[ 
https://issues.apache.org/jira/browse/FLINK-1792?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14532173#comment-14532173
 ] 

ASF GitHub Bot commented on FLINK-1792:
---------------------------------------

Github user rmetzger commented on the pull request:

    https://github.com/apache/flink/pull/553#issuecomment-99753925
  
    The code looks much better now. The only thing that makes me still unhappy 
is the following method
    ```diff
    +       override def getValue: Double = {
    +        try{
    +          val osMXBean = ManagementFactory.getOperatingSystemMXBean().
    +            asInstanceOf[com.sun.management.OperatingSystemMXBean]
    +          
fetchCPULoad.map(_.invoke(osMXBean).asInstanceOf[Double]).getOrElse(-1)
    +        } catch {
    +          case t: Throwable => {
    +            LOG.warn("Error retrieving CPU Load through 
OperatingSystemMXBean", t)
    +            -1
    +          }
    +        }
    +      }
    ```
    This `getValue()` method is called every 5 seconds to get the metrics from 
each machine. Users using Java 6 will get the WARN in their log + the exception 
because the cast to `com.sun.management.OperatingSystemMXBean` will fail every 
time.
    
    I would do the following: If you detect that fetchCpuLoad is null, register 
a gauge which is always returning -1. Otherwise, register the gauge which is 
calling fetchCpuLoad.


> Improve TM Monitoring: CPU utilization, hide graphs by default and show 
> summary only
> ------------------------------------------------------------------------------------
>
>                 Key: FLINK-1792
>                 URL: https://issues.apache.org/jira/browse/FLINK-1792
>             Project: Flink
>          Issue Type: Sub-task
>          Components: Webfrontend
>    Affects Versions: 0.9
>            Reporter: Robert Metzger
>            Assignee: Sachin Bhat
>
> As per https://github.com/apache/flink/pull/421 from FLINK-1501, there are 
> some enhancements to the current monitoring required
> - Get the CPU utilization in % from each TaskManager process
> - Remove the metrics graph from the overview and only show the current stats 
> as numbers (cpu load, heap utilization) and add a button to enable the 
> detailed graph.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to