[pve-devel] [PATCH qemu-server v2 1/6] migration: only migrate disks used by the guest

[Aaron Lauterer <a.lauterer@proxmox.com>]

When scanning all configured storages for disk images belonging to the
VM, the migration could easily fail if a storage is not available, but
enabled. That storage might not even be used by the VM at all.

By not doing that and only looking at the disk images referenced in the
VM config, we can avoid that.
Extra handling is needed for disk images currently in the 'pending'
section of the VM config. These disk images used to be detected by
scanning all storages before.
It is also necessary to fetch some information (size, format) about the
disk images explicitly that used to be provided by the initial scan of
all storages.

The big change regarding behavior is that disk images not referenced in
the VM config file will be ignored.  They are already orphans that used
to be migrated as well, but are now left where they are.  The tests have
been adapted to that changed behavior.

Signed-off-by: Aaron Lauterer <a.lauterer@proxmox.com>
---
 PVE/QemuMigrate.pm                    | 71 +++++++++++----------------
 test/MigrationTest/QemuMigrateMock.pm | 10 ++++
 test/run_qemu_migrate_tests.pl        | 12 ++---
 3 files changed, 44 insertions(+), 49 deletions(-)

diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm
index 09cc1d8..1d21250 100644
--- a/PVE/QemuMigrate.pm
+++ b/PVE/QemuMigrate.pm
@@ -312,49 +312,6 @@ sub scan_local_volumes {
 	    $abort = 1;
 	};
 
-	my @sids = PVE::Storage::storage_ids($storecfg);
-	foreach my $storeid (@sids) {
-	    my $scfg = PVE::Storage::storage_config($storecfg, $storeid);
-	    next if $scfg->{shared} && !$self->{opts}->{remote};
-	    next if !PVE::Storage::storage_check_enabled($storecfg, $storeid, undef, 1);
-
-	    # get list from PVE::Storage (for unused volumes)
-	    my $dl = PVE::Storage::vdisk_list($storecfg, $storeid, $vmid, undef, 'images');
-
-	    next if @{$dl->{$storeid}} == 0;
-
-	    my $targetsid = PVE::JSONSchema::map_id($self->{opts}->{storagemap}, $storeid);
-	    if (!$self->{opts}->{remote}) {
-		# check if storage is available on target node
-		my $target_scfg = PVE::Storage::storage_check_enabled(
-		    $storecfg,
-		    $targetsid,
-		    $self->{node},
-		);
-
-		die "content type 'images' is not available on storage '$targetsid'\n"
-		    if !$target_scfg->{content}->{images};
-
-	    }
-
-	    my $bwlimit = $self->get_bwlimit($storeid, $targetsid);
-
-	    PVE::Storage::foreach_volid($dl, sub {
-		my ($volid, $sid, $volinfo) = @_;
-
-		$local_volumes->{$volid}->{ref} = 'storage';
-		$local_volumes->{$volid}->{size} = $volinfo->{size};
-		$local_volumes->{$volid}->{targetsid} = $targetsid;
-		$local_volumes->{$volid}->{bwlimit} = $bwlimit;
-
-		# If with_snapshots is not set for storage migrate, it tries to use
-		# a raw+size stream, but on-the-fly conversion from qcow2 to raw+size
-		# back to qcow2 is currently not possible.
-		$local_volumes->{$volid}->{snapshots} = ($volinfo->{format} =~ /^(?:qcow2|vmdk)$/);
-		$local_volumes->{$volid}->{format} = $volinfo->{format};
-	    });
-	}
-
 	my $replicatable_volumes = !$self->{replication_jobcfg} ? {}
 	    : PVE::QemuConfig->get_replicatable_volumes($storecfg, $vmid, $conf, 0, 1);
 	foreach my $volid (keys %{$replicatable_volumes}) {
@@ -405,8 +362,23 @@ sub scan_local_volumes {
 
 	    $local_volumes->{$volid}->{ref} = $attr->{referenced_in_config} ? 'config' : 'snapshot';
 	    $local_volumes->{$volid}->{ref} = 'storage' if $attr->{is_unused};
+	    $local_volumes->{$volid}->{ref} = 'storage' if $attr->{is_pending};
 	    $local_volumes->{$volid}->{ref} = 'generated' if $attr->{is_tpmstate};
 
