Released: Update Rollup 3 for Exchange 2010 Service Pack 2

Earlier today, the Exchange CXP team released Update Rollup 3 for Exchange Server 2010 SP2 to the Download Center.

This update contains a number of customers reported and internally found issues. See KB2685289 Description of Update Rollup 3 for Exchange Server 2010 Service Pack 2 for more details.

Note: Some of the following KB articles may not be available at the time of publishing this post.

In particular we would like to specifically call out the following fixes which are included in this release:

  • KB2689810 Meeting Requests bodies get rendered in plain text in Outlook when created via Exchange Web Services.
  • KB2674445 Need the function to check ACL consistency during moving mailbox.
  • KB2700705 RpcClientAccess crashes with SocketException when UDP push notification is enabled.
  • KB2705425 Memory Leak in UMWorkerProcess.exe.
  • KB2698976 MRM Assistant doesnt process a mailbox with a contact created in other tenants.

General Notes:

For DST Changes:

Note for Forefront Protection for Exchange users  For those of you running Forefront Protection for Exchange, be sure you perform these important steps from the command line in the Forefront directory before and after this rollup’s installation process. Without these steps, Exchange services for Information Store and Transport will not start after you apply this update. Before installing the update, disable ForeFront by using this command: fscutility /disable. After installing the update, re-enable ForeFront by running fscutility /enable.

Exchange Team

Comments (48)
  1. Tony V says:

    KB link seems invalid.  Searching for 2685289 doesn't bring up anything either

  2. Renzo Geelhoed says:

    When is the support for Windows 2012 ready? Now, when you have DC running this, OWA does not work….

  3. @Renzo – Windows Server 2012 has not been released to the web and thus is not supported in production environments unless you are in the appropriate TAP.  


  4. Bharat Suneja [MSFT] says:

    @Tony V: When the download goes live and we publish the post, the KBs are not replicated and thus may not be available. We include a note indicating "Some of the following KB articles may not be available at the time of publishing this post.". In due course, they become available. For example, I can now access KB 2685289 as linked.

  5. Fred M says:

    2671128 – RPC Client Access Cross-Site connectivity issues occur in an Exchange Server 2010 environment

    Surprised this isn't highlighted because if I understand correctly, this is a big deal for people who've been missing the old Exchange behavior when moving mailboxes between servers at different sites – the Outlook profile will finally be updated properly now.

    Anything coming for SP1?

  6. markus says:

    Is the language pack for SP2 released 6 days ago included in this rollup package ?

  7. Greg Taylor [msft] says:

    @ Markus – no, it is not. The language pack was a one-off release to solve some specific issues. We do not update language packs via roll ups.

  8. Dan Sheehan says:

    Many thanks to the Exchange team for releasing this update. It actually fixes a number issues we were experiencing and/or want to proactively avoid as we expect to encounter them very soon.

    Of special note for my team are the following:

    Synchronization of an organizational forms library fails when you use Outlook in Cache mode in an Exchange Server 2010 environment (this is HUGE as this has been an issue since we migrated to Exchange 2010).…/2572029

    You cannot access a mailbox for several hours after you disconnect and then reconnect the mailbox in an Exchange Server 2010 SP2 environment (we routinely have to run clean-database to remediate this issue).…/2682047

    You cannot create a mailbox or mail-enable a mailbox for a disabled user account in an Exchange Server 2010 environment (this is a welcome modification so we can cut down on the back and forth with our customers)…/2706523

    Thanks again!

  9. Phil says:

    That 'disable Forefront' thing is so old had, so tedious done manually, when it can be done in a script instead.  So I fixed the (buggy) example script from an earlier Exchange Team blog post and made it more resilient:

    <a href="…/entry_199.html">Disabling Forefront for Exchange 2010 when Installing Exchange Service Packs and Hotfix Rollups</a>


  10. Bharat Suneja [MSFT] says:

    @Fred M: Check out today’s post by RS4:

    Client Access Cross-Site Connectivity Changes

  11. Phil says:

    Oops, mangled url…

    That 'disable Forefront' thing is so old had, so tedious done manually, when it can be done in a script instead.  So I fixed the (buggy) example script from an earlier Exchange Team blog post and made it more resilient:…/entry_199.html


  12. Roady [MVP] says:

    KB2700705 doesn't seem to exist (yet?) nor is there a reference to it in the description of the rollup (KB2685289). Is this a publishing oversight or is the fix not included in this rollup after all?

  13. @Roady – KB2700705 article hasn't been published and we are working to fix that.  Yes the fix referenced in the above title is included in SP2 RU3.


  14. Roady [MVP] says:


    Thanks! It's live now and also listed in the main description (KB2680383 is newly listed as well).

  15. Novice says:

    Can we apply RU3 to some of the CAS servers in an CAS Array? Can we apply RU3 to some of the CAS servers which are not part of the CAS Array?

