hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "subramanian raghunathan (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-4215) RS requestsPerSecond counter seems to be off
Date Mon, 22 Aug 2011 09:01:31 GMT

    [ https://issues.apache.org/jira/browse/HBASE-4215?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13088577#comment-13088577

subramanian raghunathan commented on HBASE-4215:

1)  Modifying the requestPersecond value to integer rather than float since the value is used
in several VO's
1) MasterMetrics metrics 
2) AClusterStatus
3) StorageClusterStatusModel
4) AServerLoad etc. 

Also to maintain consistensy the regionservermetrics and HServerLoad using integer values
for requestPersecond.

Advantages: Consitency throughout the system . Changes are minimal.
DisAdvantages: Percision loss, also integer can accomadate less value than the float.

2)HServerLoad.getNumberOfRequests() is used in multiple places.

{color:green}StorageClusterStatusModel - Expects request per second 
ClusterStatus - Expects request per second 
MasterStatusTmplImpl - UI part also Expects request per second {color}

{color:red}AvroUtil - Not sure {color} 

public void regionServerReport(final byte [] sn, final HServerLoad hsl)
  throws IOException {
    this.serverManager.regionServerReport(new ServerName(sn), hsl);
    if (hsl != null && this.metrics != null) {
      // Up our metrics.
The metric expects the total request in the all the region servers and the metric computes
the per second value of the aggregation of all the region servers.{color}

So following change is sugested to keep the existing attribute as the requestPersecond.Also
introduce additional attribute that reports the total request made in RS as totalNumberOfRequests
This will be used by the Master Metrics.

Will upload the patch ASAP 

> RS requestsPerSecond counter seems to be off
> --------------------------------------------
>                 Key: HBASE-4215
>                 URL: https://issues.apache.org/jira/browse/HBASE-4215
>             Project: HBase
>          Issue Type: Bug
>          Components: metrics
>    Affects Versions: 0.92.0
>            Reporter: Todd Lipcon
>            Assignee: ramkrishna.s.vasudevan
>            Priority: Critical
>             Fix For: 0.92.0
> In testing trunk, I had YCSB reporting some 40,000 requests/second, but the summary info
on the master webpage was consistently indicating somewhere around 3x that. I'm guessing that
we may have a bug where we forgot to divide by time.

This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira


View raw message