Details
-
Bug
-
Resolution: Fixed
-
Major
-
1.2.1
-
None
Description
We are using https://github.com/kubernetes/autoscaler/tree/master/cluster-autoscaler
which scales the k8s cluster (running on aws cloud) when it's falling short of available compute resources( cpu and memory). This would have eased the cluster admin's job and would have reduced the maintenance steps/code, when we decide to run more couchbase workloads on k8s.
In this case, since the pod object was not created, the cluster autoscaler didn't get triggered.
I think couchbase operator called it off when the scheduler reported shortage of compute resources.
The YAML file for the 2 cb cluster I was trying to create is attached.
We are using couchbase operator 1.2.1GA
The k8s cluster-autoscaler is working fine otherwise.
Any quick workaround will sure help.