none
MSFTESQL-Exchange errors after Update Rollup 4 for Ex2010 SP2 RRS feed

  • Question

  • Hi,

    I applied Update Rollup 4 for Ex2010 SP2 three days ago, and since then I've been having the following error messages:

    Event 4999:

    Watson report about to be sent for process id: 14116, with parameters: E12, c-RTL-AMD64, 14.02.0297.000, ExMSFTE, M.E.Search.ExSearch, M.E.S.Crash.CrashNow, M.E.Common.FailFastException, 678e, 14.02.0318.001.
    ErrorReportingEnabled: False

    ---

    Event 64772:

    The filter daemon process MSFTEFD exited unexpectedly. Here is the stack trace:

    Event 1503:

    The system exception c000010a was raised at 000007FEF9B60F16. Stack: 

    These errors are coming up a couple of times an hour.

    I thought it might be something to do with corrupt DB catalog files for searching, so I used the ResetSearchIndex.ps1 script to reset them, but this hasn't fixed anything.

    Any ideas what to do next?

    Thanks,

    Adam

    Thursday, August 23, 2012 1:32 PM

All replies

  • I started having the same problem after installing Update Rollup 4. Bump for a suggestion.

    Thursday, August 23, 2012 1:52 PM
  • Same issue here, another bump for a suggestion.
    Thursday, August 23, 2012 8:57 PM
  • Bump x3, I can find lot of articles related to EX 2007 SP1 but nothing for EX 2010 SP2


    ---------------- "You can run, but you cannot hide. It will find you!"

    Friday, August 24, 2012 12:07 AM
  • Several people are definitely having this problem, but solutions are not exactly forthcoming. If it's any help: our Exchange 2010 environment is currently co-existing with Exchange 2003. In order to "resolve" this issue, I ended up uninstalling the 4th Update Rollup. We have not fully deployed 2010 yet, so it wasn't a huge deal to uninstall the update. We also jumped from update rollup 2 straight to 4 - skipping 3.
    Friday, August 24, 2012 1:31 AM
  • Another note: Our org. is all Exchange 2010, although it was migrated from 2003. SP2 rollups have been applied in order, if that matters.
    Friday, August 24, 2012 3:14 PM
  • I'm having the same exact errors on two new servers runing UR4 that I installed into an existing Exchange 2010 org.

    It also seems like during mailbox moves CPU utilization is going to 100% with a large amount consumed by msftefd.exe.


    Monday, August 27, 2012 6:54 PM
  • Same problem here. Started when I installed E2K10 SP2 RU4, got a mixed E2K3 environment. Indexing works, however it takes ages to reindex databases which I did on E2K10. Seems like a bug introduced in RU4.
    Tuesday, August 28, 2012 5:35 AM
  • Same problem here. Started as soon as I installed SP2 RU4. I'm running straight Exchange 2010 SP2 RU4 across the org.
    • Edited by Jarrett B Tuesday, August 28, 2012 9:23 PM
    Tuesday, August 28, 2012 9:22 PM
  • I was planning on running the update this weekend but it looks to be too soon? 

    Im also running a mix of 2K10 and 2k3 Exchange. 

    What is the "Event 4999" actually doing?

    Would appreciate some feedback and if RU4 is even worth installing at this point with all the mentions of KB's not even working after the update?


    Tuesday, August 28, 2012 9:40 PM
  • This deals with Exch07, but this looks very similar.

    http://support.microsoft.com/kb/970311

    I'm gettting a 4999 error from 'General' (the watson report)

    Then a 64772 from MSFTESQL Service (The filter daemon process MSFTEFD exited unexpectedly. Here is the stack trace: )

    Then a 1053 from MSFTESQL Service (The system exception c000010a was raised at 000007FEE60A0F16. Stack: )

    Nearly identical except for 64772 on mine vs. 64773 in the KB article.

    Tuesday, August 28, 2012 9:50 PM
  • Bump to remind the MCPs and MS people that nobody has addressed these errors and we would like a bit of help.
    Wednesday, August 29, 2012 4:26 PM
  • I would open a ticket with Microsoft support if you need help with these errors.

    Wednesday, August 29, 2012 5:23 PM
    Moderator
  • Opended a ticket today with MS Support. Been seeing this since the Update Rollup 4 install we did Monday night. 

    Wednesday, August 29, 2012 8:09 PM
  • Opended a ticket today with MS Support. Been seeing this since the Update Rollup 4 install we did Monday night. 


    Bikermike, please let us know how it goes.
    Wednesday, August 29, 2012 8:29 PM
  • Hello all,

    Same problem here but just on one of our DAG ...

    It seem's that this error come when the content indexing process meet a particular item in a mailbox.

    Let's see if the next ROLLUP can avoid it.

    Any how, very interested by the response from microsoft ... Bikermike, please let us know how it goes.

    Thursday, August 30, 2012 5:18 PM
  • Update 2 (4 Sept 2012):

    After getting some more of these events, I now haven't had any in 4 days, so I not sure if it had anything to do with what I did below.

    Update 1 (31 Aug 2012):

    Just got another event logged, so below wasn't the answer.

    Original Post (30 Aug 2012):

    We had this exact same issue after installing RU4 last Friday.  Yesterday we discovered the "Allow automatic replies" and "Allow automatic forward" to our Default Remote Domains had been disabled, which I can only assume was done by the installation of RU4.  None of our auto reply rules were working as we discovered yesterday.  I re-enabled the automatic replies and forwards last night and now the errors have gone.

    Go figure!

    Cheers

    Damian


    • Edited by DPW2807 Tuesday, September 4, 2012 12:00 AM Update 2
    Thursday, August 30, 2012 11:46 PM
  • Fortunately I was in the same situation as Jordan and I've uninstalled UR4. After that, everything went fine again.

    Regards Andreas

    Thursday, September 6, 2012 12:32 PM
  • I'm seeing the same errors after running resetsearchindex.ps1 on a catalog that was showing as corrupted.  
    Tuesday, September 11, 2012 7:29 AM
  • Same here as Kanien.  Everything was fine after Exchange2010/SP2/RU4 ... until I ran the above script & now I'm getting non-stop errors.
    Tuesday, September 11, 2012 11:25 PM
  • I would again recommend opening a ticket with Microsoft support.

    Tuesday, September 11, 2012 11:27 PM
    Moderator
  • How does one do that?  I don't have that information, but I'll be delighted to open a ticket if you can provide me the appropriate contact info.
    Tuesday, September 11, 2012 11:34 PM
  • I would again recommend opening a ticket with Microsoft support.


    It's hard for me to justify spending money with MS support for an issue caused by their update. How do I sell that?
    Tuesday, September 11, 2012 11:41 PM
  • We are running Exchange Server 2010 SP2 Rollup2 on Server 2008 R2 SP1 on VMWare ESX 4.1. Yesterday (Sunday) we applied Rollup4 to our systems (1 Mailbox Server, 3 CAS Servers, 1 HUB Server). We began seeing those same ERRORS in our Application Event Log of our Mailbox Server (the other servers had no errors).

    I believe we have this resolved by applying the following Hotfix to Microsoft Filter Pack 2.0:

    http://support.microsoft.com/kb/2597139

    Only time will tell - but so far clean event logs.

    Monday, September 17, 2012 6:57 PM
  • I would again recommend opening a ticket with Microsoft support.


    It's hard for me to justify spending money with MS support for an issue caused by their update. How do I sell that?

    If its a bug, you wont be charged.

    Monday, September 17, 2012 7:03 PM
    Moderator
  • Has anyone else noticed that as soon as these errors start, they start getting failures in the CI checks? And then logs stop truncating after successful backup operations?

    When we first installed UR 4, we didn't notice these, but as we finished up migrations and more users started using the databases, we noticed the CI errors on multiple databases in addition to the events everyone else is seeing.

