You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@trafodion.apache.org by "Alice Chen (JIRA)" <ji...@apache.org> on 2015/07/22 20:18:09 UTC

[jira] [Created] (TRAFODION-784) LP Bug: 1395132 - Error handling hangs parallel query while hbase blocks

Alice Chen created TRAFODION-784:
------------------------------------

             Summary: LP Bug: 1395132 - Error handling hangs parallel query while hbase blocks
                 Key: TRAFODION-784
                 URL: https://issues.apache.org/jira/browse/TRAFODION-784
             Project: Apache Trafodion
          Issue Type: Bug
          Components: sql-exe
            Reporter: Apache Trafodion
            Assignee: Apache Trafodion
             Fix For: 1.1 (pre-incubation)


Related to bug 1387189. While investigating a hang, we noticed this sequence:
1. Some error cause master executor to cancel a query.
2. For some reason, the Hbase API, which is blocking, doesn't return to ExHbaseAccessTcb operators in the ESPs.
3. The master waits forever for the ESPs to reply to the canceled request
4. The ESPs don't process the cancel, because they are blocked in the Hbase API.

The result is a hang. With ESP parallel queries, it should be possible for the master to timeout it wait, stop unresponsive ESPs and then continue normally.  Of course, this will not prevent hangs if the query does not use ESPs.
Assigned to LaunchPad User Mike Hanlon



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)