VKTM background process terminating the DB instance.
Date: Wed, 11 Aug 2021 12:24:14 +0530
Message-ID: <CAOzfMup81SoFUQqvYWsk9rv_QcrbgL3sigCvjO+8yhmxcogqfg_at_mail.gmail.com>
Hi All,
Envt Details: DB version 12.1.0.2 Apr 2019 PSU , Running on VM RHEL 7.7
We have an issue wherein we have 6 DB instances running on different VM's sharing the same host getting terminated. The background process VKTM dies and subsequently PMON terminates the instance. Raised an SR with Oracle a couple of weeks back but so far they have not been able to find the root cause.Also i did follow some of the recommendations given the below MOS notes, but that also didn't help. Prior to VKTM getting terminated , the strace just shows "nanosleep({0, 1000000}, 0x7fff29abd7f0) = 0" in its trace file. Please find the below gist file which contains strace information of VKTM process just before it got terminated.
https://gist.github.com/aryangoti/7fd0882352e4d6e306d212ff1ab0a432
MOS Notes:
Customer RecommendedVKTM processes consume High CPU (Doc ID 1992652.1)
Patch For Excessive VKTM And VKRM Traces In Bug 9382956 Requires Events
(Doc ID 1381270.1)
Alert log snippet:
Wed Aug 11 01:30:55 2021
Instance Critical Process (pid: 4, ospid: 8566, VKTM) died unexpectedly PMON (ospid: 8561): terminating the instance due to error 56703 Wed Aug 11 01:30:55 2021
System state dump requested by (instance=1, osid=8561 (PMON)), summary=[abnormal instance termination]. System State dumped to trace file
/xxx/xxxxx/diag/rdbms/xxxx/xxxxx/trace/xxxx_diag_8576_20210811013055.trc Wed Aug 11 01:30:56 2021
Dumping diagnostic data in directory=[cdmp_20210811013055], requested by
(instance=1, osid=8561 (PMON)), summary=[abnormal instance termination].
Wed Aug 11 01:30:57 2021
Instance terminated by PMON, pid = 8561
Appreciate of i can get some help here!
Thanks,
Goti
-- http://www.freelists.org/webpage/oracle-lReceived on Wed Aug 11 2021 - 08:54:14 CEST