Websphere 6.1 - Configuring Security

Question

Hi

When i try to configure security through the admin console of Websphere it just hangs. Its at the last step of the below 4 steps Step 1: Specify extent of protection
Step 2: Select user repository
Step 3: Configure user repository
Step 4: Summary

Here are the extracts from my console [26/09/08 13:50:56:539 IST] 0000001f ServletWrappe I SRVE0242I: [isclite] [/ibm/console] [/com.ibm.ws.console.security/EnableSecurity.jsp]: Initialization successful. [26/09/08 13:50:58:616 IST] 0000001f ServletWrappe I SRVE0242I: [isclite] [/ibm/console] [/com.ibm.ws.console.security/SelectRegistry.jsp]: Initialization successful. [26/09/08 13:51:00:126 IST] 0000001f ServletWrappe I SRVE0242I: [isclite] [/ibm/console] [/com.ibm.ws.console.security/ConfigureRegistry.jsp]: Initialization successful. [26/09/08 13:51:00:126 IST] 0000001f ServletWrappe I SRVE0242I: [isclite] [/ibm/console] [/com.ibm.ws.console.security/LocalRegistry.jsp]: Initialization successful. [26/09/08 13:51:36:202 IST] 0000001f ServletWrappe I SRVE0242I: [isclite] [/ibm/console] [/com.ibm.ws.console.security/ConfirmEnableSecurity.jsp]: Initialization successful. [26/09/08 13:52:20:255 IST] 0000001f UserRegistryI A SECJ0136I: Custom Registry:com.ibm.ws.security.registry.nt.NTLocalDomainRegistryImpl has been initialized [26/09/08 13:52:21:025 IST] 0000001f UserRegistryI A SECJ0136I: Custom Registry:com.ibm.ws.security.registry.nt.NTLocalDomainRegistryImpl has been initialized [26/09/08 14:04:03:127 IST] 00000019 ThreadMonitor W WSVR0605W: Thread "WebContainer : 2" (0000001f) has been active for 746076 milliseconds and may be hung. There is/are 1 thread(s) in total in the server that may be hung.

Any idea what could be up here?

Thanks Damien

Answer 1

It sounds to me like your LDAP server is not responding in a timely manor. The fact that you have a hung thread rather than a network error indicates to me that you are successfully communicating to LDAP. If you are attaching this to the Active Directory backed LDAP, it could be overtaxed. We have used ADAM servers in the past to get around slow LDAP response time. However our responce times were around 3-5 seconds instead of 12+ minutes. This could also be a deadlock. I have seen hung threads as a result of implementing XD extensions where the nodes deadlocked their communication (But these were not WebContainer threads). To fix this issue, we ditched XD extensions in favor of a NetScaler setup.