WSO2 APIM pod takes a fixed time to be ready due to initialDelaySeconds

[matoruru]

Description:
I realized that WSO2 pod is taking some fixed time like 3 mins to be ready. And the reason could be readinessProbe.initialDelaySeconds. Can't we have WSO2 Kubernetes manifest have startupProbe instead?

The reason why I suggest this is that, startupProbe should give this problem dynamic and flexible time to be ready just depends on the machine spec and the time that some WSO2 processes actually take.

Reference: playground-prove-initialdelayseconds-is-bad-startupprobe-is-better

Suggested Labels:
Suggestion, Improvement

Suggested Assignees:
@matoruru

Affected Product Version:
Not sure

OS, DB, other environment details and versions:

Steps to reproduce:
Restart the Deployment resource of APIM.

Related Issues:

[matoruru]

There are livenessProbe and readinessProbe but no startupProbe.

• https://github.com/wso2/kubernetes-apim/blob/master/simple/am-single/templates/am/instance/wso2am-deployment.yaml#L57-L68
• https://github.com/wso2/kubernetes-apim/blob/master/simple/am-single/values.yaml#L67C7-L77

[matoruru]

We, user, could extend this manifest by forking but I think it worth to have in this official-provided manifest.

[matoruru]

If I got assigned to this issue I would be happy to work on this.

[matoruru]

For notification please forgive me to tag you two as most contributors in this repo @chirangaalwis @prasa7