Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

Invoking the metrics API (e.g., /cluster/metrics/node/os-cpu-idle?start=1397746020&function=average), I get all null values:

{
  "192.168.4.101": {
    "AVERAGE": [
      [
        1397746020,
        null
      ],
      [
        1397746080,
        null
      ],
. . .
      [
        1397746440,
        null
      ]
    ]
  }
}

The same goes for all os-cpu-* metrics. Other metrics, like memory, disk, network, etc., return valid values.

There are no cpu/proc-related errors in the agent log. Appreciate any insights.

share|improve this question
    
Do the CPU graphs in the UI also show "NO DATA"? Trying to determine if it's an issue with the data or your API call –  mbulman Apr 18 at 13:24
    
The CPU graphs are completely blank, for all CPU metrics (user/nice/system/steal, etc.). They do not show the "NO DATA" label. –  rkiesler Apr 19 at 19:29
    
What OS platform/flavor/version/etc are your nodes on? –  mbulman Apr 21 at 14:56
    
Ubuntu 14.04 Server (VirtualBox) –  rkiesler Apr 23 at 4:58
    
We haven't done any official testing on Ubuntu 14.04, so it's possible there is an incompatibility with how the agent is parsing iostat. Can you paste the output of "iostat -c -m 1 2" somewhere that will retain the formatting and link here? You can also set the log level of the agent to TRACE to get some additional logging around how the agent is running that command and parsing it. –  mbulman Apr 24 at 1:55

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.