    I've started a rebuild of the indexes, will let everyone know how that goes. If that fails, it's time to call Microsoft as our logs haven't truncated in 4 days and things could get ugly.

    Tuesday, September 18, 2012 10:14 PM
  • Yip, having same issue...

    Exch 2010, Exch 2003 with SP2 UR4

    Wednesday, September 19, 2012 4:02 AM
  • We are running Exchange Server 2010 SP2 Rollup2 on Server 2008 R2 SP1 on VMWare ESX 4.1. Yesterday (Sunday) we applied Rollup4 to our systems (1 Mailbox Server, 3 CAS Servers, 1 HUB Server). We began seeing those same ERRORS in our Application Event Log of our Mailbox Server (the other servers had no errors).

    I believe we have this resolved by applying the following Hotfix to Microsoft Filter Pack 2.0:

    http://support.microsoft.com/kb/2597139

    Only time will tell - but so far clean event logs.

    Rebuilding the content indexes didn't help. I've just installed the above mentioned hot fix, we'll see if that helps.

    I've also noticed a huge increase in the number of logs being generated on all databases, at the rate of about 1 every two seconds.

    ***Update: The Filter Pack hot fix did nothing, the  errors are still all over the place and even after running for 10 hours, the databases are still not indexed. Granted, they are about 300GB each and there are 16 of them, so I'm going to chalk it up to that.

    • Edited by alextesla Wednesday, September 19, 2012 4:52 PM Update
    Wednesday, September 19, 2012 7:13 AM
  • I am having the same issue, installed the RU over the weekend and the errors are now there.

    Has anyone found a way to resolve this issue yet?

    Thank you,

    Tom


    Just an FYI, I stopped the Microsoft Search (Exchange) service and the errors have stopped. I am thinking that there was something in this RU to try to fix the search function in OWA and let alone did it not fix that it has caused other issues.
    • Edited by thoff616 Monday, September 24, 2012 4:02 PM FYI
    Monday, September 24, 2012 3:49 PM
  • Just an update for everyone, I've been working with Microsoft since this afternoon. We are going to uninstall UR4 and go back to SP2 UR3. I've been giving Microsoft a lot of diagnostic information before this, so hopefully they can help address the issue with UR4. They don't want to call it a bug yet, but it is a known issue in Microsoft.

    I will report back tonight once I've completed the rollback to UR3.

    Tuesday, September 25, 2012 1:27 AM
  • Just an update for everyone, I've been working with Microsoft since this afternoon. We are going to uninstall UR4 and go back to SP2 UR3. I've been giving Microsoft a lot of diagnostic information before this, so hopefully they can help address the issue with UR4. They don't want to call it a bug yet, but it is a known issue in Microsoft.

    I will report back tonight once I've completed the rollback to UR3.

    Thank you for the update. Please let us know how the unistall works out also.

    Tom

    Tuesday, September 25, 2012 1:37 AM
  • Just an update for everyone, I've been working with Microsoft since this afternoon. We are going to uninstall UR4 and go back to SP2 UR3. I've been giving Microsoft a lot of diagnostic information before this, so hopefully they can help address the issue with UR4. They don't want to call it a bug yet, but it is a known issue in Microsoft.

    I will report back tonight once I've completed the rollback to UR3.

    The uninstall of UR4/rollback to UR3 worked, the errors are no longer showing up. However, the content indexes are corrupt from the errors and are requiring a rebuild, getting 5301 warning  from the CI Troubleshooter in the application logs. So far this seems to have worked for the ones I've tested. I have a follow-up with Microsoft tomorrow, hopefully they can resolve this soon and get UR4 back in working order.
    • Proposed as answer by alextesla Wednesday, September 26, 2012 12:09 AM
    Wednesday, September 26, 2012 12:09 AM
  • Thank you again for the update.

    Looks like I will be unistalling UR4 and waiting.

    Tom

    Wednesday, September 26, 2012 12:47 PM
  • One of our servers has the same issue, though UR4 was installed some time ago (and the errors started only today). 
    Thursday, September 27, 2012 7:14 PM
  • Hi, is there any progress in your case? We experience the same issue here.

    Freek

    Monday, October 1, 2012 8:30 AM
  • Hi,

    Same here. RU4 was installed some time ago, but those erros showed up today.


    rgds Sven


    Deem.. it seems that I'm unable to remove RU4.. during uninstall it complains some missing "exchangeserver.msi" or smth. like that..
    • Edited by Sven J Tuesday, October 2, 2012 8:01 PM
    Tuesday, October 2, 2012 4:14 PM
  • I'm having the same exact errors on two new servers runing UR4 that I installed into an existing Exchange 2010 org.

    It also seems like during mailbox moves CPU utilization is going to 100% with a large amount consumed by msftefd.exe.


    I just updated to RU4 a few days ago and I'm in the middle of a 2003 to 2010 migration.  I noticed that these errors happen during mailbox moves and the errors stop after mailbox moves are complete.  I didn't notice the msftefd.exe using more than 15% CPU and my content indexes appear to be in a Healthy state, so I'm not sure if this is a cosmetic bug or not.

    We would appreciate responses from admins who have cases with Microsoft to reply with their resolution.  Right now, I'm just viewing this as a cosmetic bug as I don't see anything wrong with the operations of Exchange.

    Saturday, October 6, 2012 3:33 PM
  • I'm also in the middle of an Exchange 2003 to 2010 migration and I'm only seeing these errors during mailbox moves. I'm also hoping that these are just cosmetic since I haven't seen any other issues.  Mailboxes all seem to migrate successfully.
    Sunday, October 7, 2012 5:01 PM
  • In our case (2007 to 2010 migration) i also only see the messages during the mailbox move.

    How can I check if the search index is healthy? 

    I tried searching within OWA and with Outlook and it seems to function all correctly...

    Monday, October 8, 2012 6:48 AM
  • In our case (2007 to 2010 migration) i also only see the messages during the mailbox move.

    How can I check if the search index is healthy? 

    I tried searching within OWA and with Outlook and it seems to function all correctly...


    Run "get-mailboxdatabasecopystatus *" to check the ContentIndexState.
    Monday, October 8, 2012 6:40 PM
  • Something interesting that I might add.  On Saturday night I did not have any mailbox moves, but during a period of 4 hours, I got these errors.  On Sunday night I also did not have any mailbox moves and no errors.
    Monday, October 8, 2012 6:42 PM
  • Run "get-mailboxdatabasecopystatus *" to check the ContentIndexState.

    After we got the errors a week ago while moving the mailboxes , yesterday i decided to move some other mailboxes. The errors didn't appeared, also not afterwards. During the mailbox move the msfte was running at 100% CPU usage.

    This morning I ran the check and the ContentIndexState was healthy. 



    Tuesday, October 9, 2012 9:18 AM
  • Has anyone gotten and installed SP2 UR4-v2 yet?  We have 13,000 mailboxes on ten servers and are supposed to install Ex2010 SP2 UR4 this weekend (from SP1 UR6).  We aren't going to if there are issues with the update - we'll go with UR3 if we need to.

    http://www.microsoft.com/en-us/download/details.aspx?id=34956

    Wednesday, October 10, 2012 1:44 PM
  • In our case (2007 to 2010 migration) i also only see the messages during the mailbox move.

    How can I check if the search index is healthy? 

    I tried searching within OWA and with Outlook and it seems to function all correctly...

    Can you please open up a case so this can be tracked? 

     If you want, let me know the case # via the contact me page on my blog .