  16. Dan Sheehan says:

    @Ross – has there been any modifications to Exchange 2010 in recent rollups to allow admins to specify the URL to use for anonymous connections to published calendars should start with HTTPS:// and not http://?

    Putting a hardware load balancer in front of the CASs where the default configuration is to not support direct HTTP requests (but rather redirect customers to HTTPS) seems to break this default URL generation for a lot of customers with no real way to fix it other than writing some crazy rules on the hardware load balancer (assuming that works which we have not even tried to do).


  17. @Novice – yes you can deploy SP2 RU3 on some servers in the CAS array (obviously from a deployment perspective this is how you would do it – you would take a set of servers out of teh LB pool, update them, return them to service, remove the next set, etc.).  Our guidance for deployment is CAS, then HT, then MBX.  However, if you deploy it on certain machines keep in mind that some clients will get updated functionality, while others will not.  This could increase troubleshooting time.

  18. @Dan – We have not changed the calendar publishing feature to use HTTPS instead of HTTP.

  19. Dan Sheehan says:

    @Ross – thanks for the clarification.

    We would submit a DCR to ask they provide some sort of flag on CAS cmdlet to have it change from using HTTP to HTTPS in the URL formulation, but the product team is so focused on the next version of Exchange that they tend to shoot down such DCRs.

    Thanks again for the clarification.

  20. @Dan, I can't speak to how that particular issue would score out, but I would still recommend going through the process of working with Premier to submit the CDCR.  This at least ensures that the Exchange Servicing team knows about the issue and understands the customer pain.  The feedback can also be shared with the component team which can also impact future versions.

  21. Novice says:


    I read another blog entry regarding Client Access Server Connectivity changes but didn't understand it fully. As a "novice", is this rollup update introduce any changes to the Exchange environment other than bug fixes?

  22. @Novice – Yes, rollups can contain introduce new functionality into the product.  Typically these types of changes come at the request of customers in what refer to as critical design change requests (CDCRs).

  23. Karsten says:

    Although not mentioned – I think this is the first version of Exchange 2010 there the following problem is fixed:

    Cant get a client Rule against the Global Adresslist in Outlook get to work

    (should be fixed with KB2543850 in SP2 RU1)

    Thanks for including it now.

    The following problem still exists:

    If I delete many mails keeping delete Button pressed I get the Error "A clientactivity failed" after a while. This is only with activated Previw in online mode. It is ok with OLK2003 the Problem now is with OLK 2010 – Could you please correct this in the next RU?

  24. @Karsten – Please open a support ticket to troubleshoot your issue.

  25. says:

    Bug report:

    It started with SP2RU1 (and there was no change in RU2 nor in RU3): If you're running Exchange 2010 with Exchange 2003 Public Folder servers at a site, OWA is not working (inviting attendees…). You need at least one Exchange 2010 PF server (or avoid applying any RUs after SP2).


  26. Karsten says:

    Thank you Ross.

    There is one new Error in the Application Eventlog of our hub.cas servers about twenty times a day since I installed RU3:

    EventID: 1

    Source: MSExchange Autodiscover

    Nicht verarbeitete Ausnahme "Benutzereinstellung 'PreferredSite' ist nicht verfügbar. "


    Is this a known Problem?

  27. Greg Taylor [msft] says:

    It is known Karsten, we are investigating.

  28. says:

    @Karsten: Lucky you. We're getting it ~16000 times per day on our CASes…

  29. Karsten says:

    @FYDIBOHF25TVDLT: as far as I can say until now – there are no problem for the user resulting from the error. But of course it makes the logfile hard to read.

    I found a second Event in the application eventlog of our servers with the Mailbox role (this is not new with SP2 RU3 but it is anoying, too):

    Event Number: 106

    Source: MSExchange Common

    Fehler beim Aktualisieren eines Leistungsindikators. … Last worker process info : System.UnauthorizedAccessException: Der Zugriff auf den Registrierungsschlüssel "HKEY_LOCAL_MACHINESOFTWAREMicrosoftExchangeServerv14Transport" wurde verweigert.

    The server has only the Mailboxrole – so it is no wonder that the "…Transport" key in the registry does not exist. Why does exchange try to renew an performancecounter at this key on an pure Mailbox server?

  30. says:

    @Karsten: Looks like the amount of "errors" corresponds to the number of users connecting to the CAS. On a test box (1 user) we also see ~20-30 events per day.

  31. Elliot G says:

    I have seen this disable FPE advice before. But when I run the command I get:

    C:Program Files (x86)Microsoft Forefront Protection for Exchange Server>fscuti

    lity /disable


     Microsoft Forefront Protection Admin Utility – v 11.0.0713.0

     (c) Microsoft Corporation. All rights reserved.


     Microsoft Forefront Protection cannot be disabled at this time.  Please stop

     the following services and then rerun this utility.



    Is this expected??  If so why don't the instructions tell you to do these things as well?

