You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@trafficserver.apache.org by "Leif Hedstrom (JIRA)" <ji...@apache.org> on 2014/09/03 05:34:52 UTC
[jira] [Commented] (TS-149) Split ram and disk cache hit response
times into separate metrics
[ https://issues.apache.org/jira/browse/TS-149?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14119263#comment-14119263 ]
Leif Hedstrom commented on TS-149:
----------------------------------
[~bcall] I think this is actually done already, see TS-2944 and TS-3036. Please close as a dupe of TS-2944 if that's the case (and remove fix version)
> Split ram and disk cache hit response times into separate metrics
> -----------------------------------------------------------------
>
> Key: TS-149
> URL: https://issues.apache.org/jira/browse/TS-149
> Project: Traffic Server
> Issue Type: New Feature
> Components: Cache
> Reporter: Anirban
> Assignee: Bryan Call
> Priority: Minor
> Labels: A
> Fix For: 5.2.0
>
>
> Split ram and disk cache hit response times into separate metrics
> Original description
> by Stephen Bisordi from Yahoo
> It appears that the response time for ram cache hits and disk cache hits is captured as a single metric
> (transaction_totaltime.hit_fresh). Would it be possible to split this into two separate metrics, or to add a ram cache
> hit result code to the log to differentiate between ram and disk hits (similar to Squid's TCP_MEM_HIT)?
> This change would allow for quicker troubleshooting of certain issues such as poor disk performance.
>
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)