    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:     LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Wednesday, October 10, 2012 2:06 PM
  • Has anyone gotten and installed SP2 UR4-v2 yet?  We have 13,000 mailboxes on ten servers and are supposed to install Ex2010 SP2 UR4 this weekend (from SP1 UR6).  We aren't going to if there are issues with the update - we'll go with UR3 if we need to.

    http://www.microsoft.com/en-us/download/details.aspx?id=34956

    Would also want to know if UR4-v2 is all OK, since UR4 seems to be a big mess.

    Thanks!

    Wednesday, October 10, 2012 7:27 PM
  • Log Name:      Application
    Source:        MSFTESQL-Exchange
    Date:          10/10/2012 3:21:55 PM
    Event ID:      1053
    Task Category: MSFTESQL Service
    Level:         Error
    Keywords:      Classic
    User:          N/A
    Computer:      
    Description:
    The system exception c000010a was raised at 000007FEEC540F16. Stack: 
    Event Xml:
    <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
      <System>
        <Provider Name="MSFTESQL-Exchange" />
        <EventID Qualifiers="32768">1053</EventID>
        <Level>2</Level>
        <Task>1</Task>
        <Keywords>0x80000000000000</Keywords>
        <TimeCreated SystemTime="2012-10-10T19:21:55.000000000Z" />
        <EventRecordID>56668</EventRecordID>
        <Channel>Application</Channel>
        <Computer>/Computer>
        <Security />
      </System>
      <EventData>
        <Data>c000010a</Data>
        <Data>000007FEEC540F16</Data>
        <Data>
        </Data>
      </EventData>
    </Event>

    Don't Install Rollup 4 it has issues


    • Edited by ITS-ERIC Wednesday, October 10, 2012 9:04 PM
    Wednesday, October 10, 2012 9:03 PM
  • Don't Install Rollup 4 it has issues



    And those issues wont get resolved unless the people experiencing these problems open cases with Microsoft Support. They will get the attention they need if you open a case.
    Wednesday, October 10, 2012 10:53 PM
    Moderator
  • Exactly, please do let us know what issues you are experiencing!

    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:     LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Wednesday, October 10, 2012 11:32 PM
  • Exactly, please do let us know what issues you are experiencing!

    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:     LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Rhoderick,

    There are plenty off issues listed in this thread, do you know if those issues have been taken care of in UR4 v2? Are there no known issues in v2, where you would recommend installing it, or should one wait for UR5?

    Thanks!

    Thursday, October 11, 2012 1:21 PM
  • Just an update from my end.

    The errors seemed to have gone away on my end. The MSFTEFD error have not shown up in Event Viewer for several days now but the error that started this whole thing still exist and the is the CIsearch error. I am afraid to rebuild the indexes again for fear that the MSFTEFD errors will come back but I will most likely rebuild them soon to try to fis the CIsearch errors.

    I would be interested to hear if anyone has installed V2 of this update though.

    Thank you,

    Tom

    Update: Well I went ahead and started a rebuild of the indexes and the errors have now returned. As long as no one wants to search in OWA I guess that is fine but I doubt that. I also have it set that the Watson errors are sent to M$ automatically so they are seeing alot of errors sent from my part of the world right now. :)

    • Edited by thoff616 Friday, October 12, 2012 3:47 PM
    Friday, October 12, 2012 3:21 PM
  • Situation (Exhange 2010 SP2 RU4v2):

    I'v reseted search indexing. After some time I got errors like belov:

    Watson report about to be sent for process id: 7980, with parameters: E12, c-RTL-AMD64, 14.02.0318.004, ExMSFTE, M.E.Search.ExSearch, M.E.S.Crash.CrashNow, M.E.Common.FailFastException, 678e, 14.02.0318.004.

    ErrorReportingEnabled: True

    Watson report about to be sent for process id: 7980, with parameters: E12N, c-rtl-AMD64, 14.02.0247.005, MsFTEFD.exe, KERNELBASE.dll, caed, e0434f4d, 706e, 6.1.7601.17932 (win7sp1_gdr.120820-0419).

    ErrorReportingEnabled: True


    rgds Sven

    Tuesday, October 16, 2012 8:05 PM
  • Thanks for letting us know Sven!

    Guess v2 isn't much better... hope UR5 will be better!?
    I'm not even going to bother with UR4 or UR4-v2... too sad... :/

    Tuesday, October 16, 2012 8:11 PM
  • bikermike04 mentioned here, that support ticket is opened.

    What about that? Any progress?


    rgds Sven

    Wednesday, October 17, 2012 11:54 AM
  • Got the same problem (RU4), curious about any news !
    Wednesday, October 17, 2012 1:13 PM
  • Currently we have a 2003 and 2010 exchange environment.
    RU4v2 has been applied and during our first mailbox migration from 2003 to 2010 we received the 4999,64772,1053 events.

    It appears that we had this only during the mailbox moves and the content index states on our databases are Healthy.  Before I migrate any more mailboxes, anybody push through migrating mailboxes with this error?  Any fallout?  Anybody have a fix?

    Thursday, October 18, 2012 2:59 AM
  • I found the detailed error messages in c:\windows\temp\ExMSFTE-unknown-0-ExWatsonReport.xml. looks like a sql error:

    "<details>FailFastException (Got exception System.IndexOutOfRangeException: Der Index war außerhalb des Arraybereichs"

    our mailbox move is done a few days ago, during the migration I stopped the service because of the errors jwr22314 mentioned. but they still occur now. Even resetting the search index (database is now in "crawling" state) and update the office filterpack to sp1 didn't help.

    • Proposed as answer by Murashid Thursday, October 18, 2012 9:12 AM
    • Unproposed as answer by Murashid Thursday, October 18, 2012 9:12 AM
    Thursday, October 18, 2012 8:27 AM
  • this is a known issue, contact microsoft for help
    Thursday, October 18, 2012 9:12 AM
  • this is a known issue, contact microsoft for help

    Your comment is based on what? I'v mentioned this thread in Exchange Team Blog regardig UR4-v2 and they answered, that this problem is not known.

    rgds Sven

    Thursday, October 18, 2012 1:18 PM
  • Well... for any new people visiting this thread, it seems safe to assume SP2, UR4 and UR4-v2 are very risky to install. And one should wait for UR5 and hope it will have all these problems resolved.

    I'm surprised (and disapointed) the MS reps here are not speaking up a little better on our questions...

    Thursday, October 18, 2012 1:43 PM
  • this is a known issue, contact microsoft for help


    Your comment is based on what? I'v mentioned this thread in Exchange Team Blog regardig UR4-v2 and they answered, that this problem is not known.

    rgds Sven

    This one:  http://blogs.technet.com/b/exchange/archive/2012/10/09/re-released-exchange-2010-and-exchange-2007-update-rollups.aspx

    Maybe it was not at the time.  That was a while back. 


    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:     LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Thursday, October 18, 2012 3:37 PM
  • I don't have details of a fix to get this resolved Perski - sorry! 

    There are several posts above saying to call in and a get a case opened up as we need to hear about this. 

    Yes we do monitor the forums, and this valuable feedback is passed on internally but we still need cases to ensure that the issues that matter are being resolved.  If no one calls 999/911 will the fire engine come?  We can all talk outside the burning building, but the calls have to be made.  Don't leave the calls up to others, that's all I'm trying to say. 

    Yes -- there is always an element of risk when making changes, be it hardware or software.  That is why we want customers to test prior to placing an update into production. 


    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:     LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Thursday, October 18, 2012 3:43 PM
  • There are several posts above saying to call in and a get a case opened up as we need to hear about this. 

    Yes we do monitor the forums, and this valuable feedback is passed on internally but we still need cases to ensure that the issues that matter are being resolved.  If no one calls 999/911 will the fire engine come?  We can all talk outside the burning building, but the calls have to be made.  Don't leave the calls up to others, that's all I'm trying to say. 

    Yes -- there is always an element of risk when making changes, be it hardware or software.  That is why we want customers to test prior to placing an update into production. 


