wal-g backup retention not working since 1.13.0

[bo0ts]

• Which image of the operator are you using? ghcr.io/zalando/postgres-operator:v1.13.0
• Where do you run it - cloud or metal? Kubernetes or OpenShift? OpenShift 4.15
• Are you running Postgres Operator in production? yes
• Type of issue? Bug

Since upgrading to postgres-operator 1.13.0 old wal-g backups are not deleted anymore.

We use the following backup related configuration:

kind: ConfigMap
apiVersion: v1
metadata:
  name: postgres-pod-config
data:
  AWS_ACCESS_KEY_ID: ...
  AWS_SECRET_ACCESS_KEY: ...
  AWS_S3_FORCE_PATH_STYLE: "true"
  AWS_REGION: ...
  AWS_ENDPOINT: ...
  USE_WALG_BACKUP: "true"
  USE_WALG_RESTORE: "true"
  WALG_DISABLE_S3_SSE: "true"
  WAL_S3_BUCKET: ...

We suspect that this is related to the following issue in Spilo zalando/spilo#1015

[sass1997]

@FxKu When do you plan to do a new release with the merged changes?

[FxKu]

Ideally next week. This release will also include Pg 17 and will drop 12 and 13.
In the meantime you can update the Spilo image in the operator configuration and switch to ghcr.io/zalando/spilo-16:3.3-p3

[bo0ts]

Ouch, could we hold of on the drop of 13? I have multiple large deployments that haven't migrated yet, but would really need that bugfix.

…

On Fri, Dec 13, 2024, 12:01 Felix Kunde ***@***.***> wrote: Ideally next week. This release will also include Pg 17 and will drop 12 and 13. In the meantime you can update the Spilo image in the operator configuration and switch to ghcr.io/zalando/spilo-16:3.3-p3 — Reply to this email directly, view it on GitHub <#2786 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAASSRUKK36DQM47K5NW5232FK5AVAVCNFSM6AAAAABQJS6FEGVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNBRGE4TIMRXGE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[FxKu]

Closing this ticket with latest release published. Workaround for 1.13.0 is to use a newer Spilo image like I have mentioned above.