NIM thread error in AIX 5.3 server !

UNIX AIX
1

Friends ,

In our production server , we are using oracle10g in IBM AIX 5.3 unix server. From last 7 days , I got the below error :

------------------------------
LABEL: TS_NIM_ERROR_STUCK_
IDENTIFIER: 864D2CE3
Date/Time: Mon Sep 7 19:34:38 NOVST 2009
Sequence Number: 7671
Machine Id: 00028EFFD600
Node Id: AIBLDCDB1
Class: S
Type: PERM
Resource Name: topsvcs
Description
NIM thread blocked
Probable Causes
A thread in a Topology Services Network Interface Module (NIM) process
was blocked
Topology Services NIM process cannot get timely access to CPU
User Causes
Excessive memory consumption is causing high memory contention
Excessive disk I/O is causing high memory contention
Recommended Actions
Examine I/O and memory activity on the system
Reduce load on the system
Tune virtual memory parameters
Call IBM Service if problem persists
Failure Causes
Excessive virtual memory activity prevents NIM from making progress
Excessive disk I/O traffic is interfering with paging I/O
Recommended Actions
Examine I/O and memory activity on the system
Reduce load on the system
Tune virtual memory parameters
Call IBM Service if problem persists
---------------------------------------------------

Also , my paging space goes to 70% . At that moment I need to restart the database or machine ro free the memory . But after few days , the problem raise again .

In this moment , would anybody plz help me , How can I resolve this problem ?

2

Your server is part of an HACMP cluster that uses heartbeats to detect all nodes are alive. The error messages tells you that one of those heartbeats is not working properly because of high (CPU) load. You said that you use 70% of the server's paging space. If this paging space is on rootvg and if rootvg disks are 100% busy during paging space usage that is the likely cause for heartbeat pakets getting lost. So the second relevant part of the error message is the hint about excessive disk I/O which is causing high memory contention

You don't not to worry as long as at least one heartbeat keeps on working. However make sure that the paging space is not being used frequently! You DB has grown over the time and you may need to get more RAM for the (all) cluster nodes