Weblogic.Admin fails 2nd time through script loop

Question

Hi
I have a script to shutdown and restart Managed servers in a cluster that the first time through the loop is successful, the second time throws:
Failed to connect to http://lnx241.domain.com:7001: Destination unreachable; nested exception is:
java.net.ProtocolException: Tunneling result unspecified - is the HTTP server at host: 'lnx241.domain.com' and port: '7001' a WebLogic Server?; No available router to destination
Here is the loop part of the script:
ADMIN_URL=t3://lnx241.domain.com:7001
#list servers that you want to shutdown with a space between names.
LIST_SERVERS="absMaximoMngd1 absMaximoMngd2"
echo "***The servers shutdown is forced. If the server won't shutdown you may have to open another SSH session and kill the managed server that is having problems shutting down***"
#Stop the servers. Clear the servers' cache, restart the servers, hopefully one at a time...:)
# the cd command is here incase the script is run through cron
cd /bea/beadev/absMaximo
. /opt/bea/weblogic81/server/bin/setWLSEnv.sh
for server in $LIST_SERVERS
do
     SERVER_NAME=$server
     echo "Stopping Weblogic Server $SERVER_NAME..."     
     java weblogic.Admin -url ${ADMIN_URL} -userconfigfile /bea/beadev/absMaximo/admin/adminKey/AdminConfig -userkeyfile /bea/beadev/absMaximo/admin/adminKey/AdminKeyFile FORCESHUTDOWN ${SERVER_NAME} 2>&1
     sleep 15
     clear_WLcache
     if [ $SERVER_NAME = absMaximoMngd1 ]
     then StartMngd1
     elif [ $SERVER_NAME = absMaximoMngd2 ]
     then StartMngd2
     fi
     sleep 5
done
If anyone has any insight into why this occurs, the help would be appreciated! Thanks!

Answer 1

The lead to this error came though verbose logs; for which the path was mentioned in server logs. After going through the verbose logs, it was found that the content from the source was getting combined with the destination and at this time, the process was failing. This was owing to huge content being stored in the source.
The solution was found by scoping the nodes which are required to be downloaded in source and thus, 'content services' were filtered out. Hence, the commit after the solution didnot try committng the content and hence, was successfull.