cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Brandon Williams (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-3242) Nodes which are not in the ring but are dead in gossip prevent truncation
Date Thu, 22 Sep 2011 10:41:26 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-3242?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13112464#comment-13112464
] 

Brandon Williams commented on CASSANDRA-3242:
---------------------------------------------

bq. Could isAnyHostDown in StorageProxy.java be modified to not care about unreachable nodes
that are not in the ring?

I'm not sure that's the best course of action since it's potentially dangerous.  Maybe a 'truncate
force' so the operator has to be aware of the danger first, but we should figure out why there
is an unreachable node listed after CASSANDRA-3071 which should have solved that.

> Nodes which are not in the ring but are dead in gossip prevent truncation
> -------------------------------------------------------------------------
>
>                 Key: CASSANDRA-3242
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-3242
>             Project: Cassandra
>          Issue Type: Bug
>    Affects Versions: 0.8.5
>            Reporter: Jason Harvey
>            Assignee: Brandon Williams
>            Priority: Trivial
>
> All of the nodes in my ring are up, however I do have an 'unreachable' node in gossip
which was removed some time ago. When I attempt to run a truncation, it does to execute since
some nodes are dead, despite those nodes not owning anything in the ring.
> Could isAnyHostDown in StorageProxy.java be modified to not care about unreachable nodes
that are not in the ring?

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message