You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Vladimir Ozerov (JIRA)" <ji...@apache.org> on 2019/03/11 08:12:00 UTC

[jira] [Commented] (IGNITE-10841) Edge-chasing deadlock detection monitoring

    [ https://issues.apache.org/jira/browse/IGNITE-10841?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16789272#comment-16789272 ] 

Vladimir Ozerov commented on IGNITE-10841:
------------------------------------------

IGNITE-11516

> Edge-chasing deadlock detection monitoring
> ------------------------------------------
>
>                 Key: IGNITE-10841
>                 URL: https://issues.apache.org/jira/browse/IGNITE-10841
>             Project: Ignite
>          Issue Type: Task
>          Components: mvcc
>            Reporter: Ivan Pavlukhin
>            Priority: Major
>
> A deadlock detection process introduces several effects in the system which should be visible for an administrator. Here are couple of such effects:
> 1. Rolling back deadlocked transactions.
> 2. Introducing messaging overhead even in case when there are not deadlocks.
> Several metrics can be exposed via JMX and/or SQL view. Here are some candidates:
> 1. Number of detected deadlocks and involved transactions.
> 2. Sent/received deadlock detection messages count.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)