Re: process M000 holds library cache lock

From: De DBA <dedba_at_tpg.com.au>
Date: Wed, 28 Mar 2012 23:50:38 +1000
Message-ID: <4F73172E.7030909_at_tpg.com.au>

Hi Remigiusz,
Most likely nothing else than a restart of the process. It is a slave process of mmon. As my test instance has no m000 at the moment, I just killed mmon to prove the point.. :-)

This is what happened:

$ pgrep -lf mmon_TST2
14079 ora_mmon_TST2
$ sudo kill -9 14079
[sudo] password for lingent:

$ pgrep -lf mmon_TST2                                                        # MMON on the TST2 instance is dead

$ tail /u00/oracle/diag/rdbms/tst2/TST2/trace/alert_TST2.log Starting background process VKRM
Wed Mar 28 22:00:00 2012
VKRM started with pid=21, OS id=20915
Wed Mar 28 22:00:30 2012
Thread 1 advanced to log sequence 59 (LGWR switch)

   Current log# 2 seq# 59 mem# 0: /u00/oracle/oradata/TST2/redo02.log Wed Mar 28 23:42:02 2012

*Restarting dead background process MMON*                        # It's dead has been noticed by the instance
Wed Mar 28 23:42:02 2012
MMON started with pid=15, OS id=22229                                # mmon is restarted.8-)

$ pgrep -lf mmon_TST2
22229 ora_mmon_TST2

Since brutally killing mmon has as only effect that it is restarted, I would conclude that any slaves of it can be safely killed as well... (but maybe others have one that they can kill to test)

Hth,
Tony

On 28/03/12 18:47, Remigiusz Sokolowski wrote:
> hi,
>
> anyone tried to kill M000 - what are possible implications?
> Will the instance die?
> It just took library cache mutex (X) and holds for very considerable
> amount of time.
>
> Regards
> Remigiusz

--
http://www.freelists.org/webpage/oracle-l

Received on Wed Mar 28 2012 - 08:50:38 CDT