    I would think they know about this being that they recommend that Error Reporting be enabled and everytime this happens it is sent to the Microsoft Mothership.

    I am glad that there are people like you that monitor these types of Forums. You are a wealth of information that we all very happy to have.

    Not all of us have the option to test things like a service pack before hand. I would think that something like this would be fully tested before being published. I understand that you will not be able to test everything and I cannot speak for everyone but my server is a standard build with no hacks on it so I would assume that a SP from Microsoft would go on this without causeing issues to a standard build SBS server.

    Again, I thank you for being here and offering the help you do.

    Tom

    ErrorReportingEnabled: True

    Thursday, October 18, 2012 3:55 PM
  • Rhoderick,<o:p></o:p>

    I have opened a case with MS as of a few minutes ago. Since there appears to be no fix
    forth coming what is the best course of action? roll the servers back by removing RU4v2 and wait till ru5?
    If we are to remove 4 and wait what RU should be installed that is known stable.
    Case Number:
    112101856494566<o:p></o:p>

    Jim<o:p></o:p>


    Thursday, October 18, 2012 3:56 PM
  • Rhoderick,<o:p></o:p>

    I have opened a case with MS as of a few minutes ago. Since there appears to be no fix
    forth coming what is the best course of action? roll the servers back by removing RU4v2 and wait till ru5?
    If we are to remove 4 and wait what RU should be installed that is known stable.
    Case Number:
    112101856494566<o:p></o:p>

    Jim<o:p></o:p>


    Hi Jim,

    Firstly, I'd recommend a Bing/Google search for problems with each RU release prior to this just as a double check (as in, if a prob is discovered draw a line through and move to next RU) - but I recently completed an upgrade of our Ex2010 environment from SP1 RU4 to SP2 RU3 and have been running stable for over 2 weeks.

    I hope that helps.

    Curby


    CurbySanPan - Still Learning Exchange

    Thursday, October 18, 2012 4:03 PM
  • Curby,

    Will see what becomes of my ticket with MS.  Looking like I might be removing RU4 and headed back to RU3.  I have not been in contact with MS yet but will update once I have more.

    Thursday, October 18, 2012 4:33 PM
  • That is a distinct possibility, but let's wait for the ticket to get worked on first. 

    And my "official" answer as to what RU to revert to?  It will be the last one that you tested and worked in your environment.  Sorry! 


    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:     LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Thursday, October 18, 2012 5:11 PM
  • I also did a MS call yesterday. The suggestions was to install RU4v2 and look, if the errors disappear. No success ? then to uninstall RU4. Let's see ...
    Friday, October 19, 2012 5:04 AM
  • BTW..

    I have more errors. Those errors are below and they started a while after instaling Update Rollup 4.
    Installing UR4v2 does not changed anything too:

    1.
    The execution time of agent 'ScanMail Routing Agent' exceeded 90000 milliseconds while handling event 'OnResolvedMessage' for message with InternetMessageId: 'Not Available'. This is an unusual amount of time for an agent to process a single event. However, Transport will continue processing this message.

    And about 10 minutes later:

    2.
    Information Store (3132) A significant portion of the database buffer cache has been written out to the system paging file.  This may result in severe performance degradation.
    See help link for complete details of possible causes.
    Resident cache has fallen by 38094 buffers (or 20%) in the last ? seconds.
    Current Total Percent Resident: 79% (149708 of 188282 buffers)

    I know, that the first one points to my antivirus TrendMicro ScanMail for Exchange 10.2 SP1 (I don't have file level AV in place). But I have not changed anything except installed UR4 and UR4v2.
    Before UR4 my exchange logs were clean as a babys a**. No significant database grows or something similar.
    So I'm inclined to think that those errors are related to UR4/UR4v2 too.


    rgds Sven

    Friday, October 19, 2012 2:06 PM
  • Im running into the same errors. They first appeared last week but RU4 was installed on release day.. Thinking about going back to RU3.  Can I savely remove it via the uninstallation at "installed update" at my DAG?

    Thanks!

    Friday, October 19, 2012 6:13 PM
  • Im running into the same errors. They first appeared last week but RU4 was installed on release day.. Thinking about going back to RU3.  Can I savely remove it via the uninstallation at "installed update" at my DAG?

    Thanks!


    Same as mine? My errors also appeared last week (RU4 insalled about 2..3 weeks ago) ..coincidence?
    I had a problem rolling back RU4. Uninstallation started and some time later uninstallation progress asked me to provide some kind of "exchangesetup.msi" (or smthg similar file). I don't have it..so I canceled the uninstallation. I just gave up and some days later updated to RU4v2, but all my problems remained.

    rgds Sven


    • Edited by Sven J Friday, October 19, 2012 7:05 PM
    Friday, October 19, 2012 7:04 PM
  • As I mentioned in a previous post, I am too having these errors, but I have yet to find anything wrong or broken which leads me to still believe this is a cosmetic issue.  For those of you who like me are experiencing, are you seeing any technical issues or is it just cosmetic?  I'm staying on SP RU4 as I have no technical reason to go backwards.
    Friday, October 19, 2012 7:16 PM
  • At the moment.. I have not seen any major impact to end users but I have identified some delays in mailflow when those errors came up..so, the big picture is not very clean in my side.

    rgds Sven

    Friday, October 19, 2012 7:33 PM
  • In our enviroment after applying RU4 (coming from SP2 without RU's), logs stop truncating after successful backup operations, but only for one particular mailstore. For now I work around it disabling circular logging before backup and re-enable it afterwords. After reading this thread I think it is pretty useless applying RU4v2 as this will not solve this problem? I am opening a support ticket at Microsoft first thing monday.



    Saturday, October 20, 2012 2:46 PM
  • Rhoderick,<o:p></o:p>

    I have opened a case with MS as of a few minutes ago. Since there appears to be no fix
    forth coming what is the best course of action? roll the servers back by removing RU4v2 and wait till ru5?
    If we are to remove 4 and wait what RU should be installed that is known stable.
    Case Number:
    112101856494566<o:p></o:p>

    Jim<o:p></o:p>


    After submitting the ticket which by the way since this is a known issue is at no charge and currently there are at least 22 logged cases regarding this, the following is where I am at.

    After reviewing my logs and looking at my specific issues, the tech had me add the following register key which turns off code added in RU4.
    HKLM\SOFTWARE\Microsoft\ExchangeServer\v14\ExSearch Create a new 32bit Dword “NeverOverrideAttachmentExtension”  Set value to 1.

    While migrating two mailboxes with the added reg entry The 1053 and 64772 events have not returned.  HOWEVER, the following additional entries were logged:

    4999, MSExchange Common: 
    Watson report about to be sent for process id: 3700, with parameters: E12, c-RTL-AMD64, 14.02.0318.004, ExMSFTE, M.E.Search.ExSearch, M.E.S.Crash.CrashNow, M.E.Common.FailFastException, 678e, 14.02.0318.004. ErrorReportingEnabled: False

    4999, MSExchange Common:
    Watson report about to be sent for process id: 3700, with parameters: E12N, c-rtl-AMD64, 14.02.0247.005, MsFTEFD.exe, KERNELBASE.dll, caed, e0434f4d, ca67, 6.1.7601.17932 (win7sp1_gdr.120820-0419).
    ErrorReportingEnabled: False

    Event 10036, MSExchangeIS Mailbox Store:
    Continuous replication block mode is unable to keep up with the data generation rate. Block mode has been suspended, and file mode has been resumed.Database: ad194699-6bc8-4816-96a8-df78c1ff59b3
    IoDepth: 3102712
    IoLatencyInMsec: 9003

    Once I reported the new finding the tech stated that MS was consolidating all these cases to one individual that will be back in the office on Monday.  Until then I will not be moving any mailboxes which of course is holding up the migration.
    As I find out more I will pass it on.

