Issue with Performance Analyzer in Centos

Hi Team,

I have installed performance-analyzer plugin in my elastic cluster but i am unable to get any metrics on running following command:
curl -XGET “localhost:9600/_opendistro/_performanceanalyzer/metrics?metrics=Latency,CPU_Utilization&agg=avg,max&dim=ShardID&nodes=all”
Response:
{“local”: {“timestamp”: 1584334145000, “data”: {“fields”:[{“name”:“ShardID”,“type”:“VARCHAR”},{“name”:“Latency”,“type”:“DOUBLE”},{“name”:“CPU_Utilization”,“type”:“DOUBLE”}],“records”:}}}

Also how can i get the default dashboards as i see the installable for PerfTop CLI only for MACOS and linux but my elastic cluster is in Centos.

Please assist.

Regards,
George

1 Like

Hi George,

Is performance analyzer enabled in your cluster? Try running the following command to enable it:

curl localhost:9200/_opendistro/_performanceanalyzer/cluster/config -H 'Content-Type: application/json' -d '{"enabled": true}'

The perftop installation comes with several default dashboards. The preferred method of installing perftop is via npm as outlined in the instructions here. If you can’t use npm, then the Linux zip distribution here should have.

If you’re still having issues, please share the output of the curl command I shared above, and please share the method you used to install performance-analyzer and which version you’re using.

Hope this helps!
Sid

Ran into the same issue and after running the following command in Dev Tools, we were able to enable performance analyzer.

POST /_opendistro/_performanceanalyzer/cluster/config
{“enabled”: true}

We are now able to view results both through the API and the dashboards.