login problem with outlook integration

I'm using the Advanced Outlook Integration component (FileSite) with 4.1SP1 and I'm getting an error logging in. The FileSite User Guide doesn't seem to have any information about troubleshooting. The error I'm getting is:

iManMPExt: WorkSite MP encountered an error while performing an operation

[CLoginCmd][Execute] WorkSite MP encountered an error while performing an operation

Stack Trace
java.lang.RuntimeException: worksitemp
at com/imanage/cms/fc/net/fma/FmaClientReplyService.findService
at com/imanage/cms/fc/net/fma/FmaClientReplyService.findServiceInCluster
at com/imanage/cms/fc/net/fma/FmaClient.findServiceOnServer
at com/imanage/cms/fc/net/fma/FmaClient.findServiceOnCluster
at com/imanage/cms/fc/net/fma/FmaClient.findService
at com/imanage/cms/fc/cmsifc/CMSClientInterfaceBase.openConnection
at com/imanage/cms/fc/cmsifc/CMSClientInterfaceBase.ensureConnection
at com/imanage/cms/fc/cmsifc/CMSClientInterfaceBase.executeMessage
at com/imanage/cms/fc/cmsifc/CMSClientInterfaceGenerated.getLibraryList
at com/imanage/cms/t2/jom/impl/SessionImpl.getPOELibraryName
at com/imanage/cms/api/workteam/impl/SimpleSessionImpl.logon
at com/imanage/cms/api/workteam/impl/SimpleSessionImpl.logon2
at com/imanage/cms/com/api/workteam/SimpleSession.logon2

I am not using trusted logon. The same user account can logon to the web interface with no problems

Find more posts tagged with

Comments

dbguy

It seems you are pointing the Outlook client to the wrong machine/cluster. You should use the MP Server cluster/machine name to connect through Outlook. A common mistake is trying to connect to the machine where the web app is running.

JTNeville

Can't mix major service releases either. I.e. 4.0 and 4.1 CMS are differnet enough a client for one or the other can fail to connect. While I haven't tested the Outlook plugin, this was the case recently for the webapp thread here. Might not apply to you but figured I'd throw it in the ring just in case.

untitled.JPG

Migrateduser

This is a test environment where the entire cluster is one server. I am currently specifying the IP address of the server under servername (which, as you might guess, isn't the same as the clustername). I'll go ahead and harass the infrastructure guys into setting me a DNS entry for the cluster and see if that makes a difference.

Thanks
Rob

Migrateduser

Adding a DNS entry for the cluster and doing it by name rather than IP got rid of the error message. However, none of the functionality is available. All of the menu items on the WorkSite MP menu are grayed out except options and it doesn't do anything when I click it. If I try to open the WorkSite MP folder in the folder list, I get an error message saying "The set of folders could not be opened".

I hoped to find something helpful in the logfiles, but the only thing that was even semi-helpful was this from the process manager log:

Tue May 03 09:55:50.141: INFO [FmaService] Sending removeClient message FMAClientReplyService at: /192.168.6.224:4762
Tue May 03 09:55:50.141: INFO [FmaService] Sending removeClient message FMAClientReplyService at: /192.168.6.224:4762
Tue May 03 09:55:50.141: INFO [FmaService] Sending removeClient message FMAClientReplyService at: /192.168.6.224:4762

I also have this from the Library Manager log:

Mon May 02 12:36:06.404: INFO [ImThreadMonitor-1] Thread FmaClientReplyService may be hung, it has been waiting for 1163 seconds for something to do
Mon May 02 12:36:21.596: INFO [ImThreadMonitor-1] Thread FmaClientReplyService may be hung, it has been waiting for 1178 seconds for something to do
Mon May 02 12:36:36.607: INFO [ImThreadMonitor-1] Thread FmaClientReplyService may be hung, it has been waiting for 1193 seconds for something to do

So, I suppose some of the questions here are what does the FMAClientReplyService actually do, where does it sit in the architecture, and is there any way I can bump up the verbosity of its log messages?

Migrateduser

Yes, the clustername used by client should exactly be the clustername configured in the config manager of CMS. In a test environment you can skip the DNS entry step by using the IP address as the clustername in CMS configuration. Then the clients can connect using IP address.

About the options being grayed out:

It happens to me when I do not select the WorkSite MP node in the outlook leftside tree. Even the 'Login' menu would be available only when I select the WorkSite MP node. This may not be the case for you though.

-bijoy

Migrateduser

I did a complete reinstall to start from scratch. I now have the clustername set to 'worksitemp' which is what the DNS entry points to. I'm still not getting a login, but I'm getting a better error message now.

Now, when I try to login I get this error:


iManMPExt:  A WorkSite MP resource required for the requested operation is currently unavailable.



[CLoginCmd][Execute] A WorkSite MP resource required for the requested operation is currently unavailable.

Stack Trace
 Failed to execute message: CmsGetDictionary
 Return code=902
 Error msg=Message for code 902 not found.
 Library=RICHMOND
 Server=WORKSITEMP

com.imanage.cms.exceptions.database.DatabaseUnavailableException:  library 'RICHMOND' not found on cluster worksitemp
 at com.imanage.cms.t1.service.CmsMsgProcessor.initialize(CmsMsgProcessor.java:117)
 at com.imanage.cms.t1.service.CmsMsgProcessor.<init>(CmsMsgProcessor.java:44)
 at com.imanage.cms.t1.service.CmsServiceBeanContainer.processMessage(CmsServiceBeanContainer.java:35)
 at com.imanage.cms.t1.net.rpc.RpcSession.processMessage(RpcSession.java:546)
 at com.imanage.cms.t1.net.rpc.RpcSession$RpcSessionThread.run(RpcSession.java:460)
 at com.imanage.cms.fc.base.ImThread.callRun(ImThread.java:808)
 at com.imanage.cms.fc.base.ImThread.executeOneLoop(ImThread.java:931)
 at com.imanage.cms.fc.thread.ThreadHWRunner.run(ThreadHWRunner.java:141)
 at java.lang.Thread.run(Unknown Source)
 at com/imanage/cms/fc/cmsifc/CMSClientInterfaceBase.executeMessage
 at com/imanage/cms/fc/cmsifc/CMSClientInterfaceBase.getDictionary
 at com/imanage/cms/t2/jom/impl/LibraryImpl.<init>
 at com/imanage/cms/t2/jom/impl/SimpleSessionImpl.getLibrary
 at com/imanage/cms/t2/jom/impl/SimpleSessionImpl.getLibrary
 at com/imanage/cms/api/workteam/impl/SimpleSessionImpl.getLibraryByName
 at com/imanage/cms/api/workteam/impl/ObjectIdImpl.getLibraryById
 at com/imanage/cms/api/workteam/impl/SessionImpl.getLibraryById
 at com/imanage/cms/api/workteam/impl/UserImpl.getPreferencesFacility
 at com/imanage/cms/com/api/workteam/User.getPreferencesFacility

I think I actually understand this error. It's looking for an authentication library that doesn't exist. In the previous installation, I had a library named Richmond, which was being used as the authentication library for the cluster. When I reinstalled, I added one library, called Corporate (see attached screenshot), and set it as the authentication library. I used the same database as before, though, which I am wondering may be the source of the issue.

Any thoughts on how to resolve this?

dbguy

It seems you are using your old RICHMOND database as CORPORATE in the new setup. If so, the easiest solution is to call it RICHMOND in the new setup. Once a database is created, it cannot be simply renamed or reused under another name in Configuration Manager.

If you really want to call it CORPORATE, call support and they can help you with the renaming process.