Remedy BRIE Process did not shut down on ARSystem stop

[Lammey, Peter A.]

Title: Remedy BRIE Process did not shut down on ARSystem stop

**

We experienced a problem this morning on one of our Remedy 6.3 servers where connection was lost to the database.
The database went down in actuality this morning for 6 minutes and then came back up.
We have a Unix service that checks for the application to be up and if it is not it stops and restarts Remedy by executing the arsystem stop and arsystem start.

We found this morning that the stop of arsystem did not bring down the BRIE process on the server.
When I contacted Remedy they indicated they have bug on this issue and a workaround that might work for us.
A while back we experienced an issue like this when we were setting up our development server and we noticed multiple BRIE processes and applied their same workaround code but it didn’t seem to work.

<<BRIEarconffix.txt>>
Has anyone else experienced this same kind of issue and found a way to correct it?

Thanks
Peter Lammey
ESPN MIT Technical Services & Applications Management
860-766-4761

__20060125_______________________This posting was submitted with HTML in it___

Modify the arsystem script and add the code below right before the line:

   return 0

in the STOP_SERVER() function.

   #stop the brsvcbrie daemon
   briepid=`$PS -aef | egrep "brsvcbrie" | egrep "${SERVER_TO_STOP}" | egrep -v 
egrep | awk '{print $2}'`
   if [ "${briepid}" != "" ]; then
   kill ${briepid} >> /dev/null 2>&1
   #give the server process a few seconds to come down
   RETRY=0
   # Quick escape if the process stopped or else wait some more
   while [ "${briepid}" != "" ] && [ "${RETRY}" -lt "${MAX_RETRY}" ]
   do
   sleep 5
   RETRY=`expr $RETRY + 1`
   briepid=`$PS | egrep "brsvcbrie" | egrep "${SERVER_TO_STOP}" | egrep -v 
egrep | awk '{print $2}'`
   done
   #make sure the server went away
   if [ "${briepid}" != "" ]; then
   # Hard kill any remaining process
   kill -9 ${briepid} >> /dev/null 2>&1
   RETRY=0
   # Quick escape if the process stopped or else wait some more
   while [ "${briepid}" != "" ] && [ "${RETRY}" -lt "${MAX_RETRY}" ]
   do
   sleep 5
   RETRY=`expr $RETRY + 1`
   briepid=`$PS | egrep "\-s ${AR_SERVER_ID}" | egrep "${SERVER_TO_STOP}" | 
egrep -v egrep | awk '{print $2}'`
   done
   fi
   [ "${briepid}" != "" ] && return 1
   fi