trafodion-codereview mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From selvaganesang <...@git.apache.org>
Subject [GitHub] incubator-trafodion pull request #931: [TRAFODION-2351] Bulk load with log e...
Date Thu, 26 Jan 2017 05:47:26 GMT
Github user selvaganesang commented on a diff in the pull request:

    https://github.com/apache/incubator-trafodion/pull/931#discussion_r97934559
  
    --- Diff: core/sql/executor/ExExeUtilLoad.cpp ---
    @@ -1442,7 +1442,8 @@ short ExExeUtilHBaseBulkLoadTcb::work()
                ComCondition *cond;
                Lng32 entryNumber;
                while ((cond = diagsArea->findCondition(EXE_ERROR_ROWS_FOUND, &entryNumber))
!= NULL) {
    -              errorRowCount = cond->getOptionalInteger(0);
    +              if (errorRowCount < cond->getOptionalInteger(0))
    --- End diff --
    
    Actually, the changes in ExHdfsScan.cpp is the one needed to take care of the core dump
issue.  To honor the command 'stop after n error rows' when the data is loaded into trafodion
table, a counter is maintained in hbase to increment the number of error rows from all the
ESP process.  Each load prepare command has its own counter id. This counter id is visible
to the ExHbaseAccessBulkLoadPrepSQTcb operator. This operator hosted in multiple ESPs will
read the counter at the end of its operation and pass it as a warning. So all the ESPs are
expected to have the same value but if an esp finishes early it may not have read the final
counter value. Hence I am taking the max error counter value passed in the warnings.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

Mime
View raw message