You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@cassandra.apache.org by "Chris Lohfink (JIRA)" <ji...@apache.org> on 2016/01/06 20:44:39 UTC

[jira] [Updated] (CASSANDRA-10975) Histogram buckets exposed in jmx are sorted by count

     [ https://issues.apache.org/jira/browse/CASSANDRA-10975?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Chris Lohfink updated CASSANDRA-10975:
--------------------------------------
    Attachment: 10975-2.2.patch

> Histogram buckets exposed in jmx are sorted by count
> ----------------------------------------------------
>
>                 Key: CASSANDRA-10975
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-10975
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Observability
>            Reporter: Chris Lohfink
>         Attachments: 10975-2.2.patch
>
>
> The estimated histogram snapshot lets its parent provide the getValues implementation which sorts the bucket array:
> https://github.com/dropwizard/metrics/blob/3.1-maintenance/metrics-core/src/main/java/com/codahale/metrics/UniformSnapshot.java#L25
> making it hard to determine what count belonged to what bucket. Along with removal of the pre 2.2 deprecated metrics this makes it nearly impossible to track latencies over time.



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