    Saturday, October 20, 2012 8:47 PM
  • I decided to ask for more information within my MS CALL regarding these errors and did not uninstall RU4. We are in production with 1300 boxes and I cannot "try and error".
    Sunday, October 21, 2012 2:32 PM
  • Rhoderick,<o:p></o:p>

    I have opened a case with MS as of a few minutes ago. Since there appears to be no fix
    forth coming what is the best course of action? roll the servers back by removing RU4v2 and wait till ru5?
    If we are to remove 4 and wait what RU should be installed that is known stable.
    Case Number:
    112101856494566<o:p></o:p>

    Jim<o:p></o:p>


    After submitting the ticket which by the way since this is a known issue is at no charge and currently there are at least 22 logged cases regarding this, the following is where I am at.

    After reviewing my logs and looking at my specific issues, the tech had me add the following register key which turns off code added in RU4.
    HKLM\SOFTWARE\Microsoft\ExchangeServer\v14\ExSearch Create a new 32bit Dword “NeverOverrideAttachmentExtension”  Set value to 1.

    While migrating two mailboxes with the added reg entry The 1053 and 64772 events have not returned.  HOWEVER, the following additional entries were logged:

    4999, MSExchange Common: 
    Watson report about to be sent for process id: 3700, with parameters: E12, c-RTL-AMD64, 14.02.0318.004, ExMSFTE, M.E.Search.ExSearch, M.E.S.Crash.CrashNow, M.E.Common.FailFastException, 678e, 14.02.0318.004. ErrorReportingEnabled: False

    4999, MSExchange Common:
    Watson report about to be sent for process id: 3700, with parameters: E12N, c-rtl-AMD64, 14.02.0247.005, MsFTEFD.exe, KERNELBASE.dll, caed, e0434f4d, ca67, 6.1.7601.17932 (win7sp1_gdr.120820-0419).
    ErrorReportingEnabled: False

    Event 10036, MSExchangeIS Mailbox Store:
    Continuous replication block mode is unable to keep up with the data generation rate. Block mode has been suspended, and file mode has been resumed.Database: ad194699-6bc8-4816-96a8-df78c1ff59b3
    IoDepth: 3102712
    IoLatencyInMsec: 9003

    Once I reported the new finding the tech stated that MS was consolidating all these cases to one individual that will be back in the office on Monday.  Until then I will not be moving any mailboxes which of course is holding up the migration.
    As I find out more I will pass it on.

    Update - the MS tech had me install the following update on all my database servers (Office 2010 hotfix package (Filterpack-x-none.msp): February 28, 2012): http://support.microsoft.com/kb/2597139
    It did ask to reboot which i let it do. Once the update was applied and rechecked the registry updates i moved two mailboxes individually with no error.

    I am going to continue to monitor and hope this has resolved the issue that I have been facing.

    Monday, October 22, 2012 7:15 PM
  • Tried this hotfix. Mine does not asked to reboot, so I did not.
    Used resetsearchindex.ps1 -force -all but after a while, the errors are not gone:

    1. Watson report about to be sent for process id: 8000, with parameters: E12, c-RTL-AMD64, 14.02.0318.004, ExMSFTE, M.E.Search.ExSearch, M.E.S.Crash.CrashNow, M.E.Common.FailFastException, 678e, 14.02.0318.004.
    ErrorReportingEnabled: True

    2. The filter daemon process MSFTEFD exited unexpectedly. Here is the stack trace: (- there's no additional information)

    3. The system exception c000010a was raised at 000007FEDC920F16. Stack: (- there's no additional information)

    So.. I'm still stuck...
    Or maybe should I reboot and reset again...


    rgds Sven

    Tuesday, October 23, 2012 3:29 PM
  • I also tried to reset my indexes and received the errors during the indexing. The errors stopped when the indexing was finished but I still receive the below error.

    Function CISearch::EcGetRowsetAndAccessor detected that content indexing was disabled for database 'DatbaseName' because of error '0x80041820' from MSSearch.

    Everything seems to be working well with the exception of not being to search in OWA. Not sure if that will ever be fixed.

    Thank you,

    Tom


    • Edited by thoff616 Tuesday, October 23, 2012 4:49 PM
    • Proposed as answer by Remigiusz_Sz Monday, October 29, 2012 11:12 AM
    Tuesday, October 23, 2012 4:49 PM
  • I also tried to reset my indexes and received the errors during the indexing. The errors stopped when the indexing was finished but I still receive the below error.

    Function CISearch::EcGetRowsetAndAccessor detected that content indexing was disabled for database 'DatbaseName' because of error '0x80041820' from MSSearch.

    Everything seems to be working well with the exception of not being to search in OWA. Not sure if that will ever be fixed.

    Thank you,

    Tom


    That looks like a different error than the others in this thread Tom. From a quick look I'm wondering if you are running into this: http://blogs.technet.com/b/exchange/archive/2011/06/30/after-installing-exchange-2010-service-pack-1-searching-via-owa-or-outlook-online-mode-fails.aspx

    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:     LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Tuesday, October 23, 2012 5:32 PM
  • I also tried to reset my indexes and received the errors during the indexing. The errors stopped when the indexing was finished but I still receive the below error.

    Function CISearch::EcGetRowsetAndAccessor detected that content indexing was disabled for database 'DatbaseName' because of error '0x80041820' from MSSearch.

    Everything seems to be working well with the exception of not being to search in OWA. Not sure if that will ever be fixed.

    Thank you,

    Tom


    That looks like a different error than the others in this thread Tom. From a quick look I'm wondering if you are running into this: http://blogs.technet.com/b/exchange/archive/2011/06/30/after-installing-exchange-2010-service-pack-1-searching-via-owa-or-outlook-online-mode-fails.aspx

    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:     LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    I have to agree with you but the errors that are mentioned above are what started it and now I only recieve those errors when I try to index, otherwise I just receive the error I mentioned now as long as the indexing does not take place.

    I did try your the fix you linked in but it did not fix anything.
    I have tried alot of different things all the way to moving boxes to a new database and it did not fix the problem but I am still open to suggestions. :)
    Thank you,

    Tom

    Tuesday, October 23, 2012 5:57 PM
  • I also tried to reset my indexes and received the errors during the indexing. The errors stopped when the indexing was finished but I still receive the below error.

    Function CISearch::EcGetRowsetAndAccessor detected that content indexing was disabled for database 'DatbaseName' because of error '0x80041820' from MSSearch.

    Everything seems to be working well with the exception of not being to search in OWA. Not sure if that will ever be fixed.

    Thank you,

    Tom


    That looks like a different error than the others in this thread Tom. From a quick look I'm wondering if you are running into this: http://blogs.technet.com/b/exchange/archive/2011/06/30/after-installing-exchange-2010-service-pack-1-searching-via-owa-or-outlook-online-mode-fails.aspx

    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:     LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    You are correct, the latest error I receive is different. The errors reported at the beginning of this message I was receiving at the beginning. They slowly stopped showing up and now they do not show up till I try to reindex the database. The error that I said I recieve now is about 30 minutes to every hour.
    I did try what your link suggested and it did not fix the issue with searching emails in OWA.

    I am open to suggestions though.  :)

    Thank you,

    Tom

    Tuesday, October 23, 2012 7:35 PM
  • Just to check one thing Tom - after running the ExchangeSearchSymlinks.ps1 from an elevated EMS, did you then rebuild the Index.  Just want to confirm....


    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

         

    Blog:http://blogs.technet.com/rmilneTwitter: LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.


