Report Kubernetes pods' conditions when they're pending
requested to merge 4569-report-container-configuration-errors-in-the-kubernetes-executor into master
What does this MR do?
Whilst a pod is pending during initialization of a job, we report the conditions that are causing it to still be in a pending state. This allows users to figure out what is causing a slow initialization, or in the event that the pod is never ready, an early indication as to why.
Why was this MR needed?
Some configurations can lead to a pod never becoming ready. This usually occurs when trying out a new configuration.
What's the best way to test this MR?
Set an improbably memory request:
[runners.kubernetes]
memory_request = "100T"
And the additional log lines should appear.
What are the relevant issue numbers?
Closes #4569 (closed)
Edited by Arran Walker