Application server 8.1 going down frequently

Application server is going down very frequently. It doesn't log anything in server.log before going down but in imq logs it logs some broker shutdown details.

imq_broker_log:

=============================================================================== =

Sun Java(tm) System Message Queue 3 2005Q4

Sun Microsystems, Inc.

Version: 3.6 SP3 (Build 02-A)

Compile: Wed Jun 22 12:36:00 PDT 2005

Copyright ?2005 Sun Microsystems, Inc. All rights reserved.

Use is subject to license terms.

This product includes code licensed from RSA Data Security.

=============================================================================== =

Java Runtime: 1.5.0_01 Sun Microsystems Inc. /usr/jdk/instances/jdk1.5.0/jre

[02/Jun/2006:22:22:10 IST] License: Sun Java(tm) System Message Queue 3.6 Enterprise Editi

on

[02/Jun/2006:22:22:10 IST]IMQ_HOME=/

[02/Jun/2006:22:22:10 IST] IMQ_VARHOME=/var/opt/SUNWappserver/domains/domain1/imq

[02/Jun/2006:22:22:10 IST] SunOS 5.10 sparc sol1 (1 cpu) root

[02/Jun/2006:22:22:10 IST] Max file descriptors: 65536 (65536)

[02/Jun/2006:22:22:10 IST] Java Heap Size: max=194688k, current=32448k

[02/Jun/2006:22:22:10 IST] Arguments: -javahome /usr/jdk/entsys-j2se -varhome /var/opt/SUN

Wappserver/domains/domain1/imq -name imqbroker -port 7676 -bgnd -silent

[02/Jun/2006:22:22:10 IST] [B1004]: Starting the portmapper service using tcp [ 7676, 50,

* ] with min threads 1 and max threads of 1

[02/Jun/2006:22:22:10 IST] [B1060]: Loading persistent data...

[02/Jun/2006:22:22:10 IST] Using built-in file-based persistent store: /var/opt/SUNWappser

ver/domains/domain1/imq/instances/imqbroker/

[02/Jun/2006:22:22:11 IST] [B1041]: Cluster initialization successful.

[02/Jun/2006:22:22:11 IST] [B1136]: Processing stored transactions

[02/Jun/2006:22:22:11 IST] [B1013]: Auto Creation of Queues is Enabled

[02/Jun/2006:22:22:11 IST] [B1151]: Loading destination mq.sys.dmq [Queue] with 0 messages

[02/Jun/2006:22:22:11 IST] [B1152]: Loading of destination mq.sys.dmq [Queue] complete

[02/Jun/2006:22:22:12 IST] [B1004]: Starting the admin service using tcp(host = *, port=0,

mode=dedicated) with min threads 4 and max threads of 10

[02/Jun/2006:22:22:12 IST] [B1004]: Starting the jms service using tcp(host = *, port=0, m

ode=dedicated) with min threads 10 and max threads of 1000

[02/Jun/2006:22:22:12 IST] [B1004]: Starting the cluster service using tcp [ 0 ] with min

threads 1 and max threads of 1

[02/Jun/2006:22:22:12 IST] [B1039]: Broker "imqbroker@sol1:7676" ready.

[02/Jun/2006:22:22:49 IST] [B1065]: Accepting: admin@127.0.0.1:33051->admin:33040. Count=1

[02/Jun/2006:22:22:49 IST] [B1066]:Closing: admin@127.0.0.1:33051->admin:33040 because

"[B0059]: Client closed the connection". Count=0

/////////////////////////

AT THIS POINT IT STAYS FOR SOME TIME THEN START SHUTTING DOWN THE BROKER AND SUBSIQUENTLY APPLICATION SERVER

//////////////////////////

[02/Jun/2006:22:25:40 IST] [B1047]: Shutting down broker...

[02/Jun/2006:22:25:40 IST] [B1007]: Stopping Service admin with protocol tcp(host = *, por

t=0, mode=dedicated)

[02/Jun/2006:22:25:40 IST] [B1007]: Stopping Service jms with protocol tcp(host = *, port=

0, mode=dedicated)

[02/Jun/2006:22:25:40 IST] [B1077]: Broadcast good-bye to all connections ...

[02/Jun/2006:22:25:40 IST] [B1078]: Flushing good-bye messages ...

[02/Jun/2006:22:25:40 IST] [B1063]: Done

[02/Jun/2006:22:25:40 IST] [B1048]: Shutdown of broker complete.

How we can fix this

[3761 byte] By [Prashant.Bhatnagar] at [2007-11-26 7:42:20]
# 1
OS is solaris 10
PrashantBhatnagar at 2007-7-6 19:50:22 > top of Java-index,Application & Integration Servers,Application Servers...
# 2

hi,

am facing now exactly the same poblem and i couldnt figure it out yet but my situation might help spotting your problem.

am accessing my solaris 10 server using an xwindow manager and when ever i close the session my server stops being accessible and checking its status again i found that the imqbroke is stopped for the same reason which is client closed the connection . in my case this is true as i log off so u really need to check whats causing the connection to close.

Best regards

Insighter at 2007-7-6 19:50:22 > top of Java-index,Application & Integration Servers,Application Servers...
# 3
Is there a core file generated in the config folder of the domain, if this is Solaris/Linux?Is there a JVM crash log (hs_...log) in the user home directory or domain's config folder?Thanks,Kedar
kedarmhaswade at 2007-7-6 19:50:22 > top of Java-index,Application & Integration Servers,Application Servers...
# 4
We are suffering through exactly the same problem,OS Solaris 10 .will appreciate for any help or clue.Many Thanks.
Sp00ky_Geek at 2007-7-6 19:50:22 > top of Java-index,Application & Integration Servers,Application Servers...
# 5
i also observer that when i close the telnet session from which i have started the server, the app server shuts down.so i started the sever with nohup and its running for last 22 hours :)
PrashantBhatnagar at 2007-7-6 19:50:22 > top of Java-index,Application & Integration Servers,Application Servers...
# 6

> i also observer that when i close the telnet session

> from which i have started the server, the app server

> shuts down.

>

> so i started the sever with nohup and its running for

> last 22 hours :)

Now I think I know. Are you hitting:

http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6356910

?

Are you using bourne shell by any chance? Can you consider using ksh?

Kedar

kedarmhaswade at 2007-7-6 19:50:22 > top of Java-index,Application & Integration Servers,Application Servers...