Can't clone from WAL (backed by the Azure storage container)

[l-maciej]

• Which image of the operator are you using? e.g. ghcr.io/zalando/postgres-operator:v1.14.0
• Where do you run it - cloud or metal? Kubernetes or OpenShift? - AKS
• Are you running Postgres Operator in production? - no
• Type of issue? - question

I have issue with setting up the WAL restore process using data stored in AZ blob storage container

Operator was configured using values.yaml

configKubernetes:
  enable_cross_namespace_secret: true
configLogicalBackup:
  logical_backup_docker_image: "ghcr.io/zalando/postgres-operator/logical-backup:v1.14.0" # PG17 support
  logical_backup_provider: "az"
  logical_backup_azure_storage_account_name: "XXXXXXXXXXXXXx"
  logical_backup_azure_storage_container: "XXXXXXXXXXXXXXXXxxxxx"
  logical_backup_azure_storage_account_key: "XXXXXXXXXXXXXXXXXXXXXXXXX"
  logical_backup_job_prefix: "logical-backup-"
  logical_backup_schedule: "37 21 * * *"
  # Fake catalog structure 
  logical_backup_s3_bucket: "XXXXXXXXXXXXX"
  logical_backup_s3_bucket_prefix: "XXXXXXXXXXXX"
  logical_backup_s3_bucket_scope_suffix: "XXXXXXXXXXXXXXXXX"
configAwsOrGcp:
  wal_az_storage_account: "XXXXXXXXXXXXXXXXX"
configGeneral:
  docker_image: ghcr.io/zalando/spilo-17:4.0-p2

First the "base" CRD was deployed

apiVersion: "acid.zalan.do/v1"
kind: postgresql
metadata:
  labels:
    team: test-team
  name: testingdb-1
  namespace: postgress-db-1
spec:
  enableLogicalBackup: true
  teamId: "test-team"
  volume:
    storageClass: "zone-redundant"
    size: 2Gi
  env:
  - name: USE_WALG_BACKUP
    value:  "true"
  - name: USE_WALG_RESTORE
    value:  "true"
  - name: CLONE_USE_WALG_RESTORE
    value:  "true"
  - name: WALG_AZ_PREFIX
    value: "azure://XXXXX/$(SCOPE)/$(PGVERSION)"
  - name: AZURE_STORAGE_ACCESS_KEY
    valueFrom:
      secretKeyRef:
        name: wal-creds
        key: AZ_KEY
  numberOfInstances: 2
  users:
    zalando:  # database owner
    - superuser
    - createdb
    pg-cross.bot: [] #role with NS level secret
    test: []
    foo_user: []  # role for application foo
  databases:
    testdb: zalando  # dbname: owner
  postgresql:
    version: "17"

The database is operational, I'm able to connect create additional DB's and clone it using the direct connection ( just by the name of the DB).
The WAL backup is visible from the inside of the pod (wal-g backup-list).

When trying to clone using UUID with CRD listed below operation fails.

apiVersion: "acid.zalan.do/v1"
kind: postgresql
metadata:
  labels:
    team: test-team
  name: wal-r-b
  namespace: postgress-db-1
spec:
  enableLogicalBackup: true
  teamId: "test-team"
  volume:
    storageClass: "zone-redundant"
    size: 1Gi
  clone: 
    cluster: testingdb-1
    uid: xxxxxxxxxxxxxxxxxxxxxxxxxxx
    timestamp: "2025-01-15T16:00:00.000+00:00"
  env:
  - name: BACKUP_SCHEDULE  # WAL backup sched to to testing
    value: "2 */1 * * *"
  - name: WALG_AZ_PREFIX
    value: "azure://xxxxxxxxxxxxx/$(SCOPE)/$(PGVERSION)"
  - name: CLONE_USE_WALG_RESTORE
    value: "true"
  - name: USE_WALG_BACKUP
    value:  "true"
  - name: USE_WALG_RESTORE
    value:  "true"
  - name: AZURE_STORAGE_ACCESS_KEY
    valueFrom:
      secretKeyRef:
        name: wal-creds
        key: AZ_KEY
  numberOfInstances: 2
  postgresql:
    version: "17"

How I can investigate what causes the errors below

2025-01-21 10:43:02,816 ERROR: Error creating replica using method wal_e: envdir /run/etc/wal-e.d/env bash /scripts/wale_restore.sh exited with code=1
2025-01-21 10:43:32,939 ERROR: Error creating replica using method basebackup_fast_xlog: /scripts/basebackup.sh exited with code=1
2025-01-21 10:43:32,939 ERROR: failed to bootstrap from leader 'wal-r-b-0'
2025-01-21 10:43:41,335 ERROR: Error creating replica using method wal_e: envdir /run/etc/wal-e.d/env bash /scripts/wale_restore.sh exited with code=1
2025-01-21 10:44:11,453 ERROR: Error creating replica using method basebackup_fast_xlog: /scripts/basebackup.sh exited with code=1
2025-01-21 10:44:11,453 ERROR: failed to bootstrap from leader 'wal-r-b-0'