hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Yu Li (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-18469) Correct RegionServer metric of totalRequestCount
Date Mon, 31 Jul 2017 04:14:00 GMT

    [ https://issues.apache.org/jira/browse/HBASE-18469?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16106788#comment-16106788
] 

Yu Li commented on HBASE-18469:
-------------------------------

bq. I do not think we should count a request with caching 10 as 10 requests. Also for multi.
It may be called requestRowsCount?
I like the idea of introducing requestRowsCount, which could clearly separate the RS metrics
into two categories:
1. requestCount: The rpc workload, which only count in rpc requests, count as only one no
matter it's single or multi.
2. requestRowsCount: The real workload, which counts per region action, and will count as
many for caching-scan or multi.

While currently for multi we have below codes in {{RsRpcServices#multi}}:
{code}
    for (RegionAction regionAction : request.getRegionActionList()) {
      this.requestCount.add(regionAction.getActionCount());
{code}

So there would be some changes required to do the metrics separation. Let's wait for others'
opinion.

btw, I could see relative efforts/discussion in HBASE-18374.

> Correct  RegionServer metric of  totalRequestCount
> --------------------------------------------------
>
>                 Key: HBASE-18469
>                 URL: https://issues.apache.org/jira/browse/HBASE-18469
>             Project: HBase
>          Issue Type: Bug
>    Affects Versions: 1.2.0
>            Reporter: Shibin Zhang
>            Priority: Critical
>
> when i get the metric ,i found  this three metric may be have some error  as follow :
>     "totalRequestCount" : 17541,
>     "readRequestCount" : 17483,
>     "writeRequestCount" : 1633,



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)

Mime
View raw message