I recently upgraded from 1.1.1 to 1.2.0 then to 1.3.0
Went to PostgreSQL database that was running and had backup, now pages shows that i don’t have any backups. saw network calls and /backup was throwing 500 error with message that cluster is not available.
Tried to upgrade the operator no change.
Checking the logs in k8 saw this
{"level":"error","T":"2024-12-02T19:02:38Z","caller":"api/proxy.go:156","msg":"invalid character '\\x1f' looking for beginning of value\nfailed to transform response body"}
can someone point to how to recover from here?
Hi @beta-auction this is a bug recently reported and fixed.
The k8s API server compresses the response body using gzip if the payload grows too large. We were not handling this response format correctly.
In your case, you are seeing this while trying to list the backups. Until the fix is released, there’s only one workaround, remove some old backups in order to decrease the payload size and avoid getting it gzipped.
Unfortunately, since the request to list them fails, you won’t be able to do this through the UI. Instead, you’ll need to delete them directly from k8s. You can do it with the following command:
kubectl delete dbb/<BackupName> -n <Namespace>
cool, thanks for the workaround
This fix will be included in v1.4.0, planned to be released in a couple of weeks.