  32. Elliot G says:

    I should add that the above example was run on the Mailbox server. If you run it on the hub transport server you get the trnasport service instead of the information store that needs to be stopped.

  33. janx says:

    After adding this RU, I'v got events like this (maybe somebody can comment):

    Starting from 25.06.2012 12:43:08 service <Unknown> has performed this activity on the server:

    RPC Operations: 720.

    Database Pages Read: 4 (of which 4 pages preread).

    Database Pages Updated: 0 (of which 0 pages reupdated).

    Database Log Records Generated: 0.

    Database Log Records Bytes Generated: 0.

    Time in Server: 32 ms.

    Time in User Mode: 32 ms.

    Time in Kernel Mode: 0 ms.

  34. Paulchen says:

    Is there a solution for the "Unhandled Exception "User setting 'PreferredSite' is not available." ???


  35. Tirtul says:

    I've noticed one thing with local mailbox move requests: My entire environment is on SP2 UR3, but when I do a local mailbox move (from one db to another) via the EMC, it doesn't recognize the correct version of Exchange.

    For example, in the move request Details, I notice the following::

    Source Version: Version 14.2 (Build 309.0)

    Target Version: Version 14.2 (Build 247.0)

    If I were to dig into the move request log, it reflects the same, so what gives???

  36. Phil says:

    @Elliot G

    If you're disabling Forefront manually, you're doing it the hard way.

    See my comment here dated May 30th.

  37. Karsten says:

    @Paulchen: I have no solution for the "User setting 'PreferredSite' " Problem. But it seem to make no further problems. As Ross stated that the ExchangeTeam is aware of the Problem I hope it will be fixed in the next RU.

  38. nico says:

    Any Updates regarding the error event 1  ? :

    Unhandled Exception "User setting 'PreferredSite' is not available. "

    Stack trace:


  39. preferredsite error says:

    same autodiscover error user preferredsite on my cas servers as well

  40. IlMuro says:

    Same Unhandled Exception "User setting 'PreferredSite' is not available. " on my newley installed Exchange 2010 SP2 RU3 servers. Events started after installing the RU3 update.

    Any news? Don't really want to migrate users onto the new systems if there is an issue.

  41. Folks asking about that Autodiscover error, in a TechEd NA session a few weeks ago Scott Schnoll said it was harmless (aside from filling up your event logs of course)

    Some more details here:

  42. Kris says:

    OK – so I didn't see the part about disabling forefront and now can't use Exchange. Is the easiest way to fix this uninstall the update. Then disable forefront and reinstall?

    Will that even work?

  43. vijay says:

    After rollout update 3 on the exchange 2010SP2 CAS servers, the following events are being logged in the event viewer.Please advice how to stop this errors.


    Exchange servers are montiored by exchange management pack using SCOM 2007R2.It is happening only active nodes which is load balanced by F5 device.

    Log Name:      Application

    Source:        MSExchange Control Panel


    Event ID:      38

    Task Category: General

    Level:         Error

    Keywords:      Classic

    User:          N/A



    Current User: '/Service Account/extest_781bf962b5f34'

    Exchange Control Panel detected an invalid canary from request for URL '/ECP/RulesEditor/InboxRules.svc/GetList'.

    Canary in cookie: 'D-V0lprQT02WUy4LzThBx_jhRYv1Ls8IwvqjJNX-7jr-diYmy8hwGVDEuWOYnDor8budENZTbZw.' mismatch with canary in header/form: ', in URL '.


  44. Amir Haque [MSFT] says:

    @ Vijay

    We've seen some cases on this issue, one resolution was … "The command appears to be case sensitive and changing '/ECP' to '/ecp' in the URL (pasted below) resolved the issue." See if this helps.

    <pasting URL from the error you have provided>

    Exchange Control Panel detected an invalid canary from request for URL '/ECP/RulesEditor/InboxRules.svc/GetList'.

  45. Arun1981 says:

    Folks, I've been reading through the issues which everyone has faced before and after applying the SP2 RU3, but after reading all the posts, for me RU3 doesnt looks to be a RU which has many bugs.  So can I expect this to be a safe RU to install it in the production environment?

    Your expert advice is highly appreciated…

  46. G12 says:

    Looks like the biggest issue I've encountered with RU3 is with moving mailboxes.  Every mailbox move to a different database within the same server ends with "CompletedWithWarning".  In going through the logs, I see the following:  MapiExceptionUnexpectedMailboxState: Unable to delete mailbox. (hr=0x80004005, ec=2634)

    This issue is mentioned here with no resolution:…/6043b74f-a41e-4674-b2b1-08c3b1a3afbd

  47. says:

    It would be awesome if somebody could check that the bug discussed in…/0e91199f-bb00-469c-a20e-0a302d2197ce

    has been fixed. Note the

     "You can contact Microsoft to ask for intern fix for this issue."

    which implies that the problem was recognized as product problem and fixed somehow.

Comments are closed.