/run/etc/wal-e.d/env-clone-* uses Instance Profile, regardless if Access Key is configured.

Please, answer some short questions which should help us to understand your problem / question better?

• Which image of the operator are you using? e.g. registry.opensource.zalan.do/acid/postgres-operator:v1.6.3
• Where do you run it - cloud or metal? Kubernetes or OpenShift? AWS K8s | Azure
• Are you running Postgres Operator in production? yes
• Type of issue? Bug report

When generating /run/etc/wal-e.d/env-clone-*, AWS_INSTANCE_PROFILE will always be set to true, and the Access Key will never be included, regardless of what the user has actually configured.

This causes the restore process to break if the instance profile is not used, and an access key is used instead.

I manually updated the env-clone-* directory to remove AWS_INSTANCE_PROFILE and add the AWS access key environment variables, and this resolves this specific issue.

Unfortunately there does not appear to be any way to do this via a pod-env-override.

Start by tracing how /run/etc/wal-e.d/env-clone-* is generated and how the pod-env-override is handled. Compare the generated environment for AWS instance-profile and access-key configurations; done means restore works with configured access keys without forcing AWS_INSTANCE_PROFILE.