The following RFE https://issues.redhat.com/browse/RFE-355 is backed by a high severity/visibility support case from one of our most important customers, <customer name in RFE>. The customer did not request this explicitly. This request is coming from support which finds the addition of an Alert would be helpful in identifying a very hard to detect issue.

—

I was entertained by the final comment in this UPS case about replacing a RAID array cache battery to address a disk i/o latency issue: 02422597. This issue originally manifested as Pod liveness probe failures.
 
From what I can tell, we do not currently ship a Cluster Monitoring Alert rule for high disk i/o latency but it is a possibility:
 
  See 2.8 and 2.9 (Disk read/write latency)
  https://awesome-prometheus-alerts.grep.to/rules.html

• 
  Options