Description of problem: When live migration failed for some reason like timeout, nova reverts attachment information in cinder, but does not properly revert connection info in block device mapping. This ends up with inconsistency between the real connection infomation and the one in database, which cause error in cleanup process when we try to hard-reboot the instance. There is a bug reported to launchpad and this should be also backported to our downstream versions. https://bugs.launchpad.net/nova/+bug/1780973 How reproducible: Always Steps to Reproduce: 1. Create a instance 2. Live-migrate the instance and cause timeout in pre_live_migration 3. Then see block device mappings table in nova database and check connection_info Actual results: block device mapping record is not reverted and has inconsistent information Expected results: block device mapping record is properly reverted
If this bug requires doc text for errata release, please set the 'Doc Type' and provide draft text according to the template in the 'Doc Text' field. The documentation team will review, edit, and approve the text. If this bug does not require doc text, please set the 'requires_doc_text' flag to -.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2019:3796
*** Bug 1986985 has been marked as a duplicate of this bug. ***