[
https://issues.apache.org/jira/browse/CASSANDRA-16364?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17835024#comment-17835024
]
Jon Haddad commented on CASSANDRA-16364:
----------------------------------------
Just ran into this with the latest 4.0, starting 9 nodes simultaneously, all
marked as seeds. Using this many seeds was an oversight, but I thought we
randomized the tokens in a way that would prevent this from happening.
{noformat}
INFO [GossipStage:1] 2024-04-08 18:51:42,228 StorageService.java:2851 - Nodes
/172.31.30.76:7000 and /172.31.32.145:7000 have the same token
-4365585967229483808. Ignoring /172.31.30.76:7000
INFO [GossipStage:1] 2024-04-08 18:51:42,228 StorageService.java:2851 - Nodes
/172.31.30.76:7000 and /172.31.32.145:7000 have the same token
156850771319184154. Ignoring /172.31.30.76:7000
INFO [GossipStage:1] 2024-04-08 18:51:42,228 StorageService.java:2851 - Nodes
/172.31.30.76:7000 and /172.31.32.145:7000 have the same token
7039551456192731860. Ignoring /172.31.30.76:7000
INFO [GossipStage:1] 2024-04-08 18:51:42,229 StorageService.java:2851 - Nodes
/172.31.30.76:7000 and /172.31.32.145:7000 have the same token
8579899636253633675. Ignoring /172.31.30.76:7000{noformat}
> Joining nodes simultaneously with auto_bootstrap:false can cause token
> collision
> --------------------------------------------------------------------------------
>
> Key: CASSANDRA-16364
> URL: https://issues.apache.org/jira/browse/CASSANDRA-16364
> Project: Cassandra
> Issue Type: Bug
> Components: Cluster/Membership
> Reporter: Paulo Motta
> Priority: Normal
> Fix For: 4.0.x
>
>
> While raising a 6-node ccm cluster to test 4.0-beta4, 2 nodes chosen the same
> tokens using the default {{allocate_tokens_for_local_rf}}. However they both
> succeeded bootstrap with colliding tokens.
> We were familiar with this issue from CASSANDRA-13701 and CASSANDRA-16079,
> and the workaround to fix this is to avoid parallel bootstrap when using
> {{allocate_tokens_for_local_rf}}.
> However, since this is the default behavior, we should try to detect and
> prevent this situation when possible, since it can break users relying on
> parallel bootstrap behavior.
> I think we could prevent this as following:
> 1. announce intent to bootstrap via gossip (ie. add node on gossip without
> token information)
> 2. wait for gossip to settle for a longer period (ie. ring delay)
> 3. allocate tokens (if multiple bootstrap attempts are detected, tie break
> via node-id)
> 4. broadcast tokens and move on with bootstrap
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
