Removing the forced awareness configuration should help.
Here it does however seem to be availability_zone and not aws_availability_zone.
Right, thats a configuration template that I picked up form our documentation. But I verified with actual settings.
I verified that with setting cluster.routing.allocation.awareness.force.aws_availability_zone.values: zone1, zone2, zone3, the cluster is assigning 3 replica-shards to each zone. No unassigned shards issue with shards (1 + 8 replica) but beyond that we seem to have nassigned shards problem.
is there any setting that enforces # number of shards per zone ?
FYI, Illl have to conduct some experiments before removing cluster.routing.allocation.awareness.force.aws_availability_zone.values: zone1, zone2, zone3 till then I am just curious how is allocation working in our cluster with this setting and 8 replicas pershard per index
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.