Hello everybody,

Today I had a strange problem. Some customers told me that they can't access
 the application which we developed. After investigating the problem I reali
zed that server1 was stopped. I don't know why this happened. I searched the
 stopServer.log to see but
there was no information regarding any stop of server1. Do you have an ideea
 why that happened?

Thank you,
Marius

[ Post a follow-up to this message ]

WAS-ND or WAS-Base flavor ?

For the first, review the Node Agent logs.

For the second, review the SystemOut.log to see if it is a graceful stop or
a crash.

Compare the logs with a normal stop if you have doubt about how to determine
it or post files.

HTH.

[ Post a follow-up to this message ]

It's was base, not ND. I looked into systemout.log and stopserver.log but th
ere was no information there regarding a stop at that our.

The last information I found in systemout.log are the last information gener
ated bye our application and it was no error!

[ Post a follow-up to this message ]

mstoian@softwin.ro wrote:
> It's was base, not ND. I looked into systemout.log and stopserver.log but 
there was no information there regarding a stop at that our.
>
> The last information I found in systemout.log are the last information generated b
ye our application and it was no error!

Did the machine get rebooted? Someone do a kill?
Ken

[ Post a follow-up to this message ]

If you don't see in SystemOut.log that it was a clean stop, then there was a
crash or someone killed the application.

You can also search for core dump files on the file system. Name and
location depends on the Operating System.

The troubleshooting guide will probably prove helpful (see MustGather
documents under Performance / Crash at
http://www-1.ibm.com/support/docvie...uid=swg27005324).

[ Post a follow-up to this message ]

Nobody killed the service, or no human person 

It's a production server at a bank so nobody can access that machine. An hou
r ago it stoped again. The machine was not rebooted, the rest of the service
s were running.

I gived startServer.sh server1 command and then the server started and is wo
rking now.

[ Post a follow-up to this message ]

Until you find the cause, you could make a daemon script to restart
automatically when it fails.

A possible cause of crash is a core crash.

Did you find dumps files on the file system ?

[ Post a follow-up to this message ]

Problem was solved.

I contacted support and they told me to install the latest FP for JRE. I've 
done this and I also modified the value of initial_heap_size to 512 MB and m
aximum_heap_size to 1024 MB and the problem dissapeared.

[ Post a follow-up to this message ]