Description of problem: VM disk contained 2 images. Base and Active/LEAF. Only the base image appears to have migrated to the dst SD and no indication that Active/LEAF move was initiated. Once the COPY task was deemed successful, the source images were purged which included the Active/LEAF that wasn't moved. This rendered VM not to boot and missing data. Error reported in vdsm.log: ~~~ VolumeDoesNotExist: Volume does not exist: (u'2ad59faa-c5c9-489f-a179-5e2c8a382c9a',) ~~~ Version-Release number of selected component (if applicable): rhevm-4.0.7.4-0.1.el7ev.noarch Red Hat Virtualization Host 4.0 (el7.2) vdsm-4.18.13-1.el7ev.x86_64 NFS - File based storage. How reproducible: Not reproducible, but occurred on end-user system. Steps to Reproduce: 1. 2. 3. Actual results: Active/LEAF image was not migrated and source SD images were purged. Expected results: Disk migration of all images should move before purging the source SD images. Additional info:
Seems a duplicate of 1538840
*** This bug has been marked as a duplicate of bug 1538840 ***
I will look at this one again to make sure this is really the same as #1538840
BZ<2>Jira re-sync