[
https://issues.apache.org/jira/browse/CASSANDRA-13540?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16033481#comment-16033481
]
Dennis Noordzij commented on CASSANDRA-13540:
---------------------------------------------
Yep, I screwed up. My bootstrap script was repeating the replacement of the
default cassandra user on an existing cluster, which is done with the default
cassandra user of course. That's where this error came from.
Sorry for consuming your time here...
> QUORUM CL is used for new superuser
> -----------------------------------
>
> Key: CASSANDRA-13540
> URL: https://issues.apache.org/jira/browse/CASSANDRA-13540
> Project: Cassandra
> Issue Type: Bug
> Components: Distributed Metadata
> Reporter: Dennis Noordzij
>
> After bootstrapping Cassandra, we're creating a new superuser, and setting RF
> of system_auth keyspace to 2, NetworkTopologyStrategy. Then a nodetool repair
> on the system_auth KS.
> Documentation says
> {quote}
> The system_auth keyspace uses a QUORUM consistency level when checking
> authentication for the default cassandra user. For all other users created,
> superuser or otherwise, a LOCAL_ONE consistency level is used for
> authenticating.
> {quote}
> But for my new superuser, new node members are rejected because QUORUM CL
> can't be achieved:
> {code}
> May 19th 2017, 17:40:14.462 Connection error: ('Unable to connect to any
> servers', {'xx.xx.xx.xx': AuthenticationFailed('Failed to authenticate to
> xx.xx.xx.xx: Error from server: code=0100 [Bad credentials]
> message="org.apache.cassandra.exceptions.UnavailableException: Cannot achieve
> consistency level QUORUM"',)})
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
