I am curious if anyone has also come across this.
For context I’m using the percona operator to build postgresql clusters in kubernetes.
I specifiy 2 instances in the config - but when it creates the persistent volume claims, it’s creating at least 3 pvcs every time or even more. It doesn’t claim them if it doesn’t use them, it remains in “pending” so not ebs volume is actually being attached, it’s just the object is created in kubernetes. I don’t understand why it soen’t just create the volumes specified and required. Is this a bug in the helm chart?
Hello @Katie_Hemus ,
I can’t reproduce it.
I deployed the database from the default values.yaml changing replicas to 2.
Here is how my PVCs look like:
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
pg-db-instance1-lxqs-pgdata Bound pvc-2ba09f2a-22e7-49b0-b97c-59b87f4236ad 5Gi RWO standard-rwo 51s
pg-db-instance1-thtr-pgdata Bound pvc-b744a67a-64bd-4099-befa-3bf2aed708dc 5Gi RWO standard-rwo 51s
pg-db-repo1 Bound pvc-51c8c4eb-00f9-4f66-942d-36d580ba43c3 1Gi RWO standard-rwo 51s
Could you please share your values.yaml?