[
https://issues.apache.org/jira/browse/CASSANDRA-18543?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17725142#comment-17725142
]
Cameron Zemek commented on CASSANDRA-18543:
-------------------------------------------
Another potential solution might be to allow shadow round to allow marking
nodes as alive (ie UP). That is if you are just starting up and talk to another
seed who has view of state of cluster, why can it not just use that as initial
state for alive state?
> Waiting for gossip to settle does not wait for live endpoints
> -------------------------------------------------------------
>
> Key: CASSANDRA-18543
> URL: https://issues.apache.org/jira/browse/CASSANDRA-18543
> Project: Cassandra
> Issue Type: Bug
> Reporter: Cameron Zemek
> Priority: Normal
> Attachments: gossip.patch
>
>
> When a node starts it will get endpoint states (via shadow round) but have
> all nodes marked as down. The problem is the wait to settle only checks the
> size of endpoint states is stable before starting Native transport. Once
> native transport starts it will receive queries and fail consistency levels
> such as LOCAL_QUORUM since it still thinks nodes are down.
> My initial solution to this was to also check live endpoints size in addition
> to size of endpoint states. This worked but I noticed in testing this fix
> that there also a lot of duplication of checking the same node (via Echo
> messages) for liveness. So the patch also removes this duplication of
> checking node is UP.
> The final problem I found while testing is sometimes could still not see a
> change in live endpoints due to only 1 second polling, so the patch allows
> for overridding the settle parameters.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org
For additional commands, e-mail: commits-h...@cassandra.apache.org
