Change the ordering of the coordinator change and the exclusion

[johscheuer]

Description

Change the ordering of the coordinator change and the exclusion. It's better to first run the exclusion and then change the coordinators, this should ensure that the processes are excluded faster, e.g. in a case where the coordinator change causes a long recovery. We previously implemented with the coordinators change first because we waited for the exclusion to complete, since we use the no_wait option now per default, the exclude command shouldn't timeout, expect in cases where the FDB cluster is in a bad state. So doing the exclude first and then the coordinator change seems like the better choice to ensure that the processes are quickly excluded.

Type of change

• Bug fix (non-breaking change which fixes an issue)

Discussion

I labeled this as a bug but arguably this is not a bug.

Testing

CI will run e2e tests.

Documentation

Follow-up

[foundationdb-ci]

Result of fdb-kubernetes-operator-pr on Linux CentOS 7

• Commit ID: 2c4c85c
• Duration 2:40:41
• Result: ✅ SUCCEEDED
• Error: N/A
• Build Log terminal output (available for 30 days)
• Build Workspace zip file of the working directory (available for 30 days)

[foundationdb-ci]

Result of fdb-kubernetes-operator-pr on Linux CentOS 7

• Commit ID: c572893
• Duration 2:53:33
• Result: ❌ FAILED
• Error: Error while executing command: if $fail_test; then exit 1; fi. Reason: exit status 1
• Build Log terminal output (available for 30 days)
• Build Workspace zip file of the working directory (available for 30 days)