First post - Hopefully in the right place...
Triton Version 6.1c.04.01

When we wanted to stop triton using rc.stop, the shared memory manager refuses to stop.

Both shmmanager -k and -uk do not get the job done.

our rc.stop log:
Stopping TRITON
Check whether we need to stop running bshell`s.
Stop Printer Daemon
Stop Database Processes of TRITON Base
Kill processes
Remove semaphores
Removing Shared Memory

Stop /opt/bse/bin/shmtimer6.1:
Shmtimer started: pid = 3317, time = 1238329995 (Sun Mar 29 14:33:15 2009)
shmtimer6.1: Shmtimer (pid 3317) stopped.
Sleeping 10 seconds to wait for shm_timer to finish ... done
Stopping successful
# of attaches = 0
# of attaches = 0
# of attaches = 0
# of attaches = 0
# of attaches = 0
shmmanager6.1: core_dumped() signal 11 cwd: '/'

The problem occurred last weekend during our weekend backup. We want to avoid the same situation as this causes our backup to wait for the Triton stop.

Hey gerhey, welcome to baanboard,

don't know whether I may be of help.
If this situation occurs again, what does "ipcs -m" show?
I have often seen situations where shmmanager couldn't free shared memory, but usually this is shown by "# of attachments > 0".
What did you do to restart triton? If you rebooted the system and locked shared memory was the problem, you could free shared memory with "ipcrm" without reboot.

good luck and another "welcome" to you

Norbert

Hi Norbert,

Thanks for this update. I have discussed this with our team, and we will investigate this further. It looks like this is pointing in the right direction.
Thank you,
Gert.