Summary: After a fresh RHEL_Binary deployment, The SRM Webservice Gateway (WSG) service fails to start in SRM VM's.
Detailed Article
Symptoms:
WSG service unable to start/restart in ABE environment.
Services stopped right after fresh installation.
[xxxx bin]# ./manage-modules.sh service status all
* Checking 'topology-mapping-service Default'... [ running ]
* Checking 'webservice-gateway Default'... [ stopped ]
* Checking 'mysql Default'... [ running ]
* Checking 'backend apg1'... [ running ]
* Checking 'backend apg2'... [ running ]
* Checking 'backend apg3'... [ running ]
* Checking 'backend apg4'... [ running ]
* Checking 'collector-manager emc-watch4net-health'... [ running ]
* Checking 'event-processing-manager emc-watch4net-health'... [ running ]
* Checking 'task-scheduler Default'... [ running ]
Error logs (gateway-0-0.log) showed:
SEVERE -- HttpServer::start(): an error occurred starting the server
java.net.BindException: Address already in use
SEVERE -- [2025-12-21 19:13:40 EST] -- Bootstrap::main(): An error occured starting the Webservice Gateway !
com.watch4net.apg.webservice.server.ServerException: an error occured starting the server
Cause:
The error was traced to a port conflict. The WSG service attempted to bind to port 48443, but another process was already using it. This resulted in the java.net.BindException: Address already in use error, preventing the service from starting.
Resolution Steps:
1. Identify the process occupying the port
Run the following command to check which PID is bound to port 48443:
[xxxx logs]# sudo lsof -i :48443
COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
java 19694 apg 49u IPv6 52334 0t0 TCP *:48443 (LISTEN)
java 19694 apg 67u IPv6 645041 0t0 TCP <ABE_FQDN>:48443-><PBE_FQDN>:36744 (ESTABLISHED)
java 19694 apg 69u IPv6 645658 0t0 TCP <ABE_FQDN>:48443-><PBE_FQDN>:36770 (ESTABLISHED)
java 19694 apg 72u IPv6 645880 0t0 TCP <ABE_FQDN>:48443-><PBE_FQDN>:36788 (ESTABLISHED)
java 19694 apg 74u IPv6 647343 0t0 TCP <ABE_FQDN>:48443-><PBE_FQDN>:55076 (ESTABLISHED)
2. Terminate the conflicting process
Kill the identified PID to free up the port:
kill -9 <PID>
3. Restart the services
./manage-modules.sh service restart all
Once the port was released, restarting the services brought them up successfully
Result: After restart all Services are up and running.