I had been running 8.2 on Linux 2.6 for about a month without any
problems. Suddenly yesterday, my database connection was severed. I
tried to re-connect but could not (I've forgotten the error message).
I tried db2stop && db2start, which returned successfully, but I could
not connect to the database, got an error message regarding RESTARTing.
I tried RESTARTing but received the same error message. I rebooted
the box, and now:
$ db2start
02/10/2006 14:57:16 0 0 SQL1220N The database manager shared
memory set cannot be allocated.
SQL1032N No start database manager command was issued. SQLSTATE=57019
Shashi Mannepalli - 10 Feb 2006 22:23 GMT
Your host is running out of memory...etc. Try starting the instance
after doing a db2_kill.
If it is not successfull.....try reboot of the host. Later reduce
memory related configuration paramters.
cheers
Shashi Mannepalli
JasonFriedman80238 - 10 Feb 2006 23:34 GMT
I should have mentioned that there's plenty of memory:
$ free
total used free shared buffers
cached
Mem: 4151320 604644 3546676 0 231616
172164
-/+ buffers/cache: 200864 3950456
Swap: 8385920 0 8385920
Serge Rielau - 11 Feb 2006 00:57 GMT
> I should have mentioned that there's plenty of memory:
> $ free
[quoted text clipped - 4 lines]
> -/+ buffers/cache: 200864 3950456
> Swap: 8385920 0 8385920
Could it be you changed the DB or DBM CFG before it happened.
Check your DB and DBM CFG fo any out of wack heap sizes.
Cheers
Serge

Signature
Serge Rielau
DB2 Solutions Development
IBM Toronto Lab
cheng.j3@gmail.com - 11 Feb 2006 03:14 GMT
Could you check the shmmax on that box ?
eg :
> cat /proc/sys/kernel/shmmax
1073741824
If your one is too small , you may need increase it .
> echo 1073741824 >/proc/sys/kernel/shmmax
You need add this to /etc/sysctl.conf, so this can be picked up
whenever system reboot .
Cheng.