[jira] [Comment Edited] (ZOOKEEPER-1777) Missing ephemeral nodes in one of the members of the ensemble

["Flavio Junqueira (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/ZOOKEEPER-1777?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13784591#comment-13784591 ] 

Flavio Junqueira edited comment on ZOOKEEPER-1777 at 10/2/13 10:57 PM:
-----------------------------------------------------------------------

This is a bit confusing. Since you wiped out the state of B, you lost quorum for the transactions that A and B had committed. It is possible that A had a snapshot containing txns in 7d to 9a which confused it and got you in this broken scenario. If A didn't have such a snapshot, then it should have truncated and received a diff. Forcing a snapshot transfer shouldn't be necessary. 


was (Author: fpj):
This is a bit confusing. Since you wiped out the state of B, you lost quorum for the transactions that A and B had committed. It is possible that A had a snapshot containing txns in 7d to 9a which confused it and got you in this broken scenario. If A didn't have such a snapshot, then it should have truncated and received a diff. Forcing a snapshot shouldn't be necessary. 

> Missing ephemeral nodes in one of the members of the ensemble
> -------------------------------------------------------------
>
>                 Key: ZOOKEEPER-1777
>                 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-1777
>             Project: ZooKeeper
>          Issue Type: Bug
>          Components: quorum
>    Affects Versions: 3.4.5
>         Environment: Linux, Java 1.7
>            Reporter: Germán Blanco
>            Assignee: Germán Blanco
>            Priority: Blocker
>             Fix For: 3.4.6, 3.5.0
>
>         Attachments: snaps.tar
>
>
> In a 3-servers ensemble, one of the followers doesn't see part of the ephemeral nodes that are present in the leader and the other follower. 
> The 8 missing nodes in "the follower that is not ok" were created in the end of epoch 1, the ensemble is running in epoch 2.



--
This message was sent by Atlassian JIRA
(v6.1#6144)