VDSM maintainers don't want to incorporate that fix for 3.5.5 as there is a workaround, after https://bugzilla.redhat.com/show_bug.cgi?id=1260429 is solved the user can deactivate the problematic domain and then proceed with the upgrade. postponing to 3.6
Liron - the attached patch is abandoned. Is this something we'd EVER want to solve? The workaround seems sufficient to me - if there's really a problem accessing the domain, you can't use it, and should probably remove it from the pool.
Allon, imo it is. When we have a problematic domain it may become available later, so we won't always want to force the user to remove it and then add it again in order to perform an upgrade - the upgrade should be allowed regardless, when the domain will become available it'll be upgraded to the needed version (as happens today).
This bug was accidentally moved from POST to MODIFIED via an error in automation, please see mmccune with any questions
Moving from 4.0 alpha to 4.0 beta since 4.0 alpha has been already released and bug is not ON_QA.
oVirt 4.0 beta has been released, moving to RC milestone.
Any news?
Upgrade is always a hot topic. Liron - can you please supply some doctext for this BZ? Thanks!
Should this be ON_QA?
Tested with the following code: ------------------------------------- rhevm-4.0.2-0.1.rc.el7ev.noarch vdsm-4.18.8-1.el7ev.x86_64 Verified with the following scenario: ------------------------------------- Steps to Reproduce: 1. Create new v3.0 DC, Cluster, Add new host 2. Add new storage domain to DC 3. Move host to maintenance 4. block connection between the host and the non master domain 5. activate the host, wait for it to become the spm 6. try to upgrade the DC -> The upgrade is successfull Actual results: The DC upgrade is successfull Moving to VERIFIED!