locked
Response Group Configuration Tool Failure - Unknown Error Occurred RRS feed

  • Question

  • Hello,

    I'm currently working with a customer which has some issues regarding OCS RGS. They want to implement RGS, but halted at the stage of configuring the workflows.

    In their lab environment, after applying the well-known fixes to get the Response Group Service website running (setSPN, member of RTCUniversalServerAdmins, kerberos authentication), the Workflow-site works perfectly, showing 0 workflows and the buttons for creating the workflow(s).

    However, in their live environment I keep getting the following error when opening the website. It is showing the blue RGS-logo bar at top, but where the groups/buttons should appear a red cross and the following tekst:

    Response Group Configuration Tool Failure
      An unknown error occurred. The operation cannot complete successfully. Please contact the administrator if this problem persists.
      Click here to return Home.

    Response Group Configuration Tool Failure

    An unknown error occurred. The operation cannot complete successfully. Please contact the administrator if this problem persists.

    Click here to return Home.

    On the Front-End server, I get the following errors in the event viewer;

    In the application log:

    Event code: 3005 
    Event message: An unhandled exception has occurred. 
    Event time: 23-8-2010 13:26:52 
    Event time (UTC): 23-8-2010 11:26:52 
    Event ID: 9fd6239f34234927bbb887c702f04197 
    Event sequence: 9 
    Event occurrence: 2 
    Event detail code: 0 
     
    Application information: 
      Application domain: /LM/W3SVC/1/ROOT/Rgs-2-129270362192244711 
      Trust level: Full 
      Application Virtual Path: /Rgs 
      Application Path: C:\Program Files\Microsoft Office Communications Server 2007 R2\Web Components\Acd Files\ 
      Machine name: OCS64 
     
    Process information: 
      Process ID: 748 
      Process name: w3wp.exe 
      Account name: <domain>\RTCComponentService 
     
    Exception information: 
      Exception type: NullReferenceException 
      Exception message: Object reference not set to an instance of an object. 
     
    Request information: 
      Request URL: https://<Pool FQDN>:443/Rgs/Deploy/Default.aspx 
      Request path: /Rgs/Deploy/Default.aspx 
      User host address: 10.100.99.1 
      User: <domain>\<Member of RTC Universal Server Admins> 
      Is authenticated: True 
      Authentication Type: Negotiate 
      Thread account name: <domain>\RTCComponentService 
     
    Thread information: 
      Thread ID: 7 
      Thread account name: <domain>\RTCComponentService 
      Is impersonating: False 
      Stack trace:  at Microsoft.Rtc.Acd.Web.AcdUserContext.LoadFromHttpContext(HttpContext context)
      at Microsoft.Rtc.Acd.Web.AcdUserContext.FromHttpContext(HttpContext context)
      at Microsoft.Rtc.Acd.Web.UI.AcdUserView.CreateChildControls()
      at System.Web.UI.Control.EnsureChildControls()
      at System.Web.UI.Control.PreRenderRecursiveInternal()
      at System.Web.UI.Control.PreRenderRecursiveInternal()
      at System.Web.UI.Control.PreRenderRecursiveInternal()
      at System.Web.UI.Control.PreRenderRecursiveInternal()
      at System.Web.UI.Control.PreRenderRecursiveInternal()
      at System.Web.UI.Page.ProcessRequestMain(Boolean includeStagesBeforeAsyncPoint, Boolean includeStagesAfterAsyncPoint)
     
     
    Custom event details: 

    In the Office Communications Server log:

    Unhandled exception occurred in the Response Group Service Configuration Tool. The operation could not complete successfully.
    
    Requested URL: /Rgs/Deploy/default.aspx
    User / Host making the Request: <domain>\<Member of RTC Universal Server Admins> / <IP>
    
    Unhandled exception: System.Web.HttpUnhandledException - Exception of type 'System.Web.HttpUnhandledException' was thrown.
    Inner Exception: System.NullReferenceException - Object reference not set to an instance of an object.
    
    Cause: An unhandled exception occurred.
    Resolution:
    Check the exception.

    It is running on an enterprise pool (2007 R2), single front-end server with separate SQL server, mediation server and proxy server (RCC). Web components server etc. is all configured on the Frond End server.

