ESM login process times out after upgrading to SLM

  • 7007007
  • 05-Oct-2010
  • 30-Apr-2012


Novell Sentinel Log Manager 1.1
SUSE Linux Enterprise Server 11


The Linux server's primary IP address is configured on a bonded interface (ie. Bond0) and is not set with the typical "ethx" naming.

After upgrading to Novell Sentinel Log Manager 1.1, one is not able to log into the Event Source Management tool, and receives the message: "The login process timed out. The most likely causes are: 1) Sentinel Server is not fully initialized: Try again later. 2) Sentinel Server is not running: Start Sentinel Server. 3) Misconfiguration: Check that your configuration is set to the correct Sentinel Server."

The contents of the ESMWebStart.jnlp file show that ESM is attempting to use an alternate IP address of the server, and not the primary bonded IP address.


This appears to be happening because Tomcat is not able recognize the bonded interface and assign the proper IP address.

To work around the problem:
1. Locate the script in "/opt/novell/sentinel_log_mgr/bin/".
2. Create a backup of the file.
3. Locate the existing IF statement that assigns the IP address to the SERVER_IP variable, and comment it out.
4. Manually assign the server's bonded IP address to the SERVER_IP variable.
5. Restart the sentinel_log_mgr service.

Example for manually assigning the bonded IP address:
#SERVER_IP=$(/sbin/ifconfig | grep eth -A 2 | grep 'inet' | grep -v 'inet6' | grep -v '' | head -1 | cut -d: -f2 |
 awk '{print $1}')
#if [ -z "$SERVER_IP"]
#       SERVER_IP=$(/sbin/ifconfig | grep 'inet' | grep -v 'inet6' | grep -v '' | head -1 | cut -d: -f2 | awk '{pr
int $1}')

The issue has been reported to Novell engineering, and should be fixed in a future release of Novell Sentinel Log Manager.