $volid) if $@; + }; + + # first unused / lost volumes owned by this container + my @sids = PVE::Storage::storage_ids($self->{storecfg}); + foreach my $storeid (@sids) { + my $scfg = PVE::Storage::storage_config($self->{storecfg}, $storeid); + next if $scfg->{shared}; + next if !PVE::Storage::storage_check_enabled($self->{storecfg}, $storeid, undef, 1); + + # get list from PVE::Storage (for unreferenced volumes) + my $dl = PVE::Storage::vdisk_list($self->{storecfg}, $storeid, $vmid, undef, 'rootdir'); - if (!$scfg->{shared}) { + next if @{$dl->{$storeid}} == 0; - $self->log('info', "copy mointpoint '$ms' ($volid) to node ' $self->{node}'"); - PVE::Storage::storage_migrate($self->{storecfg}, $volid, $self->{nodeip}, $storage); - push @{$self->{volumes}}, $volid; + # check if storage is available on target node + my $targetsid = PVE::JSONSchema::map_id($self->{opts}->{storagemap}, $storeid); + my $target_scfg = PVE::Storage::storage_check_enabled($self->{storecfg}, $targetsid, $self->{node}); + + die "content type 'rootdir' is not available on storage '$targetsid'\n" + if !$target_scfg->{content}->{rootdir}; + + PVE::Storage::foreach_volid($dl, sub { + my ($volid, $sid, $volname) = @_; + + $volhash->{$volid}->{ref} = 'storage'; + $volhash->{$volid}->{targetsid} = $targetsid; + }); + } + + # then all volumes referenced in snapshots + foreach my $snapname (keys %{$conf->{snapshots}}) { + &$test_volid($conf->{snapshots}->{$snapname}->{'vmstate'}, 0, undef) + if defined($conf->{snapshots}->{$snapname}->{'vmstate'}); + PVE::LXC::Config->foreach_volume($conf->{snapshots}->{$snapname}, $test_mp, $snapname); + } + + # finally all current volumes + PVE::LXC::Config->foreach_volume_full($conf, { include_unused => 1 }, $test_mp); + + # additional checks for local storage + foreach my $volid (keys %$volhash) { + eval { + my ($sid, $volname) = PVE::Storage::parse_volume_id($volid); + my $scfg = PVE::Storage::storage_config($self->{storecfg}, $sid); + + my $migratable = ($scfg->{type} eq 'dir') || ($scfg->{type} eq 'zfspool') + || ($scfg->{type} eq 'lvmthin') || ($scfg->{type} eq 'lvm') + || ($scfg->{type} eq 'btrfs'); + + die "storage type '$scfg->{type}' not supported\n" + if !$migratable; + + # image is a linked clone on local storage, se we can't migrate. + if (my $basename = (PVE::Storage::parse_volname($self->{storecfg}, $volid))[3]) { + die "clone of '$basename'"; + } + }; + &$log_error($@, $volid) if $@; + } + + foreach my $volid (sort keys %$volhash) { + my $ref = $volhash->{$volid}->{ref}; + if ($ref eq 'storage') { + $self->log('info', "found local volume '$volid' (via storage)\n"); + } elsif ($ref eq 'config') { + $self->log('info', "found local volume '$volid' (in current VM config)\n"); + } elsif ($ref eq 'snapshot') { + $self->log('info', "found local volume '$volid' (referenced by snapshot(s))\n"); } else { - $self->log('info', "mointpoint '$ms' is on shared storage '$storage'"); + $self->log('info', "found local volume '$volid'\n"); } - }); + } + + foreach my $volid (sort keys %$volhash_errors) { + $self->log('warn', "can't migrate local volume '$volid': $volhash_errors->{$volid}"); + } + + if ($abort) { + die "can't migrate CT - check log\n"; + } - my $conffile = PVE::LXC::Config->config_file($vmid); - my $newconffile = PVE::LXC::Config->config_file($vmid, $self->{node}); + my $rep_volumes; + + my $rep_cfg = PVE::ReplicationConfig->new(); + + if (my $jobcfg = $rep_cfg->find_local_replication_job($vmid, $self->{node})) { + die "can't live migrate VM with replicated volumes\n" if $self->{running}; + my $start_time = time(); + my $logfunc = sub { my ($msg) = @_; $self->log('info', $msg); }; + $rep_volumes = PVE::Replication::run_replication( + 'PVE::LXC::Config', $jobcfg, $start_time, $start_time, $logfunc); + } + + my $opts = $self->{opts}; + foreach my $volid (keys %$volhash) { + next if $rep_volumes->{$volid}; + my ($sid, $volname) = PVE::Storage::parse_volume_id($volid); + push @{$self->{volumes}}, $volid; + + # JSONSchema and get_bandwidth_limit use kbps - storage_migrate bps + my $bwlimit = $volhash->{$volid}->{bwlimit}; + $bwlimit = $bwlimit * 1024 if defined($bwlimit); + + my $targetsid = $volhash->{$volid}->{targetsid}; + + my $new_volid = eval { + my $storage_migrate_opts = { + 'ratelimit_bps' => $bwlimit, + 'insecure' => $opts->{migration_type} eq 'insecure', + 'with_snapshots' => $volhash->{$volid}->{snapshots}, + 'allow_rename' => 1, + }; + + my $logfunc = sub { $self->log('info', $_[0]); }; + return PVE::Storage::storage_migrate( + $self->{storecfg}, + $volid, + $self->{ssh_info}, + $targetsid, + $storage_migrate_opts, + $logfunc, + ); + }; + + if (my $err = $@) { + die "storage migration for '$volid' to storage '$targetsid' failed - $err\n"; + } + + $self->{volume_map}->{$volid} = $new_volid; + $self->log('info', "volume '$volid' is '$new_volid' on the target\n"); + + eval { PVE::Storage::deactivate_volumes($self->{storecfg}, [$volid]); }; + if (my $err = $@) { + $self->log('warn', $err); + } + } if ($self->{running}) { die "implement me"; @@ -117,11 +346,13 @@ sub phase1 { my $vollist = PVE::LXC::Config->get_vm_volumes($conf); PVE::Storage::deactivate_volumes($self->{storecfg}, $vollist); - # move config - die "Failed to move config to node '$self->{node}' - rename failed: $!\n" - if !rename($conffile, $newconffile); - + # transfer replication state before moving config + $self->transfer_replication_state() if $rep_volumes; + PVE::LXC::Config->update_volume_ids($conf, $self->{volume_map}); + PVE::LXC::Config->write_config($vmid, $conf); + PVE::LXC::Config->move_config_to_node($vmid, $self->{node}); $self->{conf_migrated} = 1; + $self->switch_replication_job_target() if $rep_volumes; } sub phase1_cleanup { @@ -159,17 +390,30 @@ sub final_cleanup { $self->log('info', "start final cleanup"); if (!$self->{conf_migrated}) { - my $conf = $self->{vmconf}; - delete $conf->{lock}; - - eval { PVE::LXC::Config->write_config($vmid, $conf); }; + eval { PVE::LXC::Config->remove_lock($vmid, 'migrate'); }; if (my $err = $@) { $self->log('err', $err); } + # in restart mode, we start the container on the source node + # on migration error + if ($self->{opts}->{restart} && $self->{was_running}) { + $self->log('info', "start container on source node"); + my $skiplock = 1; + PVE::LXC::vm_start($vmid, $self->{vmconf}, $skiplock); + } } else { my $cmd = [ @{$self->{rem_ssh}}, 'pct', 'unlock', $vmid ]; - $self->cmd_logerr($cmd, errmsg => "failed to clear migrate lock"); + $self->cmd_logerr($cmd, errmsg => "failed to clear migrate lock"); + + # in restart mode, we start the container on the target node + # after migration + if ($self->{opts}->{restart} && $self->{was_running}) { + $self->log('info', "start container on target node"); + my $cmd = [ @{$self->{rem_ssh}}, 'pct', 'start', $vmid]; + $self->cmd($cmd); + } } + } 1;