Log sidecars request far too much memory/cpu

[jackmtpt]

nxrm3-ha-repository/nxrm-ha/values.yaml

Lines 143 to 145 in 5d519b3

	requests:
	cpu: "0.1"
	memory: "256Mi"

All they do is tail a log file. They don't need 256Mi to do that.

1m CPU and 32Mi memory is plenty.

[vswaminathan777]

@jackmtpt These are example values provided in the values.yaml file and are intentionally left configurable so that users can adjust them based on the scale of their deployment.

[jackmtpt]

Sure, and I have tweaked them in my deployment, but wouldn't it be better if the values were somewhat sensible to start with? On what basis was 256Mi decided as a reasonable default for tail -f? I seriously doubt that even if you had 1 billion requests/second hitting your nexus instance that you would need that much memory for tail.