+	    my $bwlimit = $self->get_bwlimit($sid, $targetsid);
+	    $local_volumes->{$volid}->{targetsid} = $targetsid;
+	    $local_volumes->{$volid}->{bwlimit} = $bwlimit;
+
+	    my $volume_list = PVE::Storage::volume_list($storecfg, $sid, $vmid, 'images');
+	    # TODO could probably be done better than just iterating
+	    for my $volume (@$volume_list) {
+		if ($volume->{volid} eq $volid) {
+		    $local_volumes->{$volid}->{size} = $volume->{size};
+		    $local_volumes->{$volid}->{format} = $volume->{format};
+		    last;
+		}
+	    }
+
 	    $local_volumes->{$volid}->{is_vmstate} = $attr->{is_vmstate} ? 1 : 0;
 
 	    $local_volumes->{$volid}->{drivename} = $attr->{drivename}
@@ -450,6 +422,19 @@ sub scan_local_volumes {
 		if PVE::Storage::volume_is_base_and_used($storecfg, $volid);
 	};
 
+	# add pending disks first
+	if (defined $conf->{pending} && %{$conf->{pending}}) {
+	    PVE::QemuServer::foreach_volid($conf->{pending}, sub {
+		    my ($volid, $attr) = @_;
+		    $attr->{is_pending} = 1;
+		    eval { $test_volid->($volid, $attr); };
+		    if (my $err = $@) {
+			&$log_error($err, $volid);
+		    }
+		});
+	}
+
+	# add non-pending referenced disks
 	PVE::QemuServer::foreach_volid($conf, sub {
 	    my ($volid, $attr) = @_;
 	    eval { $test_volid->($volid, $attr); };
diff --git a/test/MigrationTest/QemuMigrateMock.pm b/test/MigrationTest/QemuMigrateMock.pm
index 94fe686..46af62a 100644
--- a/test/MigrationTest/QemuMigrateMock.pm
+++ b/test/MigrationTest/QemuMigrateMock.pm
@@ -230,6 +230,16 @@ $MigrationTest::Shared::storage_module->mock(
 	}
 	return $res;
     },
+    volume_list => sub {
+	my ($cfg, $sid, $vmid) = @_;
+
+	my $res = [];
+
+	for my $volume (@{$source_vdisks->{$sid}}) {
+	    push @$res, $volume if $volume->{vmid} eq $vmid;
+	}
+	return $res;
+    },
     vdisk_free => sub {
 	my ($scfg, $volid) = @_;
 
diff --git a/test/run_qemu_migrate_tests.pl b/test/run_qemu_migrate_tests.pl
index 3a3049d..fbe87e6 100755
--- a/test/run_qemu_migrate_tests.pl
+++ b/test/run_qemu_migrate_tests.pl
@@ -707,7 +707,6 @@ my $tests = [
 	},
     },
     {
-	# FIXME: Maybe add orphaned drives as unused?
 	name => '149_running_orphaned_disk_targetstorage_zfs',
 	target => 'pve1',
 	vmid => 149,
@@ -728,10 +727,11 @@ my $tests = [
 	},
 	expected_calls => $default_expected_calls_online,
 	expected => {
-	    source_volids => {},
+	    source_volids => {
+		'local-dir:149/vm-149-disk-0.qcow2' => 1,
+	    },
 	    target_volids => {
 		'local-zfs:vm-149-disk-10' => 1,
-		'local-zfs:vm-149-disk-0' => 1,
 	    },
 	    vm_config => get_patched_config(149, {
 		scsi0 => 'local-zfs:vm-149-disk-10,format=raw,size=4G',
@@ -744,7 +744,6 @@ my $tests = [
 	},
     },
     {
-	# FIXME: Maybe add orphaned drives as unused?
 	name => '149_running_orphaned_disk',
 	target => 'pve1',
 	vmid => 149,
@@ -764,10 +763,11 @@ my $tests = [
 	},
 	expected_calls => $default_expected_calls_online,
 	expected => {
-	    source_volids => {},
+	    source_volids => {
+		'local-dir:149/vm-149-disk-0.qcow2' => 1,
+	    },
 	    target_volids => {
 		'local-lvm:vm-149-disk-10' => 1,
-		'local-dir:149/vm-149-disk-0.qcow2' => 1,
 	    },
 	    vm_config => get_patched_config(149, {
 		scsi0 => 'local-lvm:vm-149-disk-10,format=raw,size=4G',
-- 
2.30.2