AnsweredAssumed Answered

Meaning of CLDB metrics collected by Ganglia

Question asked by acetao on Dec 22, 2015
I've set up Ganglia to show CLDB metrics and File Server metrics, but wondering their meaning and usage for cluster monitoring.

First, are these below metrics accumulated values after a cluster is alive, or something for the configured time period by, e.g. cldb.period=10

Second, what's the exact meaning of these metrics, how it can help administrators to monitor the health of a cluster?
Can anybody shed some light on them? Appreciated.

Container Location (/Size) Table Lookup 
Container Location  (/Size) Table Scan 
Container Location  (/Size) Table Update 
MFS Heartbeat processing time (It gives a value of 8G or so, what does it mean?) 
NFS heartbeat processing time 
Number of Active Container Reports 
Number of Client Container Failure Reports 
Number of FS RW Container Reports 
äó_


Outcomes