on 05-02-2014 8:15 AM
Hi,
We are using SAP Netweaver Portal 7.3.
We are getting the "cannot connect to provider " warning in frontend of portal while user is trying to view his/her inbox. though this issue is intermittent we have the following logs in trace files :
1. Cannot clear ICM server cache by [718ac393010411e3a1fa000000d23c1e] etagIP address.
2. Empty or null time zone value from Backend.
3. Attempting to create outgoing ssl connection without trusted certificates.
But we arent sure wether the warning message is due to the above mentioned logs.
Can any one help us regarding this?
Is the 3 mentioned logs the reason for getting the warning "cannot connect to provider "?
I have also attached the log file screenshot.
Thanks and regards
Arghya
Hi Arghya!
I'm sending here the same update which I've sent to you in the other thread.
I'm afraid that the trace messages which you are mentioning are not related to the actual problem that you're having in the worklist.
I would advise you to use the portal Log Watch tool according to the following KBA:
1593107 - Logwatch Tool - Collecting Portal DefaultTrace logs
You should select com.sap.netweaver.bc.uwl as the trace location to be recored and ALL as the trace severity. After hitting Record then recreate the issue, checking the logs afterwards.
You should see an error from UWL, when connecting to one or more of the connectors which you have defined.
Keep us posted on the outcome.
Regards,
Armando Zaro
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
HI Armando,
Thank you for your suggestion. i will let you know the about the outcome.
I also need a few more information,
As this issue is intermittent, it is difficult for us to reproduce the scenario,hence if we change the severity of the UWL logs, will it have a severe effect on the servers?
Can you please suggest?
thanks
Arghya
Hi Arghya!
The best would be if you could somehow isolate a scenario where the issue can be reproduced on demand, and then use the Log Watch tool which I've mentioned to you as it only raises the priority for the time when you're collecting the traces.
Regarding activating in a server basis, we should know that the trace severity which you would configure would be effective then for all the users up to when you deactivate it. Meaning that we should use it with caution. It might be the case that it doesn't cause any problem but first maybe we should try other options.
Before raising the traces, you could try colleting the correct default trace excerpts without having the need of raising the UWL trace severity. Please use the following KBA to help you into this:
1596214 - How to find the latest default trace file of the NW AS Java right after reproducing an iss...
Once with the correct file, you should be able to check it and see the errors coming from com.sap.netweaver.bc.uwl
Eventually we'll get some clue about the root cause in this case!
Looking forward to checking your updates.
Regards,
Armando
Hi Armando..
We checked using log config tool and got the below logs :
Jun 13, 2014 3:00:36 PM com.sap.netweaver.bc.uwl.core.connect.bpem.subst.SubstitutionRequestGenerator.handleSubstitutionRequest [Thread[UWL Pooled Thread:1,5,SystemThreadGroup]] Error: Cannot connect to provider system SAP_LocalSystem
[EXCEPTION]
java.io.IOException: Cannot create connection.
at com.sap.netweaver.bc.uwl.core.connect.bpem.subst.SubstitutionRequestGenerator.getConnection(SubstitutionRequestGenerator.java:50)
at com.sap.netweaver.bc.uwl.core.connect.bpem.subst.SubstitutionRequestGenerator.handleSubstitutionRequest(SubstitutionRequestGenerator.java:90)
at com.sap.netweaver.bc.uwl.core.connect.bpem.subst.SubstitutionRequestGenerator.getSubstitutions(SubstitutionRequestGenerator.java:36)
at com.sap.netweaver.bc.uwl.core.connect.bpem.subst.BPEMSubstitutionManager.getAllSubstitutions(BPEMSubstitutionManager.java:55)
at com.sap.netweaver.bc.uwl.core.SubstitutionCheckThread$1.run(SubstitutionCheckThread.java:51)
at com.sap.netweaver.bc.uwl.core.SubstitutionCheckThread$1.run(SubstitutionCheckThread.java:40)
at java.security.AccessController.doPrivileged(Native Method)
at javax.security.auth.Subject.doAs(Subject.java:337)
at com.sap.netweaver.bc.uwl.core.SubstitutionCheckThread.run(SubstitutionCheckThread.java:40)
at com.sap.netweaver.bc.uwl.core.ThreadManager$PoolWorker.run(ThreadManager.java:209)
Could you please help us on this?
Regards,
Arghya
Hi Arghya
Just to ensure, is the BPM running in the same system as the UWL one? Or the BPM system is remote to the UWL one?
I know that according to you the problem is intermittent, but I was wondering if perhaps the problem was with the configuration of the BPM connector within UWL.
Looking at the exception (java.io.IOException), do you have anything in front of the portal system such as a load balancer or reverse proxy? It seems that there is something which is interrupting the communcation between UWL and the BPM servlet.
Kind regards,
Armando Zaro
Hi,
I just wanted to elaborate on this.
Consider a UWL configuration where you have defined 3 webflow connectors:
SAP_ECC
SAP_SRM
SAP_CRM
Now consider a user who logs in using his portal or AD userid. This gentleman has ID in ECC system but not in say SRM.
So the connector with SRM will fail and display this warning message.
This is perfectly OK and natural as long as user is able to see the workitems for which he is authorized.
We had an issue sometime back. while user tries to logoff, the DSM terminator was trying to log the user off from SRM system as well, by virtue of this UWL.
however since no ID exist in SRM for this user, it was publishing a lot of exceptions on the portal logs.
SAP suggested to : Patch1 for UWL COLL PROCESS ENGINE 7.31
Please analyze and take a call on this as well.
Regards,
Ankur
Hi Ankur
To solve the concern that you're bringing, you could consider configuring User Roles in a connector level.
This is also one of the guidelines given in the following KBA:
1577547- UWL Performance Tips and Considerations
By going through this way the UWL wouldn't even execute a call to the back end system in case the user does not have the appropriate role assigned in the portal.
Please configure this as it will help you to adress the problem.
Regards,
Armando
User | Count |
---|---|
90 | |
10 | |
10 | |
10 | |
7 | |
7 | |
6 | |
5 | |
4 | |
3 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.