[ 
https://issues.apache.org/jira/browse/CASSANDRA-21496?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

George updated CASSANDRA-21496:
-------------------------------
     Bug Category: Parent values: Degradation(12984)Level 1 values: Performance 
Bug/Regression(12997)
       Complexity: Low Hanging Fruit
      Component/s: Consistency/Streaming
    Discovered By: Code Inspection
         Severity: Normal
           Status: Open  (was: Triage Needed)

> RemoveNodeStreams and ReconfigureCMS RESTORE_REPLICA_COUNT streams received 
> by replicas are mis-parsed as OTHER.
> ----------------------------------------------------------------------------------------------------------------
>
>                 Key: CASSANDRA-21496
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-21496
>             Project: Apache Cassandra
>          Issue Type: Bug
>          Components: Consistency/Streaming
>            Reporter: Aleksey Yeschenko
>            Assignee: George
>            Priority: Normal
>          Time Spent: 10m
>  Remaining Estimate: 0h
>
> {{RemoveNodeStreams#startStreaming()}} and {{ReconfigureCMS#streamRanges()}} 
> build an {{INITIATE_DATA_MOVEMENTS_REQ}} payload
> whose {{streamOperation}} field is set to 
> {{StreamOperation.RESTORE_REPLICA_COUNT.name()}}, i.e. the enum constant name 
> {{RESTORE_REPLICA_COUNT}}.
> {{DataMovementVerbHandler#doVerb()}} on the receiver parses that string with 
> {{StreamOperation.fromString(message.payload.streamOperation))}},
> incorrectly getting {{StreamOperation.OTHER}} (since 
> {{"RESTORE_REPLICA_COUNT" != "Restore replica count"}}).
> Current impact: this causes MVs to require rebuild unnecessarily on 
> removenode.
> Future impact: this affects CEP-45 topology change logic. I have a temporary 
> workaround there but it's inelegant.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to