Re: MMNL absent message in alert log

On Dec 13, 3:23 pm, eAddict <eAdd..._at_gmail.com> wrote:
> I have shut down SAP but can't even get in at an SQLplus prompt...
> There are 3 processes that
> are taking up A HUGE amount of resources.
>
> CPU TTY PID USERNAME PRI NI SIZE RES STATE TIME %WCPU %CPU
> COMMAND
> 0 ? 27589 orabpr 241 20 4053M 9536K run 111:29 100.09 99.92
> oracleBPR
> 4 ? 14547 orabpr 241 20 4049M 6720K run 102:27 99.77 99.60
> oracleBPR
> 1 ? 14451 orabpr 241 20 4071M 15760K run 282:38 99.67 99.49
> oracleBPR
>
> What happens if I just outright kill the processes? There are no
> users connected since SAP and the listener are down.

A similar message appeared once in the alert logs of all database instances on my server (at roughly the same time) one week after I applied the daylight saving patch earlier this year. I am not using the features provided by MMNL, so the loss of functionality provided by MMNL was not noticed. According to "Expert Oracle Database 10g Administration", "The MMNL process... performs lightweight manageabiity tasks, including computing metrics and capturing session history information for the ASH feature in some circumstances. For example, MMNL will flush ASH data to disk if the ASH memory buffer fills up before the one-hour interval that would normally cause MMON to flush it."

This raises the question, have you hit a maximum specified size for a tablespace, or run out of room on the disk for a tablespace to autoextend? Are the redo logs failing to be archived due to lack of disk space?

Are you able to connect as the internal user, which should be able to connect even when SYS cannot (for instance, if all online redo logs are full and insufficient space is available in the archived redo location)? If you bounce the listener, are you able to connect as the internal user?

Charles Hooper
IT Manager/Oracle DBA
K&M Machine-Fabricating, Inc. Received on Fri Dec 14 2007 - 08:36:26 CST