[libvirt] [PATCH 4/7] qemu: Kill domain when migration finish fails
Peter Krempa
pkrempa at redhat.com
Thu Jul 9 13:19:09 UTC 2015
On Wed, Jul 08, 2015 at 19:36:03 +0200, Jiri Denemark wrote:
> Whenever something fails during incoming migration in Finish phase
> before we started guest CPUs, we need to kill the domain in addition to
> reporting the failure.
>
> Signed-off-by: Jiri Denemark <jdenemar at redhat.com>
> ---
> src/qemu/qemu_migration.c | 32 ++++++++++++--------------------
> 1 file changed, 12 insertions(+), 20 deletions(-)
>
> diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c
> index 9439954..576b32d 100644
> --- a/src/qemu/qemu_migration.c
> +++ b/src/qemu/qemu_migration.c
> @@ -5589,6 +5589,7 @@ qemuMigrationFinish(virQEMUDriverPtr driver,
> qemuDomainObjPrivatePtr priv = vm->privateData;
> virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(driver);
> unsigned short port;
> + bool keep = false;
>
> VIR_DEBUG("driver=%p, dconn=%p, vm=%p, cookiein=%s, cookieinlen=%d, "
> "cookieout=%p, cookieoutlen=%p, flags=%lx, retcode=%d",
> @@ -5647,15 +5648,9 @@ qemuMigrationFinish(virQEMUDriverPtr driver,
> }
> }
>
> - if (qemuMigrationVPAssociatePortProfiles(vm->def) < 0) {
> - qemuProcessStop(driver, vm, VIR_DOMAIN_SHUTOFF_FAILED,
> - VIR_QEMU_PROCESS_STOP_MIGRATED);
> - virDomainAuditStop(vm, "failed");
> - event = virDomainEventLifecycleNewFromObj(vm,
> - VIR_DOMAIN_EVENT_STOPPED,
> - VIR_DOMAIN_EVENT_STOPPED_FAILED);
> + if (qemuMigrationVPAssociatePortProfiles(vm->def) < 0)
> goto endjob;
> - }
> +
> if (mig->network && qemuDomainMigrateOPDRelocate(driver, vm, mig) < 0)
> VIR_WARN("unable to provide network data for relocation");
>
> @@ -5681,11 +5676,10 @@ qemuMigrationFinish(virQEMUDriverPtr driver,
> * to restart during confirm() step, so we kill it off now.
> */
> if (v3proto) {
> - qemuProcessStop(driver, vm, VIR_DOMAIN_SHUTOFF_FAILED,
> - VIR_QEMU_PROCESS_STOP_MIGRATED);
> - virDomainAuditStop(vm, "failed");
> if (newVM)
> vm->persistent = 0;
If you choose to fix qemuMigrationPersist so that it sets the persistent
flag only on succes to fix the previous patch then this will create a
conflict since the above statement shouldn't be necessary.
> + } else {
> + keep = true;
> }
> goto endjob;
> }
ACK,
Peter
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://listman.redhat.com/archives/libvir-list/attachments/20150709/dd6c2ecb/attachment-0001.sig>
More information about the libvir-list
mailing list