On 01/13/2015 12:54 AM, John Ferlan wrote: > > > On 01/07/2015 10:42 AM, Ján Tomko wrote: >> Error out if the domain has disappeared in the meantime. >> >> This prevents writing the persistent definition as the domain >> status and calling qemuDomainRemoveDevice on devices >> that already have been dealt with in qemuProcessStop. >> --- >> src/qemu/qemu_domain.c | 9 +++++---- >> src/qemu/qemu_driver.c | 4 ++-- >> src/qemu/qemu_hotplug.c | 38 +++++++++++++++++++------------------- >> 3 files changed, 26 insertions(+), 25 deletions(-) >> >> diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c >> index a4e4d6b..6b108bf 100644 >> --- a/src/qemu/qemu_hotplug.c >> +++ b/src/qemu/qemu_hotplug.c >> @@ -193,7 +193,8 @@ qemuDomainChangeEjectableMedia(virQEMUDriverPtr driver, >> >> qemuDomainObjEnterMonitor(driver, vm); >> ret = qemuMonitorEjectMedia(priv->mon, driveAlias, force); >> - qemuDomainObjExitMonitor(driver, vm); >> + if (qemuDomainObjExitMonitor(driver, vm) < 0) >> + goto cleanup; > > if "ret == 0" we get to cleanup... with ejected media, no Audit and the > rest of this not done... >> I've added ret = -1 in front of that. Both audit and qemuDomainPrepareDisk on the error path would need access to the disk that's now freed. Jan
Attachment:
signature.asc
Description: OpenPGP digital signature
-- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list