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

Posted to issues@ignite.apache.org by "Ignite TC Bot (JIRA)" <ji...@apache.org> on 2019/04/16 08:19:00 UTC

[jira] [Commented] (IGNITE-11745) Rebalancing overwhelmingly prefers some supplier nodes to others

    [ https://issues.apache.org/jira/browse/IGNITE-11745?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16818760#comment-16818760 ] 

Ignite TC Bot commented on IGNITE-11745:
----------------------------------------

{panel:title=--&gt; Run :: All: Possible Blockers|borderStyle=dashed|borderColor=#ccc|titleBGColor=#F7D6C1}
{color:#d04437}Cache 2{color} [[tests 0 TIMEOUT , Exit Code |https://ci.ignite.apache.org/viewLog.html?buildId=3608333]]
* IgniteClientCacheStartFailoverTest.testClientStartCloseServersRestart (last started)

{color:#d04437}Cache 6{color} [[tests 0 Exit Code |https://ci.ignite.apache.org/viewLog.html?buildId=3608337]]
* ClientFastReplyCoordinatorFailureTest.testClientFastReply (last started)

{color:#d04437}MVCC Cache 2{color} [[tests 1|https://ci.ignite.apache.org/viewLog.html?buildId=3608371]]
* IgniteCacheMvccTestSuite2: IgniteCacheEntryProcessorNodeJoinTest.testEntryProcessorNodeLeave - 0,0% fails in last 146 master runs.

{panel}
[TeamCity *--&gt; Run :: All* Results|https://ci.ignite.apache.org/viewLog.html?buildId=3608384&amp;buildTypeId=IgniteTests24Java8_RunAll]

> Rebalancing overwhelmingly prefers some supplier nodes to others
> ----------------------------------------------------------------
>
>                 Key: IGNITE-11745
>                 URL: https://issues.apache.org/jira/browse/IGNITE-11745
>             Project: Ignite
>          Issue Type: Bug
>          Components: cache
>    Affects Versions: 2.7
>            Reporter: Ilya Kasnacheev
>            Assignee: Ilya Kasnacheev
>            Priority: Critical
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> When cache has backups, and you add third node to cluster, Ignite will only rebalance data from single node.
> When you add n-th node, Ignite will not rebalance from some nodes and it will pull 10x as much data from some nodes than from others.
> This is because we filter static nodes list by partition availability and then pick the first one. Overwhelmingly it is the first nodes in list and nodes towards the end of list will never get to supply partitions.



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