Problem: 503 Service Unavailable
SAP WebAS Engine is starting...
If this state does not change within a few minutes,
please contact your system administrator.
Check the recommendations in SAP Notes: 943498, 764417
Message: Dispatcher running but no server connected!
Details: No details available
This problem occurs suddenly after a time interval when working on Sap Enterprise Portal. Please Suggest me Number of possible solutions for this type of problem.....
Thank's in Advance
Your server might be rebooting , wait for some time .
In case its not coming up provide log for dev_disp or server etc from work directory.
Possible troubleshooting steps are check Database, status of dispatcher , server and then read the crresponding developer trace.
Without the relevent logs, its all shooting in the dark :-)..
Could you please paste an excerpt from the std_server0.out file from the time of the server crash? The same excerpt from dev_server0 file will also help us.
Also, You could search for the term "Outofmemory" in the std_server*.out file and see if that's whats causing the crash. If so, we need to figure out whats causing this OOM. It could be either perm exhaustion or heap exhaustion. Once you paste the relevant excerpt, we'll be able to tell what caused it.
Sometimes even if the system is configured correctly for the JVM settings, the OOM crash could occur because there is a
memory leak. In that case you need to ensure that your system is generating heapdumps as per notes:
1004255 for Windows
1267126 IBM i: How to get a heapdump which can by analyzed with MAT
1265455 How can you generate a system dump for analysis using MAT? (LNX)
1263258 LXN / 64
1259465 How to get a heapdump which can by analyzed with MAT (AIX) (1053495 heapdump settings on AIX)
Once you have the .dmp or the .hprof file you can analyze it using the MAT Tool which can be downloaded from www.eclipse.org/mat
The generation of leak suspect report shows the exact culprits.
However, first step here is to figure out whats causing the server restarts. So, to figure that please paste the relevant excerpt from the std_server file.
Another reason why the server could crash is due to the jvm crash. If that's the case you'll see hs_err files in case of SUN JVM in cluster\server or cluster\dispatcher directory.
Hope this helps.
SAP / Palo Alto, USA
You have not told us in detail what made you to restart the J2EE Engine, did you change any parameters? If the are semi-deployed SDM files the system won't start.
Anyway the best way to start if you still having the problems is to check the status of your processes.
http://host:5(sys_nr)13 this will open the SAP Management Console and you will be able to see which process are still to start and on double-clicking you can check the logs.
USE: http://mysap.com:50013 as example or on OS Level: /usr/sap/<SID>/JC00/work
*dev_ms > check if your license is still valid, this one of the common issues
*dev_dispatcher > this will detail one or 2 possible errors
*dev_jcontrol (depending on your version of SA NW EP)
All of the above are very important to the running of your system.
As per the symptom, I am supposing your java system is getting OutOfMemory, and with this OutOfMemory, the serverX
process restarts automatically from time to time.
To resolve this, it will be necessary to do a details anslysis of genereted Heapdump with the occurance of OutOfMemory error.
Hence I would like to suggest you report a customer message to SAP support, as the heapdump is quite large and can not be analyzed here.
That is not the problem, when the server node is in waiting status this message will be appear when you use the URL, just refresh your server node status at OS level by using JCmon and check out the status of the server node. If any error occurs in server node please check the server_std.out file at /usr/sap/<SID>/<instance name>/work directory.