Description of problem: Currently hosted engine has its own flows to maintenance the engine VM and another flow to maintenance an HA-enabled host. These flows should be unified with the existing host maintenance flows.
This BZ was created for the engine side, but another may be needed for VDSM to support HA-maintenance request from the engine.
setting target release to current version for consideration and review. please do not push non-RFE bugs to an undefined target release to make sure bugs are reviewed for relevancy, fix, closure, etc.
Merged Change-Ids: Ib62849ffe0f7392c1feb1f2a0536a23410b58210 I0f76f7ad63bcf6d7871c362b46cfa6e928eb9c74 I17f0eff120b2ab99b4badc53da1780c1e758299c
Changes merged upstream master, still pending merge in 3.4 branch (... moving back to post for now)
This BZ should be fixed in oVirt 3.4.0 RC
this is an automated message: moving to Closed CURRENT RELEASE since oVirt 3.4.0 has been released
So, if I understand this bug correctly, putting HE host into maintenance should apply local HE maintenance on the host. However, it is not the case on my 3.6.3 environment. rhevm-3.6.3.4-0.1.el6 ovirt-hosted-engine-ha-1.3.4.3-1.el7ev After putting a host in maintenance, running the following on the host: # hosted-engine --vm-status # does not report any maintenance: ... Local maintenance : False ... # systemctl status ovirt-ha-agent.service # systemctl status ovirt-ha-broker.service Both services are running. Is this a bug or I misunderstand the behavior?
(In reply to Marina from comment #7) > So, if I understand this bug correctly, putting HE host into maintenance > should apply local HE maintenance on the host. > > However, it is not the case on my 3.6.3 environment. > rhevm-3.6.3.4-0.1.el6 > ovirt-hosted-engine-ha-1.3.4.3-1.el7ev > > After putting a host in maintenance, running the following on the host: > # hosted-engine --vm-status # does not report any maintenance: > ... > Local maintenance : False > ... > > # systemctl status ovirt-ha-agent.service > # systemctl status ovirt-ha-broker.service > > Both services are running. > > Is this a bug or I misunderstand the behavior? This was the original behavior, but I'm not sure if something may have changed since then. Martin?
I am pretty certain that it worked for me when I tried. Perhaps it just takes time? Did you try checking status continuously, or at least 1-2 minutes later? Did you have VMs there? It might be that the status changes only some time after all VMs are migrated away, not sure. If this does not work no matter what, I'd suggest to open a new bug and attach engine and HA logs.
Yeah it should still behave like that.
Marina, if you see something different, please open a new BZ and provide all the information. You can add this bz as a general reference.
Ok, last question before I open the bug. Indeed, after flipping back and forward, final status of the host became correct: # hosted-engine --vm-status ... Local maintenance : True ... However, both HA services are still running. Is this correct?
Yes, hosted engine services are still up during local maintenance. The services are responsible for telling other nodes about the status.