    Tuesday, October 23, 2012 7:45 PM
  • Tried this hotfix. Mine does not asked to reboot, so I did not.
    Used resetsearchindex.ps1 -force -all but after a while, the errors are not gone:

    1. Watson report about to be sent for process id: 8000, with parameters: E12, c-RTL-AMD64, 14.02.0318.004, ExMSFTE, M.E.Search.ExSearch, M.E.S.Crash.CrashNow, M.E.Common.FailFastException, 678e, 14.02.0318.004.
    ErrorReportingEnabled: True

    2. The filter daemon process MSFTEFD exited unexpectedly. Here is the stack trace: (- there's no additional information)

    3. The system exception c000010a was raised at 000007FEDC920F16. Stack: (- there's no additional information)

    So.. I'm still stuck...
    Or maybe should I reboot and reset again...


    rgds Sven


    Reboot and reset did'nt help...

    rgds Sven

    Tuesday, October 23, 2012 9:01 PM
  • Just to check one thing Tom - after running the ExchangeSearchSymlinks.ps1 from an elevated EMS, did you then rebuild the Index.  Just want to confirm....


    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

         

    Blog:http://blogs.technet.com/rmilneTwitter: LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.


    Yes sir.

    Tom

    Tuesday, October 23, 2012 9:32 PM
  • As you know I opened a ticket and I was informed that these errors are caused by a known bug. Even RU4v2 will not help. I can decide wheter to uninstall RU4 or to wait for a fix in the near future. Well I think I can live with it and wait.

    • Proposed as answer by SKapitzke Monday, November 12, 2012 1:23 PM
    Wednesday, October 24, 2012 5:03 AM
  • same here,

    the microsoft told me that is a know bug, and RU4v2 is a workarround ou something, we created the registry entry for

    NeverOverrideAttachmentExtension

    and then this issue will be fixed on next rollup update.


    Foi util? Classifique ^^ MCP - MCDST - MCTS - MCSA - MCITP www.nrv.com.br www.alexandrenakagawa.com.br

    Wednesday, October 24, 2012 4:42 PM
  • When is RU5 due?
    Wednesday, October 24, 2012 5:08 PM
  • Rhoderick,<o:p></o:p>

    I have opened a case with MS as of a few minutes ago. Since there appears to be no fix
    forth coming what is the best course of action? roll the servers back by removing RU4v2 and wait till ru5?
    If we are to remove 4 and wait what RU should be installed that is known stable.
    Case Number:
    112101856494566<o:p></o:p>

    Jim<o:p></o:p>


    After submitting the ticket which by the way since this is a known issue is at no charge and currently there are at least 22 logged cases regarding this, the following is where I am at.

    After reviewing my logs and looking at my specific issues, the tech had me add the following register key which turns off code added in RU4.
    HKLM\SOFTWARE\Microsoft\ExchangeServer\v14\ExSearch Create a new 32bit Dword “NeverOverrideAttachmentExtension”  Set value to 1.

    While migrating two mailboxes with the added reg entry The 1053 and 64772 events have not returned.  HOWEVER, the following additional entries were logged:

    4999, MSExchange Common: 
    Watson report about to be sent for process id: 3700, with parameters: E12, c-RTL-AMD64, 14.02.0318.004, ExMSFTE, M.E.Search.ExSearch, M.E.S.Crash.CrashNow, M.E.Common.FailFastException, 678e, 14.02.0318.004. ErrorReportingEnabled: False

    4999, MSExchange Common:
    Watson report about to be sent for process id: 3700, with parameters: E12N, c-rtl-AMD64, 14.02.0247.005, MsFTEFD.exe, KERNELBASE.dll, caed, e0434f4d, ca67, 6.1.7601.17932 (win7sp1_gdr.120820-0419).
    ErrorReportingEnabled: False

    Event 10036, MSExchangeIS Mailbox Store:
    Continuous replication block mode is unable to keep up with the data generation rate. Block mode has been suspended, and file mode has been resumed.Database: ad194699-6bc8-4816-96a8-df78c1ff59b3
    IoDepth: 3102712
    IoLatencyInMsec: 9003

    Once I reported the new finding the tech stated that MS was consolidating all these cases to one individual that will be back in the office on Monday.  Until then I will not be moving any mailboxes which of course is holding up the migration.
    As I find out more I will pass it on.

    Update - the MS tech had me install the following update on all my database servers (Office 2010 hotfix package (Filterpack-x-none.msp): February 28, 2012): http://support.microsoft.com/kb/2597139
    It did ask to reboot which i let it do. Once the update was applied and rechecked the registry updates i moved two mailboxes individually with no error.

    I am going to continue to monitor and hope this has resolved the issue that I have been facing.


    I have not had the events that I was seeing in my case return. I have moved a few mailboxes and they appear to be moving cleanly. The hotfix and reg entry seems to have done the trick.

    James Robson

    Wednesday, October 24, 2012 7:56 PM
  • James,

    Just curious if you can search an email box for email in OWA or if you force the server to reindex the database do the errors show up with the things you have done ?

    Thank you,

    Tom

    Thursday, October 25, 2012 12:42 AM
  • I don't have details of a fix to get this resolved Perski - sorry! 

    There are several posts above saying to call in and a get a case opened up as we need to hear about this. 

    Yes we do monitor the forums, and this valuable feedback is passed on internally but we still need cases to ensure that the issues that matter are being resolved.  If no one calls 999/911 will the fire engine come?  We can all talk outside the burning building, but the calls have to be made.  Don't leave the calls up to others, that's all I'm trying to say. 

    Yes -- there is always an element of risk when making changes, be it hardware or software.  That is why we want customers to test prior to placing an update into production. 


    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:     LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Rhoderick,

    That might be fine for people with a Test environment (similar enough to Prod) but we don't have that. And I can't risk going through this bug and opening a case while our users are suffering. That's why I just wanted to know if UR4-v2 specifically was supposed to fix this. I won't update until a release has been out long enough, and I know there are no bugs like this being reported here in the forums.

    I will at least wait until UR5 now.

    Thanks!

    Thursday, October 25, 2012 4:06 PM
  • I would like to know does UR$-V2 address this issue.

    Thanks

    Thursday, October 25, 2012 6:28 PM
  • Sejod,

    To quote SKapitzke:

    "As you know I opened a ticket and I was informed that these errors are caused by a known bug. Even RU4v2 will not help. I can decide wheter to uninstall RU4 or to wait for a fix in the near future. Well I think I can live with it and wait."

    From reading this thread, UR4-v2 does not seem to fully fix the problems, so if you are not yet using UR4, better wait for UR5, or if you are using UR4 and have problems, may be worth rolling back to UR3. At least if you want to stay problem free. We have lots of OWA only users, so I can't live with this issue or any work arounds, I'm waiting for UR5.

    Thursday, October 25, 2012 6:45 PM
  • James,

    Just curious if you can search an email box for email in OWA or if you force the server to reindex the database do the errors show up with the things you have done ?

    Thank you,

    Tom

    Tom,

    Searching an email box for email via OWA does not appear to cause the errors.  I will check on the reindex and let you know.

    Jim


    James Robson

    Thursday, October 25, 2012 6:55 PM
  • Hello,

    In my server , I tried theese steps and resolved my problem ;

    Set-MailboxDatabase ' Mailbox Database Name '-IndexEnabled $False

    Exchange scripts ;

    .\GetDatabaseForSearchIndex.ps1

    .\ResetSearchIndex.ps1 -all

    Set-MailboxDatabase ' Mailbox Database Name '-IndexEnabled $True
    • Edited by dosdibi Friday, October 26, 2012 4:36 PM
    • Proposed as answer by dosdibi Monday, November 12, 2012 7:43 PM
    • Unproposed as answer by adfrad Tuesday, November 13, 2012 2:01 PM
    Friday, October 26, 2012 1:53 PM
  • For the benefit of others that will totally disable indexing on the database that you run it on. 

