SRVCTL shows instance down while instance is up [message #464436] |
Thu, 08 July 2010 01:55 |
ridhi_sundar
Messages: 184 Registered: November 2007 Location: Bangalore
|
Senior Member |
|
|
Hi All
when I use srvct status to check the status of the instances it shows the instnce is down while the instance is actually up.
[qaoraerp@qa1-rora ~]$ srvctl status database -d QAERPRPT
Instance QAERPRPT1 is not running on node qa1-rora
Instance QAERPRPT2 is running on node qa2-rora
[qaoraerp@qa1-rora ~]$
ps -ef|grep pmon shows the instance up
[qaoraerp@qa1-rora ~]$ ps -ef|grep pmon
qaoraerp 1314 27205 0 01:51 pts/3 00:00:00 grep pmon
qaoraerp 10798 1 0 Jun27 ? 00:06:27 ora_pmon_D1ERP1
qaoraerp 10800 1 0 Jun27 ? 00:07:42 ora_pmon_D1ERPRPT1
qaoraerp 10802 1 0 Jun27 ? 00:08:00 ora_pmon_QAERP1
d1oraerp 10958 1 0 Jun27 ? 00:01:03 ora_pmon_D1P2P1
qaoraerp 10987 1 0 Jun27 ? 00:00:17 asm_pmon_+ASM1
qaoraerp 20682 1 0 Jun28 ? 00:04:25 ora_pmon_QAERPRPT1
qaoraerp 21260 1 0 Jun28 ? 00:00:35 ora_pmon_QAP2P1
even I can log in to the instance and status is open.
SQL> select instance_name,status from v$instance;
INSTANCE_NAME STATUS
---------------- ------------
QAERPRPT1 OPEN
So it is clear that the instance is up. Can anyone tell what might be the reason for srvctl not to detect it?
Thanks
Ridhi Sundar
|
|
|
Re: SRVCTL shows instance down while instance is up [message #464445 is a reply to message #464436] |
Thu, 08 July 2010 02:08 |
John Watson
Messages: 8960 Registered: January 2010 Location: Global Village
|
Senior Member |
|
|
The clusterware does sometimes get it wrong. For example, when you first register your instance in the OCR withsrvctl add instance -d dbname -i instancename it will show as "not running", I guess because at the time it was started the clusterware didn't know it existed. Just issue srvctl start instance -d QAERPRPT -i QAERPRPT1 nothing will be started (because it already is) but iut will be reported correctly from then.
|
|
|
Re: SRVCTL shows instance down while instance is up [message #464495 is a reply to message #464445] |
Thu, 08 July 2010 03:42 |
ridhi_sundar
Messages: 184 Registered: November 2007 Location: Bangalore
|
Senior Member |
|
|
Hi John
Thanks for the responce. But it did not work.
[qaoraerp@qa1-rora ~]$ srvctl start instance -d QAERPRPT -i QAERPRPT1
PRKP-1001 : Error starting instance QAERPRPT1 on node qa1-rora
CRS-0233: Resource or relatives are currently involved with another operation.
[qaoraerp@qa1-rora ~]$ srvctl status database -d QAERPRPT
Instance QAERPRPT1 is not running on node qa1-rora
Instance QAERPRPT2 is running on node qa2-rora
[qaoraerp@qa1-rora ~]$
|
|
|
Re: SRVCTL shows instance down while instance is up [message #464507 is a reply to message #464495] |
Thu, 08 July 2010 04:14 |
John Watson
Messages: 8960 Registered: January 2010 Location: Global Village
|
Senior Member |
|
|
Hello again - this looks as though you have a real problem, not just confused clusterware. I certainly can't give a simple answer.
If it were one of my databases, I would start by looking at all the log files. And probably try to shutdown the all the resources, using sqlplus, srvctl, crs_stop, and so on. Since you have several instances on the node, I guess you can't re-start the clusterware?
Sometimes, I have had databases in a similar state: everything is working perfectly for the users, but from our side it is un-manageable. So just leave it running, and re-start the entire environment at your next maintenance window. Good luck - and perhaps your users won't notice any problems.
|
|
|
|
|
|