Hide Forgot
Description of problem: Break migration during migrating, unexpected error is caused Version-Release number of selected component (if applicable): - libvirt-0.8.7-7.el6.x86_64 - kernel-2.6.32-113.el6.x86_64 - qemu-kvm-0.12.1.2-2.146.el6.x86_64 How reproducible: always Scenario 1: Very quickly press ctrl+c after executing migration command Test step 1. On source # virsh migrate --live rhel61_i386_1 qemu+ssh://10.66.65.132/system ^Cerror: Requested operation is not valid: no job is active on the domain error: Reconnected to the hypervisor Acutal result 1. On source host The guest is in running status and works well # virsh list --all Id Name State ---------------------------------- 9 6z running 16 rhel61_i386_1 running On target host The guest is in paused status and cannot destroy it # virsh list --all Id Name State ---------------------------------- 88 rhel61_i386_1 paused # virsh destroy rhel61_i386_1 error: Failed to destroy domain rhel61_i386_1 error: Timed out during operation: cannot acquire state change lock Expected result It looks like the virsh code gives up at this point - it should probably continue, and let the user press Ctrl-C again, at which point we'd be in Step 2 and it would work. Scenario 2: After executing migration command for about 2 seconds, surely before the migration is finished, press ctrl+c Test step 1. On source host # virsh migrate --live rhel61_i386_2 qemu+ssh://10.66.65.132/system ^Cerror: operation failed: migration job: canceled by client error: Reconnected to the hypervisor Actual result 1. On source host The guest is in running status but could see I/O error and the guest becomes a read-only system (please refer to rhel61_i386_2.png) # virsh list --all Id Name State ---------------------------------- 9 6z running 16 rhel61_i386_1 running 17 rhel61_i386_2 running On target host No this guest Expected result No error is caused to guest on source host Note: no /var/log/message for both scenario
*** This bug has been marked as a duplicate of bug 682953 ***