Github user remibergsma commented on the pull request:
https://github.com/apache/cloudstack/pull/885#issuecomment-144446693
The functionality of this PR also seems to work:
```
WARN [c.c.v.VirtualMachinePowerStateSyncImpl]
(AgentManager-Handler-5:null) Detected missing VM but power state is outdated,
wait for another process report run for VM id: 93
WARN [c.c.v.VirtualMachinePowerStateSyncImpl]
(AgentManager-Handler-5:null) Detected missing VM but power state is outdated,
wait for another process report run for VM id: 94
INFO [c.c.v.VirtualMachineManagerImpl] (AgentManager-Handler-2:null) VM
r-93-VM is at Running and we received a power-off report while there is no
pending jobs on it
INFO [c.c.v.VirtualMachineManagerImpl] (AgentManager-Handler-2:null)
Detected out-of-band stop of a HA enabled VM r-93-VM, will schedule restart
INFO [c.c.v.VirtualMachineManagerImpl] (AgentManager-Handler-2:null) VM
i-2-94-VM is at Running and we received a power-off report while there is no
pending jobs on it
INFO [c.c.v.VirtualMachineManagerImpl] (AgentManager-Handler-2:null) VM
i-2-94-VM is sync-ed to at Stopped state according to power-off report from
hypervisor
```
I tested it with a router and an instance. The router was restarted, the
instance was sync'ed to stopped (it was non-HA).
This is a HA-enabled instance:
```
INFO [c.c.v.VirtualMachineManagerImpl] (AgentManager-Handler-6:null) There
is pending job or HA tasks working on the VM. vm id: 93, postpone power-change
report by resetting power-change counters
WARN [c.c.v.VirtualMachinePowerStateSyncImpl]
(AgentManager-Handler-1:null) Detected missing VM but power state is outdated,
wait for another process report run for VM id: 96
INFO [c.c.v.VirtualMachineManagerImpl] (AgentManager-Handler-13:null) VM
i-2-96-VM is at Running and we received a power-off report while there is no
pending jobs on it
INFO [c.c.v.VirtualMachineManagerImpl] (AgentManager-Handler-13:null)
Detected out-of-band stop of a HA enabled VM i-2-96-VM, will schedule restart
INFO [c.c.h.HighAvailabilityManagerImpl] (AgentManager-Handler-13:null)
Schedule vm for HA: VM[User|i-2-96-VM]
INFO [c.c.h.HighAvailabilityManagerImpl] (HA-Worker-2:ctx-3c02c86b work-4)
Processing work HAWork[4-HA-96-Running-Investigating]
INFO [c.c.h.HighAvailabilityManagerImpl] (HA-Worker-2:ctx-3c02c86b work-4)
HA on VM[User|i-2-96-VM]
INFO [c.c.h.HighAvailabilityManagerImpl] (HA-Worker-2:ctx-3c02c86b work-4)
SimpleInvestigator found VM[User|i-2-96-VM] to be alive? false
INFO [c.c.h.HighAvailabilityManagerImpl] (HA-Worker-2:ctx-3c02c86b work-4)
VM is now restarted: 96 on 4
INFO [c.c.h.HighAvailabilityManagerImpl] (HA-Worker-2:ctx-3c02c86b work-4)
Completed work HAWork[4-HA-96-Running-Scheduled]
```
This gets started again as expected, so that also works fine. The only
thing I notice, and that is to be expected, that is now takes longer before HA
kicks in and recovery starts.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
