You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Javier Akira Luca de Tena (Jira)" <ji...@apache.org> on 2020/08/19 04:24:00 UTC

[jira] [Created] (HBASE-24903) 'scandetail' log message is missing when responseTooSlow happens in the rpc that closes the scanner

Javier Akira Luca de Tena created HBASE-24903:
-------------------------------------------------

             Summary: 'scandetail' log message is missing when responseTooSlow happens in the rpc that closes the scanner
                 Key: HBASE-24903
                 URL: https://issues.apache.org/jira/browse/HBASE-24903
             Project: HBase
          Issue Type: Bug
    Affects Versions: 1.4.13
            Reporter: Javier Akira Luca de Tena


'scandetail' log message is missing when responseTooSlow happens in the rpc that closes the scanner.

RSRpcServices.closeScanner is called before logging the slowLog in RPCServer.logResponse.

Since closeScanner removes the scanner entry from scanners map, logResponse can't find scanDetails when calling RSRpcServices.getScanDetailsWithId.

 

I have reproduced it by exhausting the region (no more results in the region), which sets moreResultsInRegion = false and cause to closeScanner in same rpc [https://github.com/apache/hbase/blob/c2e0cf989e4a86169219161d4d889db80288e636/hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/RSRpcServices.java#L3235-L3237]

At least, this seems the behavior in branch-1.

 

I think this bug was introduced at https://issues.apache.org/jira/browse/HBASE-17489.

Note that this is a completely different case than https://issues.apache.org/jira/browse/HBASE-24282.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)