    Just to be clear, That is a workaround rather than a fix  :)


    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:     LinkedIn:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Friday, October 26, 2012 3:25 PM
  • I forgot to write last command. I added to post.
    Friday, October 26, 2012 4:38 PM
  • We are migrating from 2007 to 2010 SP2 RU4 as well and are running in to these event log errors.    I have opened support case with MS:  112102946620820

    The filter daemon process MSFTEFD exited unexpectedly. Here is the stack trace: 

    Watson report about to be sent for process id: 344, with parameters: E12, c-RTL-AMD64, 14.02.0318.004, ExMSFTE, M.E.Search.ExSearch, M.E.S.Crash.CrashNow, M.E.Common.FailFastException, 678e, 14.02.0318.004.
    ErrorReportingEnabled: True

    The system exception c000010a was raised at 000007FEF9DB0F16. Stack: 

    Watson report about to be sent for process id: 9152, with parameters: E12, c-RTL-AMD64, 14.02.0318.004, ExMSFTE, M.E.Search.ExSearch, M.E.S.Crash.CrashNow, M.E.Common.FailFastException, 678e, 14.02.0318.004.
    ErrorReportingEnabled: True

    Monday, October 29, 2012 1:00 PM
  • Ok, just an update from my corner of the world.
    I was doing maintenance on another server over the weekend and decided to go ahead and in stall Version 2 of RU 4. Figured it could not hurt anything.

    Well after it installed and went and checked to see if the search function would work in OWA, and low and behold it worked for the first time.

    Usually when you try to search in OWA you would start to get errors in Event Viewer so I went and check there and no errors.

    As of about 15 minutes ago it is still working and I do not plan to reboot the server, scared it will stop working if I do. :)
    Thank you,

    Tom

    Monday, October 29, 2012 4:42 PM
  • Hello,

    I had the same problems.
    Here my workaround:
    - deinstall RU4
    - install RU3
    - reset search index \ResetSearchIndex.ps1 - force -all

    Everything ist working fine. Now i´m waiting for RU5.

    Günther

    Friday, November 2, 2012 7:56 AM
  • So, basically no additional information? Some people here are saying that this problem is known for MS at the moment.. any hotfix? Workaround? Any additional informaton? Results from opened tickets? Anything?

    rgds Sven

    Tuesday, November 6, 2012 7:00 AM
  • You can test if the \ResetSearchIndex.ps1 - force -all will assist you environment by...cycling 2 Microsoft search services in this order.<o:p></o:p>

    Restart 1. Microsoft Search (Exchange) then restart the 2. Microsoft Exchange Search Indexer. {Once cycled OWA searching was fixed}

    After validating that the users were able to search OWA, Running the \ResetSearchIndex.ps1 - force -all...on our 132 mail-db's which assisted with resolving reported issues of searching across the board.<o:p></o:p>

    once the\ResetSearchIndex.ps1 - force -all completed test mailbox health via PS
    cmdlet -> C:\>get-mailboxdatabasecopystatus *<o:p></o:p>


    Tuesday, November 6, 2012 2:28 PM
  • Tanya,

    I am experiencing a little more than cosmetic issues.  We are getting the errors typically when doing a mailbox migration (2003_>2010).  Then seemingly the CAS array thinks it loses a member and migrates mailbox database.  After which any new activesync devices (which we initially quarantine), continuously act as if not approved.


    Minoo Jalal

    Thursday, November 8, 2012 9:53 PM
  • I just really hope Microsoft gets these issues corrected in Update Rollup 5.
    Thursday, November 8, 2012 10:13 PM
  • I just really hope Microsoft gets these issues corrected in Update Rollup 5.

    Update is available since 13th November: http://support.microsoft.com/kb/2719800/en-us

    I must say i don't see anything related to this in the KB article :-(


    • Edited by mcv Friday, November 16, 2012 9:34 AM
    Friday, November 16, 2012 9:29 AM
  • WTF.. there are over 100 posts.. and no helpful responce from MS..and I also don't see anything related to this problem in UR5...

    rgds Sven

    Friday, November 16, 2012 10:07 AM
  • unfortunaltelly this update Rollup 5 causes new issues so I don't recommend to install this Rollup 5 now in production environment.

    Exchange 2010 SP2 Error after upgrading to Rollup 5

     


    Remigiusz
    ExchangeBlog


    Friday, November 16, 2012 11:01 AM
  • That's not totally accurate Sven,

    The workaround is posted above, and yes this is not in RU5. 

    Please also ensure your office filter pack is the 2010 edition, and you have it up to date. 

    http://support.microsoft.com/kb/2460041?wa=wsignin1.0


    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:    LinkedIn:    Facebook:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Friday, November 16, 2012 2:04 PM
  • That's not totally accurate Sven,

    The workaround is posted above, and yes this is not in RU5. 

    Please also ensure your office filter pack is the 2010 edition, and you have it up to date. 

    http://support.microsoft.com/kb/2460041?wa=wsignin1.0


    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Where is it posted above?


    • Edited by zoltank Friday, November 16, 2012 2:07 PM
    Friday, November 16, 2012 2:06 PM
  • The registry edit.  Try searching for NeverOverrideAttachmentExtension.


    Cheers,

    Rhoderick

    Microsoft Premier Field Engineer, Exchange

    Blog: http://blogs.technet.com/rmilne  Twitter:    LinkedIn:    Facebook:

    Note: My posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Friday, November 16, 2012 2:20 PM
  • This is really unbelievable... makes one wonder if Microsoft doesn't have a test environment themselves?! (I'm sure they do, but then they got one real poor testing team behind it!)

    The RU4 problem hasn't been fixed in RU5, and RU5 has a bunch of new problems on top of that...! :(
    So what's next, wait for SP3?! Maybe they test Service Pack releases better??

    Big time disappointed...

    Friday, November 16, 2012 2:34 PM
  • That's not totally accurate Sven,

    The workaround is posted above, and yes this is not in RU5. 

    Please also ensure your office filter pack is the 2010 edition, and you have it up to date. 

    http://support.microsoft.com/kb/2460041?wa=wsignin1.0


    Cheers,

    Rhoderick

    Okay.. my apologies. I have not tested yet this registry trick. I will. But I'm still not satisfied with Your responce. First post on this thread is about 23 August, UR5 was released 13 November. Two months and all You can provide is some kind of workaround to turn off "something". What exactly I'm turning off? What this setting "NeverOverrideAttachmentExtension to 1" is doing? What I'm losing if this reg. key is set to "1" ?

    BTW: I have MS Office FilterPack SP1 + Hotfix for MS FilterPack 2.0 (KB2597139) installed.


    rgds Sven

    Friday, November 16, 2012 2:53 PM
  • UR5 can no longer be downloaded due to bug that affects DAG's.

    Looks like we have to wait for version 2 


    Remigiusz
    ExchangeBlog


    • Edited by Remigiusz_Sz Saturday, November 17, 2012 8:15 AM
    Saturday, November 17, 2012 8:15 AM
  • UR5 can no longer be downloaded due to bug that affects DAG's.

    Looks like we have to wait for version 2 


    Remigiusz
    ExchangeBlog



    This must be joke....

    rgds Sven

    Saturday, November 17, 2012 8:37 AM
  • Amazingly poor job.

    He's right, they pulled the KB http://support.microsoft.com/kb/2719800

    Colossal failure. Thank god I don't trust MS Rollups any more.

    Monday, November 19, 2012 6:18 PM
  • Hi,

    I'm seeing these errors after migrating some mailboxes from 2007 sp3 to 2010 sp2 ru4v2,

    but what are the symptoms to indicate that this is the same problem?

    We're not getting these errors repeatedly; so far, only during a mailbox move.

    I'm not hearing reports of issues with search in Outlook or OWA.  The errors appeared during mailbox moves and stopped.  None of the other servers have these errors.

