Unity integrated with Exchange 2010 - Users receive "Your messages are not available"

Document

Sep 2, 2010 10:19 AM
Sep 2nd, 2010

Updated October 26th, 2010

This is a new issue being investigated by Cisco and Microsoft, please check back for additional details as they become available.

A Cisco defect, CSCti70702, has been created to track this, but the problem has been found to be with the Microsoft CDO software that Unity leverages to connect to Exchange. The Microsoft KB article tracking this issue is KB 2405611. Microsoft released a patch on 10/25 that should resolve this. They are still in the process of updating their documentation, but the patch is available for download at the following URL:

/* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman","serif";} http://www.microsoft.com/downloads/en/details.aspx?FamilyID=94274318-27c4-4d8d-9bc5-3e6484286b1f&displaylang=en

Please see details below under the Solution heading for upgrade and verification procedures.

Products Affected:

Cisco Unity 5.0(1), 7.0(2), and 8.0(3) integrated with one or more Microsoft Exchange 2010 servers


Problem Description:

Users with mailboxes on an Exchange 2010 server that dial in to Unity to check their messages may intermittently hear the prompt "Your messages are not available". Upon calling back in, the user is then able to listen to their messages. Please note that the nature of this problem makes it intermittent. If a user receives this message everytime they call in, it is unlikely they are encountering this issue.


The windows application event log on the Unity server will contain errors similar to the following:

Event Type:    Error
Event Source:    CiscoUnity_MALEx
Event Category:    Error
Event ID:    30020
Date:        8/26/2010
Time:        9:07:35 AM
User:        N/A
Computer:    UNITY
Description:
An attempt to create a search-result folder for account cn=jdoe,cn=Recipients,ou=First Administrative Group,o=company has failed.  The MAPI subsystem returned the following error: 80040115.


This is typically an indication of configuration issues with Unity, Exchange, or the MAPI subsystem.


Additionally, the similar WAV errors to the one below may be seen interspersed throughout the application log:


Event Type:    Error
Event Source:    CiscoUnity_Wav
Event Category:    Error
Event ID:    801
Date:        8/20/2010
Time:        9:43:03 AM
User:        N/A
Computer:    UNITY
Description:
Cisco Unity's multi-media component has encountered a serious error.


EXPLANATION:
A serious failure has occurred on port 4 during wav media processing.  Depending upon the
severity of the failure, parties on this call could experience unexpected behavior or be
disconnected.  In some cases, further calls on this port will not be handled correctly.



TECHNICAL DETAILS:
Thread 0x00001420 had a Failure on Port 4 in AvWav


DESCRIPTION:
File: E:\views\CU7.0.1.28\un_Miu\UnityAvWav\WAV.cpp(4718)
Method: WavReadFormatPlay



Failure: call to WavRead failed.

The specific errors may vary slightly, but most will include the error return code "80040115". A packet capture taken at the time that the end user symptoms are observed can also be used to verify that this specific issue is being encountered. If a capture filter must be applied, ensure that it includes all traffic between the active Unity server and all CAS servers. Use one of the following view filters, depending on what application you are using:

Wireshark: dcerpc.pkt_type == 19 or dcerpc.cn_status == 0x1c00001a

Netmon: MSRPC.Orphaned || MSRPC.Fault.Status == 0x1c00001a

If you observe the end user symptoms (callers intermittently hear "Your messages are not available now"), the event log entries, and the above packet capture filters identify suspect traffic, it is save to assume you are encountering this issue. Matching a single symptom is not sufficient evidence to conclude this issue is occurring.


Solution:

A problem has been identified with the Microsoft CDO library that Unity uses to communicate with Exchange 2010. This is documented in Microsoft KB article 2405611:

http://support.microsoft.com/kb/2405611

A patch has been provided by Microsoft, it is available at the following URL:

/* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0in 5.4pt 0in 5.4pt; mso-para-margin:0in; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman","serif";} http://www.microsoft.com/downloads/en/details.aspx?FamilyID=94274318-27c4-4d8d-9bc5-3e6484286b1f&displaylang=en

This is an updated version of the Microsoft Exchange CDO tools that Unity leverages to communicate with Exchange. The patch only needs to be applied to your Unity server(s). To verify that the patch has been applied, please download the latest version of the Gather Unity System Info tool and verify that the output under "Environment Information" is similar to the following (note that older versions of GUSI may not properly detect the version):

