[stable/rabbitmq-ha] High CPU usage at "idle" from livelinessProbe and readinessProbe

[mjrsnyder]

Is this a request for help?: Yes

---

Is this a BUG REPORT or FEATURE REQUEST? (choose one): BUG REPORT

Version of Helm and Kubernetes:
helm: 2.7.2
kubernetes: 1.9.6

Which chart:
stable/rabbitmq-ha

What happened:
After running helm install the running pod consumes ~100% of a cpu core.

Removing the livelinessProbe and readinessProbe after helm finishes the deployment and restarting the rabbitmq pod reduces the cpu usage dramatically. I think this issue is related to #3855 in [stable/rabbitmq]

The issue seems to stem from a high max number of processes. Adding an option to set ulimit -n {{ .Values.processesUlimit }} in the container command may provide a temporary solution. Here's a link to a report of this issue (it is rather old) https://groups.google.com/forum/#!topic/rabbitmq-users/hO06SB-QBqc

What you expected to happen:
not use 100% cpu

How to reproduce it (as minimally and precisely as possible):

helm install --name hungry-rabbit \ 
stable/rabbitmq-ha \
--set rbac.create=false \
--set replicaCount=1 

Anything else we need to know:
I am running helm from a Windows machine.
RabbitMQ functions fine, it just consumes way more resources than necessary.

[thomas-riccardi]

Related to #3855 ? In the end, even with the ulimit applied, I still had high CPU usage on probes: rabbitmqctl status is slow (see #3855 (comment))

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Any further update will cause the issue/pull request to no longer be considered stale. Thank you for your contributions.

[mjrsnyder]

I'm still having this issue