On 05/27/2014 05:24 PM, Eric Blake wrote:
On 05/27/2014 08:06 AM, Jason J. Herne wrote:
From: "Jason J. Herne" <jjherne@us.ibm.com>
qemuDomainObjStart is checking the return code from qemuDomainObjRestore for errors even after determining that the return code is 0. This causes the following error message to appear even when the restore was successful.
+ if (ret > 0) { + VIR_WARN("Ignoring incomplete managed state %s", managed_save); + } else { + VIR_WARN("Unable to restore from managed state %s. " + "Maybe the file is corrupted?", managed_save); + } } + goto cleanup;
This goto is placed wrong; it causes us to skip starting the domain even when loading managed state was successful.
Actually, it looks like we WANT to goto cleanup for ret == 0, and that the real problem is that commit cfc28c66 botched up for being noisy on success. Maybe the fix we want is: diff --git i/src/qemu/qemu_driver.c w/src/qemu/qemu_driver.c index f008763..86190ff 100644 --- i/src/qemu/qemu_driver.c +++ w/src/qemu/qemu_driver.c @@ -6085,8 +6085,9 @@ qemuDomainObjStart(virConnectPtr conn, if (ret > 0) { VIR_WARN("Ignoring incomplete managed state %s", managed_save); } else { - VIR_WARN("Unable to restore from managed state %s. " - "Maybe the file is corrupted?", managed_save); + if (ret < 0) + VIR_WARN("Unable to restore from managed state %s. " + "Maybe the file is corrupted?", managed_save); goto cleanup; } } -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org