[
https://issues.apache.org/jira/browse/CASSANDRA-17682?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17552814#comment-17552814
]
Brandon Williams commented on CASSANDRA-17682:
----------------------------------------------
I will look when I get a chance but you should probably reach for help from
community, since I personally don't see a clear bug here (what does the
authorizer have to do with anything?) and this jira is for C* development, not
support.
https://cassandra.apache.org/_/community.html#discussions
> index read is from one node when authenticator is passwordAuthenticator
> -----------------------------------------------------------------------
>
> Key: CASSANDRA-17682
> URL: https://issues.apache.org/jira/browse/CASSANDRA-17682
> Project: Cassandra
> Issue Type: Bug
> Reporter: wei song
> Priority: Normal
>
> I have a cluster which replication_factor is 3 like:
> {code:java}
> Datacenter: DC1
> ================
> Status=Up/Down
> |/ State=Normal/Leaving/Joining/Moving
> -- Address Load Tokens Owns Host ID
> Rack
> UN 192.168.242.129 271.1 MiB 256 ?
> bda86425-b6cd-479d-a410-c9bd8826d505 rack1
> UN 192.168.242.134 256.13 MiB 256 ?
> a1935992-3443-4f02-9496-23def15d950d rack1
> UN 192.168.242.135 270.81 MiB 256 ?
> 8031e6fd-6d60-4fb2-8323-14f93f4c1468 rack1
> {code}
> While authenticator is passwordAuthenticator，all read disk IO for index reads
> comes from a node。Read disk IO of other node is 0. it is just like that all
> index read is from one node.
> By the way, index read connot get from nodetool tablestats of cassandra3.11.x
>
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
