2014-07-08 12:38 PM
In recent weeks we've seen an influx of protocol latency events. Typically we sit around sub 5ms protocol latency (FC only), but recently I've seen sustained spikes of 50ms or higher. At the same time our CPU will report 100% utilization. I'm having a hard time tracking down where to go, in regards to finding the actual culprit. I've looked at volume ops in comparison to the protocol latency but cannot find a particular volume that sticks out as the culprit.
Are there any helpful tools that can help me narrow this down? Right now I'm using a combination of System Manager, Management Console, and Unified Manager.