[bitnami/wordpress] Wordpress maintenance mode returns 503 health check fails

[lwj5]

Which chart:
wordpress

Describe the bug
When updating WordPress plugins or themes. WordPress is automatically put into maintenance mode.
wp-login.php will return 503, thus health check will fail.

To Reproduce
Steps to reproduce the behavior:

1. Go to 'WordPress updates'
2. Click on 'update all'
3. See pod die

Expected behavior
The pod is not supposed to be killed, the health check URL should return 200 even in maintenance mode.

Version of Helm and Kubernetes:
Does not matter

[javsalgar]

Hi,

I think that, in the case of a long upgrade, it may make sense to change the liveness/readiness probe of the container. In other charts we are allowing a customLivenessProbe and customReadinessProbe section that deals with this kind of situations. In this chart it is not implemented yet, so while you perform this maintenance operation I would advise you to disable the liveness/readiness probes and then reenable them. Would it make sense for your case?

[lwj5]

I see okay, thanks for the reply. I'll try to find something for a long term solution.

[lwj5]

@javsalgar wp-admin/install.php might be a good health check path. It doesn't get affected by maintenance mode and fails when DB connection fails. You could consider the path change.

[javsalgar]

Hi,

Thank you very much for the input! Feel free to create a PR with the change. If not, we will add it to our backlog (which would take a bit longer because of priorities)

[lwj5]

PR added.

[javsalgar]

Thank you so much! We will review it very soon :)