Running "ipcs" you can see if there are any orphaned shared memory, semaphore, or message queues allocated. If there are, and you're *sure* that you've stopped all the db2 instance(s) (including the administration instance) use ipcrm to manually clean-up. You'll probably have to be root to execute ipcrm. And be careful, it's very easy to remove something.
A ksh script like this should work, but do your own testing. for i in `ipcs -q | grep $USER | cut -c2-8` do ipcrm -q $i done for i in `ipcs -m | grep $USER | cut -c2-8` do ipcrm -s $i done for i in `ipcs -s | grep $USER | cut -c2-8` do ipcrm -s $i done This used to be more common back in v1&2 than it does now (I have not had to do this in years), and am not sure it's even required anymore. > explain the use of ipcrm? I have needed > to use db2_kill when TSM was taken down before DB2 finished > archiving a log file and DB2 refused to stop. ===== To unsubscribe, send 'unsubscribe' to [EMAIL PROTECTED] For other info (and scripts), see http://people.mn.mediaone.net/scottrmcleod
