[ClusterLabs] HA problem: No live migration when setting node on standby

[Philip Schiller]

On clusterlabs.org i found this little detail:

https://clusterlabs.org/pacemaker/doc/2.1/Pacemaker_Explained/epub/advanced-options.html

If an otherwise migratable resource depends on another resource via an 
ordering constraint, there are special situations in which it will be 
restarted rather than migrated.

For example, if the resource depends on a clone, and at the time the 
resource needs to be moved, the clone has instances that are stopping 
and instances that are starting, then the resource will be restarted. 
The scheduler is not yet able to model this situation correctly and so 
takes the safer (if less optimal) path.

Also, if a migratable resource depends on a non-migratable resource, and 
both need to be moved, the migratable resource will be restarted.


My drbd-resource is a clone. I also have the order constraint promote 
drbd then start vm. When i dismiss the order constraint in the crm 
configure it seems to migrate correctly.

Still this isn't optimal at all because with this missing order 
constraint the cluster won't cleanly boot from zero. The cluster doesn't 
wait for drbd promotion before the vm starts.

Does anyone have some information on this? Does it just not work or do i 
miss something.


Kind regards Philip.