Closed jochenehret closed 2 years ago
@ramonskie do you have any feedback on this?
my personal opinion is we should not over engineer this. as it is only going to be used for cf working groups. who should have enough knowledge to debug if something goes wrong. instead of catching every failure/upgrade scenario
this issue will be closed as there has been no activity for a long time. you can reopen this issue if its still necessary to investigate this
We should make the init script idempotent so that we can easily re-run it if a configuration parameter has changed. For example, the database CPU or memory may need to be adapted or the number of nodes in the GKE cluster.
It could also happen that the init script fails in the middle and then must be re-run from the start.
Possible solutions:
|| true
after each command (but this can hide real errors)gcloud create
, first check if the resource already exists or not (more coding effort)