cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jaakko Laine (JIRA)" <j...@apache.org>
Subject [jira] Updated: (CASSANDRA-150) multiple seeds (only when seed count = node count?) can cause cluster partition
Date Wed, 25 Nov 2009 12:58:39 GMT

     [ https://issues.apache.org/jira/browse/CASSANDRA-150?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Jaakko Laine updated CASSANDRA-150:
-----------------------------------

    Attachment: 150.patch

Added extra condition to send gossip to random seed also if liveEndpoints.size + unreachableEndpoints.size
is less than seeds.size. This will cause us to send same gossip to same seed twice occasionally
(when the seed is live, but we have not yet seen enough nodes), but I think this is OK, as
this is quite special case and will go away as soon as we've seen enough nodes.

Another option would be to add extra parameter excludeThisNode to sendGossip and not send
gossip if random returns that address, but IMHO this option is messy and gains very little.


> multiple seeds (only when seed count = node count?) can cause cluster partition
> -------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-150
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-150
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>            Reporter: Jonathan Ellis
>            Priority: Minor
>         Attachments: 150.patch
>
>
> happens fairly frequently on my test cluster of 5 nodes.  (i normally restart all nodes
at once when updating the code.  haven't tested w/ restarting one machine at a time.)

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message