    So far I have troubleshooted a lot, and fixed a few settings:

    - errors of RGS: RGS was not able to register to pool by MTLS. certificate was not containing an entry for the FE server
    - "known fixes" as SetSPN were not done
    - server was not accepting browsing by using DNS alias for the poolname (security setting in registry)
    - Response Group Service service was running with the domain administrator account instead of the RTCComponentService
    - LSGroupExpAppPool was running in integrated mode with administrator account. Changed this back to classic and the RTCComponentService

    After these fixes, the check tool from the reskit as also the validation were indicating everything was 100% operative and successfull. but the workload website, not working.

    Then I tried some other things, and also found some flaws and errors in the IIS configuration. Not the correct app.pool and username, for example. Compared these to the IIS in the lab environment (where the website is running properly) In the end, decided to re-configure the complete web components by removing RGS and the component services, removing the component services directory, and re-installing the services again. Activating the RGS app, and re-run the EE setup for final activation.
    However, still getting the same error.

    Tried to track the OCS log, but this gives me the same errors as I see in the event logs. The IIS log shows me it succesfully opens the error ASPX, but nothing why.

    To me, it seems that some ASP or IIS issue may be, as OCS tells me everything is up and running. But of course, without a workflow it won't work properly :-) I didn't build the environment myself, and it looks like some errors occured or things were 'forgotten' during the implementation. Also some troubleshooting by the customer might have had this situation as result. However, I can't get to the cause, the customer didn't track the installation and the troubleshooting.

    I'm about out of options, and wanted to consult you before deciding to advise to completely re-build the OCS environment. The pool is an existing pool where hundreds of users already use the IM- and Presence-functionalities and some use Remote Call Control over several locations; it's not an environment which isn't in use already, so re-building the OCS environment will have a lot of impact.

    Please let me know if any of you ever had the same situation and know the solution, and/or knows anything else I still can try. Thanks in advance.

    Monday, August 23, 2010 12:19 PM

All replies

  • I don't think I'm much help to you - but here's my comments...

    Can you install the RGS web components on another server to test?  If so, you can prove if it's an issue just on this server.  I don't know if you can do this - I've never tested.

     

    I had some minor issues with RGS - but all my issues were fixed by installing all the updates.

    I have had to re-install my live OCS R2 setup from scratch but I found I could uninstall the software, remove the servers from the domain (Front end and Med), rebuild the servers from scratch, name them the same and rejoin them back to the domain - and AD would have all the user details already in-place, the dial plan, etc.

    Saved a TON of work. I 'think' the entire process took approx 6 hours to get the servers reinstalled, running and re-patched.

    Not something you want to rely on - but when the poo hits the fan...

     

    regards

    Paul Adams

     

    Wednesday, August 25, 2010 7:29 PM
  • Dear Paul,

    Thanks for your comment.  Yes installing the role on another server was indeed one of my considerations. however, I haven't tested it - I would have had to request a new server, built it there, and even if it worked - the problem is not resolved on the "real" server. We would still have to solve it, or re-install the complete server (or replace it) in that case.

    I found (a lot) of other bugs and issues on that environment, which I all was able to resolve. However, I doubt the stability of it - who knows what else is wrong. Besides, it contains virtual elements, with EV thats not supported.

    Therefore I have suggested to completely re-view and re-build the environment (the correct way and documented), if possible in parallel to the existing pool. On physical servers, etc.

    They agreed to that plan, so I'm working on a design now. However, I'm still curious about this particular problem, and what the hack is wrong...

    Thursday, August 26, 2010 12:13 PM
  •  

    Hi Louis,


    While searching for the same issue, I came across your post. I have had this problem once but it got resolved simply by resetting 'iis' (without force).


    However, today again I am facing the same issue with exactly the same events in Application Logs and iisreset does not help as well. So if you figured a way to resolve this error, please do post so.


    Shall be grateful.

    Thank you.

    • Proposed as answer by Sanaullah Mir Tuesday, January 4, 2011 4:11 AM
    Tuesday, January 4, 2011 4:03 AM
  •  

     

    'iisreset /noforce' could not do it this time but yes 'iisreset /force' did it. :)

    My RGS Configuration Tool Page is back... 

    • Proposed as answer by Sanaullah Mir Tuesday, January 4, 2011 4:11 AM
    Tuesday, January 4, 2011 4:10 AM