fix #2774: add early check for non-managed volumes

when checking whether a to-be-added drive's and the VM's replication
status are matching. otherwise, we end up in a failing generic
'parse_volume_id' with no mention of the actual reason.

adding 'replicate=0' to the new drive string fixes the underlying issue
with and without this patch, so this is just a cosmetic/usability
improvement.

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>

diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm
index 5e6fd42513dbaf509e9ae128c85a6eab84276018..974ee3b1e0b41259a5b605c15c060217d317df3e 100644 (file)
--- a/PVE/API2/Qemu.pm
+++ b/PVE/API2/Qemu.pm
@@ -1095,6 +1095,9 @@ my $update_vm_api  = sub {
        return if PVE::QemuServer::drive_is_cdrom($drive);
 
        my ($storeid, $volname) = PVE::Storage::parse_volume_id($volid, 1);
+       die "cannot add non-managed/pass-through volume to a replicated VM\n"
+           if !defined($storeid);
+
        return if defined($volname) && $volname eq 'cloudinit';
 
        my $format;