aws-efs operator pod in CrashLoopBackOff state about aws-efs-operator

[2uasimojo]

It looks like CSIDriver v1beta1 was removed in 4.9. Several older resource versions were, but either this one wasn't called out in the master list or we missed it when we were doing these fixes. (@camilamacedo86 FYI)
This will need someone to migrate the version of CSIDriver we use in aws-efs-operator. It doesn't have to be us -- community contributions are welcome.

[camilamacedo86]

Hi @2uasimojo and @pushkarmahakal1982,

It seems that the operator aws-efs operator has as a dependency the operator efs-operator. Then, seems that it is trying to install or was installed on the cluster previously a version of the efs-operator which uses the removed APIs on 4.9.

Then note that before you migrate your cluster to 4.9 you must ensure that all operators installed are compatible with 4.9 ( you will be notified in the biggest part of the cases from 4.8 to 4.9 in the case of not, however, it's recommended also check the events before upgrade the cluster to avoid this scenario)

To solve this problem is required to contact the authors and ask for them to publish a new version of their solutions that do not use the APIs which were removed from 4.9 and also ensure that their project depends on available distributions which are also workable on 4.9. More info: redhat-openshift-ecosystem/community-operators-prod#138

[jharrington22]

Hi all, this will be resolved once 0.0.8 is released redhat-openshift-ecosystem/community-operators-prod#551

Version 0.0.7 (referenced) will lock the current broken image with OCP 4.8 where it wont crashloop. 0.0.8 will remove the lock and fix the crashloop for >= OCP 4.9

[jharrington22]

This specific issue was fixed in #44 but will require the above OperatorHub release to roll out to clusters.

[jharrington22]

This is now resolved on master and with the community operator version 0.0.8

Closing this issue.