Created attachment 335238 [details] cpu info in the HP boxes Description of problem: tried to do the migration between two AMD boxes,but got failed to start listening VM error Version-Release number of selected component (if applicable): ovirt-node-image-1.0-1.snap4.el5ovirt libvirt-0.5.1-3.el5ovirt HP xw4500 workstation How reproducible: Always Steps to Reproduce: [prerequisite] shut off the sasl authentication on the destination box for there is a bug 489250 [on the destination box] 1.# service libvirtd restart [on the source box] 2.# virsh migrate $running_guest qemu+tcp://IP/system libvir: QEMU error :operation failed :failed to start listening VM [on the destination box] 1.# service libvirtd stop 2.# LIBVIRT_DEBUG=1 libvirtd --verbose -l [on the source box] 3.# virsh migrate $running_guest qemu+tcp://IP/system libvir: QEMU error :operation failed :failed to start listening VM Actual results: libvir: QEMU error :operation failed :failed to start listening VM Expected results: Additional info:
Created attachment 335240 [details] guest defination xml
migration between two AMD boxes is ok now,but still need to close the sasl authentication . bug 489250
this bug 490304 is fixed in snap5
migration now is not a stable function : for currently there are still some kinds of system could not be migrated, they are Fedora10,rhel5u3,win2k3 .
(In reply to comment #3) > this bug 490304 is fixed in snap5 This comment could be ignored for it's really a mistake. Currently I migrated win2k8 guest stably between the two boxed. But still another issue,for the live migration is not a really live one.
Moving to RHEL since libvirt is not supported in RHEV1.0 production builds and this problem will need to be addressed in RHEL5.4 release
It's unclear this problem is still present in 5.4 as migration over TCP should have been tested heavilly at this point, and some bugs in that area have been fixed. This doesn't seems AMD specific anymore and the remaining SASL issue has been processed independently https://bugzilla.redhat.com/show_bug.cgi?id=489250 So I think we can close this bug at this point, the original issues have both been fixed, and if there were still something remaining it would be better to open a new bug with an adequate title for 5.5, Daniel