[jira] [Created] (TRAFODION-1109) LP Bug: 1438775 - mxosrvr core in SRVR::FETCH2bulk when columns in particular order

["Alice Chen (JIRA)" <ji...@apache.org>]

Alice Chen created TRAFODION-1109:
-------------------------------------

             Summary: LP Bug: 1438775 - mxosrvr core in SRVR::FETCH2bulk when columns in particular order
                 Key: TRAFODION-1109
                 URL: https://issues.apache.org/jira/browse/TRAFODION-1109
             Project: Apache Trafodion
          Issue Type: Bug
          Components: sql-exe
            Reporter: Aruna Sadashiva
            Assignee: Tharak Capirala
            Priority: Blocker
             Fix For: 1.1 (pre-incubation)


First select crashes, but the second one works:

Select exec_start_utc_ts, exec_end_utc_ts, query_id, query_text from “_REPOS_”.metric_query_table;
Select query_text, exec_start_utc_ts, exec_end_utc_ts, query_id  from “_REPOS_”.metric_query_table;

This system had about 350 rows in metric_query_table. It worked ok on a system where metric_query_table had 40 rows. 

Stack trace: 
#0  0x00007ffff4829625 in raise () from /lib64/libc.so.6
#1  0x00007ffff482ad8d in abort () from /lib64/libc.so.6
#2  0x00007ffff4867537 in __libc_message () from /lib64/libc.so.6
#3  0x00007ffff486ce66 in malloc_printerr () from /lib64/libc.so.6
#4  0x00007ffff486f9b3 in _int_free () from /lib64/libc.so.6
#5  0x00007ffff685f53b in SRVR::FETCH2bulk (pSrvrStmt=0x1ca8650) at sqlinterface.cpp:6111
#6  0x00007ffff688dbda in odbc_SQLSrvr_FetchPerf_sme_ (objtag_=<optimized out>, call_id_=<optimized out>, 
    returnCode=0x7fffe4fde67c, dialogueId=<optimized out>, stmtLabel=0xed7f04 "SQL_CUR_2", maxRowCnt=100, maxRowLen=0, 
    sqlAsyncEnable=0, queryTimeout=0, rowsAffected=0x7fffe4fde678, outValuesFormat=0x7fffe4fde674, 
    outputDataValue=0x7fffe4fde650, sqlWarningOrErrorLength=0x7fffe4fde670, sqlWarningOrError=@0x7fffe4fde668: 0x0)
    at srvrothers.cpp:5815
#7  0x00000000004b6907 in odbc_SQLSrvr_Fetch_ame_ (objtag_=0xed8160, call_id_=0xed81b8, dialogueId=1449409261, 
    operation_id=<optimized out>, sqlAsyncEnable=0, queryTimeout=0, stmtHandle=0, stmtLabel=0xed7f04 "SQL_CUR_2", 
    maxRowCnt=100, maxRowLen=0) at SrvrConnect.cpp:7498
#8  0x000000000048b852 in SQLFETCH_IOMessage (objtag_=0xed8160, call_id_=0xed81b8, operation_id=3009)
    at Interface/odbcs_srvr.cpp:1021
#9  0x0000000000493d7e in DISPATCH_TCPIPRequest (objtag_=0xed8160, call_id_=0xed81b8, operation_id=<optimized out>)
    at Interface/odbcs_srvr.cpp:1803
#10 0x0000000000433872 in BUILD_TCPIP_REQUEST (pnode=0xed8160) at ../Common/TCPIPSystemSrvr.cpp:603
#11 0x000000000043420d in PROCESS_TCPIP_REQUEST (pnode=0xed8160) at ../Common/TCPIPSystemSrvr.cpp:581
#12 0x00000000004623f6 in CNSKListenerSrvr::tcpip_listener (arg=0xda1a50) at Interface/linux/Listener_srvr_ps.cpp:400
#13 0x00007ffff45dd290 in sb_thread_sthr_disp (pp_arg=0xed74d0) at threadl.cpp:253
#14 0x00007ffff43a99d1 in start_thread () from /lib64/libpthread.so.0
#15 0x00007ffff48df8fd in clone () from /lib64/libc.so.6

According to Arvind, crash happens in this part of the code (sqlinterface.cpp):

        //Changes due to cursor issue
        if (pSrvrStmt->maxRowCnt > 0)
        {
                if (pSrvrStmt->outputDescVarBufferLen > 0)
                {
                        if( srvrGlobal->m_FetchBufferSize/pSrvrStmt->outputDescVarBufferLen  < pSrvrStmt->maxRowCnt )
                        {
                                if (pSrvrStmt->outputDescVarBuffer != NULL)
                                        delete pSrvrStmt->outputDescVarBuffer;
                                pSrvrStmt->outputDescVarBuffer = NULL;
                                markNewOperator,pSrvrStmt->outputDescVarBuffer =
 new BYTE[pSrvrStmt->maxRowCnt*pSrvrStmt->outputDescVarBufferLen];
                                if (pSrvrStmt->outputDescVarBuffer == NULL)



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