The LOGSEG871 is known to pop up when your recovery log is full. Try creating a
new recovery log volume (doesn't have to be very large, maybe 20 - 100 MB),
then use the "dsmserv extend log <volumename> <size>". For example:

   dsmfmt  -m  -log  /usr/adsmlog/newlogvol  21            <== multiple of four
+ 1, usable size will be 20 MB
   dsmserv  extend  log  /usr/adsmlog/newlogvol  20

Once the log has been extended, you should be able to restart your ADSM server.

If the recovery log is normal mode (use the QUERY STATUS command to determine
this, if you are unsure), then it might be that your ADSM activity has
increased and your recovery log needs to be enlarged. If your recovery log is
in roll-forward mode, then make sure you have the DBBACKUPTRIGGER defined,
which is the safety valve that helps prevent the recovery log from filling up
between regularly scheduled database backups.

You might also consider going out to the ftp site index.storsys.ibm.com and
obtaining the latest level of server code. Earlier levels had problems where
the recovery log would fill up for no apparent reason.

Andy Raibeck
ADSM Level 2 Support

        ADSM-L AT VM.MARIST DOT EDU
        11-24-97 06:12 PM
Please respond to ADSM-L AT VM.MARIST DOT EDU @ internet

To: ADSM-L AT VM.MARIST DOT EDU @ internet
cc:
Subject: Re: Unable to restart AIX ADSM-Server :((

Normally there are some information messages indicating what part of
the startup cycle the server is in.  This gives you a clue as to where
it's dying.  My guess is that it's a corrupt logfile.  If you have them
mirrored, try renaming the first copy so that the server can't find it,
forcing it to use the second copy.

       /
   /  *  /  Alan Hamilton
  *     *   alanh AT primenet DOT com