MAPI Provider : CDO Version - 6.5.8190.0
         TCPChimney : Disabled
         RSS : Disabled

If the version is anything prior to 6.5.8190.0, you are running a version that is vulnerable to this condition. To apply the upgrade:

1. Right click on Unity icon in task bar and select "Stop Unity"
2. Go to the Add/Remove Programs Control Panel and uninstall the current version of CDO ("Messaging API and Collaboration Data Objects 1.2.1")
3. Run ExchangeMapiCdo.msi from the patch to apply the updated version.
4. Reboot the server
5. Launch the Gather Unity System Info tool and verify the output shows the updated version.

Note that due to the intermittent nature of this problem, it is difficult for Cisco to fully verify internally. To date, all customers that have been confirmed to be encountering the symptoms described above have had their problems resolved by the application of this patch. If you continue to encounter the same difficulties after applying the patch, please contact Cisco TAC and reference this article and defect CSCti70702.

Average Rating: 5 (3 ratings)

Comments

Brad Magnani Mon, 09/27/2010 - 19:10 (reply to blukens07)

No, the information on this page is the most current you will find.  Keep an eye on this page as well as the Microsoft KB article, as we too are eagerly awaiting the release from Microsoft.

timothyhollis Tue, 10/05/2010 - 08:35

I am also eagerly awaiting the new CDO release.  Not only am I having internal users complaining about VM checking issues, I am also getting reports that external users that are calling in to leave messages are getting cut-off midway through their message.  My users have been suffering for about two weeks now and I can see the villagers in the distance carrying their pitchforks and torches.  Please hurry Microsoft!!!

emorivera2526 Fri, 10/08/2010 - 14:16 (reply to cbelcher)

Over 2 months since I first opened a TAC case about this problem.  I'm pretty sure the natives have already drawn and quartered me, but my pergatory is dealing with the same issue and waiting for Microsoft to patch

pjweppner Wed, 10/27/2010 - 12:45 (reply to williamsutton)

Installed on 10/26 @ 4:30 PM and as of 10/27 @ 3:44 PM not a single error or warning in the app log on a very busy system where I would have hundreds of errors by now using the old CDO. So far so good. No new issues being reported yet either.

blukens07 Tue, 10/26/2010 - 07:02

when is the CDO planned to be included in the Unity software or serverupdatewizard?

tollom Fri, 10/29/2010 - 09:13

any idea if this will finally be supported by TAC now that the patch from MS is out???

pahayes Fri, 10/29/2010 - 09:38 (reply to tollom)

The updated CDO tools have been supported with Unity since the day they were released by Microsoft (10/25).That is the case for nearly all applicable Microsoft updates/patches. It is only service packs and major releases that have to be qualified for Unity. For more details, see Unity's "Support Policy for Microsoft Service Packs and Updates" here:

http://www.cisco.com/en/US/docs/voice_ip_comm/unity/8x/support/8xcusupp.html#wp525581

adamcrabb Sun, 11/14/2010 - 16:34

We are currently running Unity 7.0(2) ES35, Exchange 2010 partner server, and a mixed 2003/2010 Exchange environment.  We're moving away from 2003 but that process takes time, so for now we're mixed.  A few weeks back we started having serious MWI issues.  A manual "Resynchronize Now" usually resolved the issue but only for a few minutes.  When I'd run the resync there were always MaleX errors, MAPI disconnections, and server search errors in the application event log.  After installing the Microsoft patch i have zero errors when running a manual MWI sync, so an immediate improvement.  I still have some delay (3-4 minutes) that I still need to track down with TAC and am thinking it may be related to a hardware load balancer that sits in front of Exchange. 

Our company runs a Cisco Ace 4700 hardware load balancer to load balance the connections to the Exchange CAS array.  We have configured 'affinity' or 'sticky' for services like OWA, OA, and EWS.  The original Cisco configuration document for Exchange 2010 HLB with Ace does not make mention of configuring affinity for RPC traffic.  I've read a few threads that are starting to recommend configuring "Client IP" as the affinity method for RPC traffic.  Does anyone out there have experience with a HLB in front of an Exchange 2010 environment?  Do you have affinity configured for RPC traffic?

Thanks!

Actions

Login or Register to take actions

This Document

Posted September 2, 2010 at 10:19 AM
Stats:

Related Content