On Sat, Jan 07, 2017 at 13:37:48 -0500, John Ferlan wrote:
On 12/16/2016 11:24 AM, Peter Krempa wrote:
> Images that became the backing chain of the current image due to the
> snapshot need to be unlocked in the lock manager. Also if qemu was
> paused during the snapshot the current top level images need to be
> released until qemu is resumed so that they can be acquired properly.
>
> Resolves:
https://bugzilla.redhat.com/show_bug.cgi?id=1191901
> ---
> src/qemu/qemu_driver.c | 17 +++++++++++++++++
> 1 file changed, 17 insertions(+)
>
> diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c
> index 742b6ceda..13da035c2 100644
> --- a/src/qemu/qemu_driver.c
> +++ b/src/qemu/qemu_driver.c
> @@ -14452,6 +14452,23 @@ qemuDomainSnapshotCreateDiskActive(virQEMUDriverPtr
driver,
> virStorageFileUnlink(diskdata[i].src) < 0)
> VIR_WARN("Unable to remove just-created %s",
diskdata[i].src->path);
> }
> + } else {
> + /* on successful snapshot we need to remove locks from the now-old
> + * disks and if the VM is paused release locks on the images since qemu
> + * stopped using them*/
> + bool paused = virDomainObjGetState(vm, NULL) != VIR_DOMAIN_RUNNING;
> +
> + for (i = 0; i < snap->def->ndisks; i++) {
> + if (!diskdata[i].disk)
Should this be disk.src? Or handle the _NONE case? since src only gets
filled in for that.
The entries in the 'diskdata' array are filled only if they are not
_NONE. !diskdata[i].disk->src would crash on the missing entries since
.disk would be NULL. !diskdata[i].disk || !diskdata[i].disk->src is
redundant since the src member is mandatory for a disk object, thus
always present if .disk is not NULL.
Peter