Helm chart - readinessProbe failure without errors

[aumthanki]

Hi,

Has anyone been forced to tweak the readinessprobe for the filebeat chart?

readinessProbe:
exec:
command:
- sh
- -c
- |
#!/usr/bin/env bash -e
filebeat test output
failureThreshold: 3
initialDelaySeconds: 10
periodSeconds: 10
timeoutSeconds: 5

Seeing pods fail the readiness probe check but without any errors:
3m12s Warning Unhealthy pod/filebeat-filebeat-2rfcf Readiness probe failed:

The only workaround I've found is to increase the failurethreshold, periodseconds and timeoutseconds.

Regards,
Aum

[mtojek (Marcin Tojek)]

Hi,

could you please format this post using code tags? Thanks in advance!

[aumthanki]

I can't edit the post, apologises. Here is the readiness probe settings formatted correctly:

readinessProbe:
exec:
command:
- sh
- -c
- |
#!/usr/bin/env bash -e
filebeat test output
failureThreshold: 3
initialDelaySeconds: 10
periodSeconds: 10
timeoutSeconds: 5

Our elastic cluster is hosted on Azure. It has two nodes (one master) and a tiebreaker node. The master node seems to be heavily utilised whereas the other node is not. Is this configured on the filebeat level or inside of the node config? How does log balancing work?