Filed 03-AUG-2000 Updated 15-SEP-2000 Product Oracle Server - Enterprise Edition V7 Product Version 8.1.6.1.0 Platform Generic Platform Version 2.2.14-5.0SMP Database Version 8.1.6.1.0 Affects Platforms Port-Specific Priority Severe Loss of Service Status Additional Information Requested Base Bug N/A Fixed in Product Version No Data Problem statement: PMON FAILS WITH ORA-601 -------------------------------------------------------------------------------- PROBLEM: . 1. Clear description of the problem encountered: PMON fails with "ORA-601 cleanup lock conflict" when using mts and there are some active concurrent sessions. After that instace falls down. 2. Pertinent configuration information (MTS/OPS/distributed/etc) MTS 3. Indication of the frequency and predictability of the problem Always reproducible 4. Sequence of events leading to the problem When there are some concurrent sessions with mts working in the database, the error ORA-601 is generated by pmon 5. Technical impact on the customer. Include persistent after effects. Pmon failure causes an instance failure . ========================= DIAGNOSTIC ANALYSIS: This issue is described in bug:1109425 and has not been solved. In our case we have a testcase. . The problem could not be reproduced in a Solaris environment. ========================= WORKAROUND: Do not use mts . ========================= RELATED BUGS: bug:1109425, bug:987647 . ========================= REPRODUCIBILITY: . 1. State if the problem is reproducible; indicate where and predictability Always reproducible 2. List the versions in which the problem has reproduced 8.1.5.0.2, 8.1.6.1.0 over SuSE Linux 6.4 and Red Hat Linux release 6.2. Both of them have kernel version 2.2 3. List any versions in which the problem has not reproduced . ========================= TESTCASE: I could reproduce this on otclnx1 - Enable mts. The following mts parameter were used, the rest were defined by default: mts_dispatchers=("tcp,1") mts_listener_address= "(ADDRESS=(PROTOCOL=tcp)(PORT=)(HOST=))" - import file ora601.dmp which imports tables emp (and dept - have unless 10 mts sessions of sqlplus running the following query: select b.job,sum(a.sal) from emp a, emp b, dept c where a.empno = b.mgr and a.deptno = c.deptno group by b.job; . . ======================== STACK TRACE: ksedmp ksddoa ksdpcg ksdpec ksfpec kgesev ksesec0 ksliwat kslwaitns kskthbwt kslwait ksbsrv kmmssv kmmlsa kmmlod ksucln ksbrdp opirip opidrv sou2o main libc_start_main init . ========================= 24 HOUR CONTACT INFORMATION FOR P1 BUGS: . ========================= DIAL-IN INFORMATION: . ========================= IMPACT DATE: ASAP . . ************** . Aditional information ===================== I emailed to Jamshed Patel (Developer Manager of Linux), who recommended to open a new bug about it, knowing that Bug:1109425 has not been solved. Files were put in ess30: /bug/bug1371144 . ora601.dmp and a.sql with the query. Will start looking at this.... In the meantime can you provide the ORA-601 tracefile along with alert log covering the time the instance was started to the time the error occured. thanks. Files are in ess30: /bug/bug1371144 alert_V816.log pmon_24994.trc Thanks. Please can you set EVENT="10246 trace name context forever, level 1" in the init.ora file and reproduce once more. This should give us a (large) pmon trace file telling us a little more of what pmon's up to. After the problem reproduces, please update this trace along with the additional ORA-601 tracefile. The definition of event 10246 only generated this additional message: . PMON timeout in cleaning up Waiting for event = [process startup], 2 9 0 . I put the alert and pmon trace files in ess30: alert_V816_08aug.log, pmon_30191_08aug.trc Thanks for this, I do appreciate your help. I have another customer with same problem and exactly the same trace file - mts, 8.1.6.0, Linux version is RedHat 6.0. Is there anything we could do to help you with debugging? alert and PMON's files are at ss30.us.oracle.com:8001 /US TAR1159473.999 Not at the moment, as soon as I get chance to update this I will. I'm currently in various training courses in the US, so an update probably isn't going to happen until I'm back in the UK next week. Thanks. When the server starts a MTS server process, it encounters lock conflict error(601) leading to killing of PMON process. Looking into the dump to see the root cause of the pbm. Test patch for this bug is available at tcpatch.us /u01/patch/LINUX/8.1.6.1/bug1371144. . Could you please apply this patch at the customer site and update about the status.