On 2017-12-13 05:50, seeteena wrote: > Hi David, > > While creating snapshots when tag name '0' used in the very first and > followed by tag name '1' then snapshot created with tag name '0' is getting > erased. > A snapshot is identified by a name computed either from an id, which is > basically a numerical counter starting at 1 for qcow2, or from a tag, which > is a string (provided > by the user or automatically computed). > > (qemu) savevm 0 > > This creates a snapshot with tag '0' and id '1'.
That's something someone has complained about already, as far as I remember, and this is indeed an issue. >> (qemu) savevm 1 > > This deletes snapshot with name '1' (ie, with id '1') and creates snapshot > with tag '1' and id '1'. I think this is the issue, not info snapshots. > From the output of 'info snapshots' id '1' is not seen for tag '0' instead > seeing an empty field. > > ID TAG VM SIZE DATE VM CLOCK > -- 0 338M 2017-10-16 13:44:35 00:02:07.491 > > If an ID shown as '1' in the above then user can understand the above > documented stuff i.e "If there is already a snapshot with the same tag or ID, > it is replaced". > The '--' in ID field is annoying. Maybe, but this patch is wrong still. Commit 3a1ee711904 says exactly why: > The patch uses snapshot name instead of snapshot id to determine whether a > snapshot is fully available and uses '--' instead of snapshot id in output > because the snapshot id is not guaranteed to be the same on all images. info snapshots first shows a list of snapshots that are present on all disks. However, those are matched by name and not by ID, so the ID is not necessarily the same. Therefore, we can only print it if it is. Sure, we can do that, but your patch is missing that check. Example: $ qemu-img create -f qcow2 foo.qcow2 64M Formatting 'foo.qcow2', fmt=qcow2 size=67108864 encryption=off cluster_size=65536 lazy_refcounts=off refcount_bits=16 $ qemu-img create -f qcow2 bar.qcow2 64M Formatting 'bar.qcow2', fmt=qcow2 size=67108864 encryption=off cluster_size=65536 lazy_refcounts=off refcount_bits=16 $ qemu-img snapshot -c common_snapshot foo.qcow2 $ qemu-img snapshot -c bar_snapshot bar.qcow2 $ qemu-img snapshot -c common_snapshot bar.qcow2 $ qemu-system-x86_64 -hda foo.qcow2 -hdb bar.qcow2 -monitor stdio $ qemu-img snapshot -l foo.qcow2 Snapshot list: ID TAG VM SIZE DATE VM CLOCK 1 common_snapshot 0 2017-12-15 10:09:17 00:00:00.000 $ qemu-img snapshot -l bar.qcow2 Snapshot list: ID TAG VM SIZE DATE VM CLOCK 1 bar_snapshot 0 2017-12-15 10:09:25 00:00:00.000 2 common_snapshot 0 2017-12-15 10:09:28 00:00:00.000 QEMU 2.9.1 monitor - type 'help' for more information (qemu) info snapshots List of snapshots present on all disks: ID TAG VM SIZE DATE VM CLOCK -- common_snapshot 0 2017-12-15 10:09:17 00:00:00.000 List of partial (non-loadable) snapshots on 'ide0-hd1': ID TAG VM SIZE DATE VM CLOCK 1 bar_snapshot 0 2017-12-15 10:09:25 00:00:00.000 With your patch: (qemu) info snapshots List of snapshots present on all disks: ID TAG VM SIZE DATE VM CLOCK 1 common_snapshot 0 2017-12-15 10:09:17 00:00:00.000 List of partial (non-loadable) snapshots on 'ide0-hd1': ID TAG VM SIZE DATE VM CLOCK 1 bar_snapshot 0 2017-12-15 10:09:25 00:00:00.000 Max
signature.asc
Description: OpenPGP digital signature