    Has Microsoft published anything to help me decide whether to proceed with the migration?

    Thanks,

    rudif 


    • Edited by rudi f Monday, November 19, 2012 11:00 PM
    Monday, November 19, 2012 10:57 PM
  • I am Not Sure Have you fix the problem or not but please find the workaround.

    Workaround:

    Microsoft confirmed that it’s an active bug with E2010 Sp2 RU4, a fix will be available in RU6 for exchange server 2010 Sp2.

    Added the following registry changes and and installed latest folder pack.

    HKLM\SOFTWARE\Microsoft\ExchangeServer\v14\ExSearch
    Create a new 32bit Dword “NeverOverrideAttachmentExtension”
    Set value to 1.

    Description of the Office 2010 hotfix package (Filterpack-x-none.msp): February 28, 2012
    http://support.microsoft.com/kb/2597139

    Will defenetlly resolve the problem

    Wednesday, November 21, 2012 1:31 PM
  • thanks, but what is the problem?  I have the errors appear during a migration, only several errors in the application log, but I'm not getting any other symptoms.

    Is there an official post of the issue and the workaround?

    I'm trying to figure out whether the errors can be ignored or taken seriously before I continue my migration.

    Is there something I can do to test that I need the registry key and the hotfix other than the fact that I get these errors?

    thanks,

    rudif

    Wednesday, November 21, 2012 3:23 PM
  • Update.

    I can confirm, that this trick with modifying registry, helps us. But the other questions are still unanswered.


    rgds Sven

    Thursday, November 22, 2012 9:19 AM
  • FYI:  I open SR in GTSC Support for this issue and they said me that the fix is targeted for RU6.


    • Edited by umone Wednesday, November 28, 2012 10:59 AM
    Wednesday, November 28, 2012 10:58 AM
  • I’ve had RU4 for many weeks without any errors (during my 2007 to 210 migration).  I have moved a couple dozen mailboxes without these errors.  I suddenly experienced these errors during the move of a couple mailboxes last evening.  Today I have continued moving mailboxes so far without any more errors.  I am of course quite torn about moving forward however I don’t seem to be experiencing any other side effects as of yet.  I’m going to proceed and simply monitor the event viewer.  I will also leave my exchange 2007 in place for the time being just in case.

    Wednesday, December 5, 2012 6:42 PM
  • Completed the move, no errors, no issues.
    Saturday, December 8, 2012 12:33 AM
  • I notice UR5 has been re-released as it has a security hotfix, no obvious sign of a problem for this issue, but then it does mention UR6 above.

    It would be nice to have an idea of when UR6 will come out.

    Jeff

    Wednesday, December 12, 2012 11:45 AM

  • Good stuff about RU4-v1, RU4-v2, RU5-v1, RU5-v2 and RU6


    Also good stuff on the Open Office XML file extension attachment to message bug and recommendation to update the Filter pack.


    Since RU4-v1 and RU4-v2 appear to have an Indexing problem, which was not addressed in either RU5-v1 and RU5-v2 (and those introduced DAG support issues, and we do use DAGs)


    We will be waiting on RU6 to even contemplate testing an RU rollout on SP2 production this year.


    Kind of strange that its going to be over a year since the last "safe" RU release for SP2.


    I hope the Exchange support team addresses this at one of the Microsoft Exchange events this year. It will be the first question I ask.


    I opened a separate ticket to discuss this issue.

    How would Content Indexing handle an NDR storm or Bounces for a mailbox?


    But you might consider if a user accidentally created a Forwarding rule in their mailbox and started an NDR storm that it possibly could overwhelm the msftefd.exe process until they used up all their mailbox space.

    By default Exchange does not permit Forwarding messages to external mailservers, but it is also one of the first things people enable. And exposes this possibility. Hence it could "randomly" occur and be hard to track down.

    There is a method for automatically preventing NDR storms on Technet here:

    Prevent NDRs Storms


    There is also a nice discussion on how to audit mailboxes for client inserted External Forwarding rules here:

    Ever needed to find Server Side Outlook rules that forward mail outside of your organisation


    Finally, mailbox databases on DAGs can sometimes get manipulated "before" the Indexer is aware an object to be indexed has been moved or deleted in which case it can "spin" trying to access that object at its former location which the Indexer looked up in AD before attempting to index. Discussed here:

    Microsoft.Exchange.Search.ExSearch.exe using 100% CPU for xx hours


    And Antivirus run on Exchange role server can sometimes deny access to important files at critical times. Discussed here:

    Exchange Server 2010 and Antivirus exclusions


    Most of the best practices I've read indicate its a "really, really" good idea to disable Indexing on any mailbox database you plan to do something with well in advance and give it a few minutes to complete its last tasks before "moving the database, moving a mailbox in the database, deleting a mailbox, deleting a mailbox database" or doing anything with a replicated DAG.

    The msftedt.exe "process" is only instantiated when something needs to be done, so if its in your process list it really is "in the middle of something" and yanking resources away from it could cause it to spin for a long time or at least until manually killed.

    Indexing "updates" the catalogs, so to me that means disabling Indexing on the databases probably won't affect searches while your doing maintenance type things.. and it is an "optional" service.. stopping Indexing will not shutdown mail delivery or routing. (I think) it will merely mean the catalogs will just stop updating until you turn it back on.

    "One last thing" -- I seem to recall a "race" condition between DAG services and the Indexing service. Such that if the Index service were not delayed on start up it could clobber the DAG services and prevent a database from mounting. I believe this has been fixed but can't find a reference in the Release notes for any Updates that refer to it. So if a database refuses to mount you might disable the Indexing service and see if that solves the problem. You can enable it after the database mounts.


    • Edited by jwillis84 Monday, January 7, 2013 9:04 AM
    Monday, January 7, 2013 8:34 AM
  • I just wanted to post that these errors do still persist, while moving mailboxes between databases, in Rollup 5-V2.  I am only getting these errors during the migrations.
    Saturday, January 12, 2013 7:52 PM
  • I just wanted to post that these errors do still persist, while moving mailboxes between databases, in Rollup 5-V2.  I am only getting these errors during the migrations.

    FWIW, me too.

    Did my post help? Please use "Vote As Helpful", "Mark as answer" or "Propose as answer". Thank you!


    Tuesday, February 12, 2013 1:17 PM
  • On Tue, 12 Feb 2013 13:17:47 +0000, Jetze Mellema wrote:
     
    >I just wanted to post that these errors do still persist, while moving mailboxes between databases, in Rollup 5-V2. I am only getting these errors during the migrations.FWIW, me too.
    >
    >
    >Did my post help? Please use "Vote As Helpful", "Mark as answer" or "Propose as answer". Thank you!
     
    Does this help?
    http://exchangeblog.pl/en/2012/12/exchange-2010-event-id-4999-msftefd-exe-process-crashes/
     
    ---
    Rich Matheisen
    MCSE+I, Exchange MVP
     

    --- Rich Matheisen MCSE+I, Exchange MVP
    Wednesday, February 13, 2013 3:13 AM
  • I think it is too early for asking. After the problems with the RU.v2 versions ... I will wait before installing. But our thread problem and the fix is mentioned here: http://support.microsoft.com/kb/2746164/en-us
    Wednesday, February 13, 2013 2:17 PM
  • Installed RU6. So far, so good. If I find anything related to this topic I will post here.

    rgds Sven

    Thursday, February 14, 2013 9:45 AM
  • I can confirm the issue has been resolved by installing UR6 for Exchange 2010 SP2. The fix was not mentioned in the list of problems fixed in this UR but it still solved my issue.

    After installing UR6 I saw event id 5604 and 5611 from the CI Troubleshooter for a couple of hours. After CI had processed all items in the queue the event disappeared and everything seems working again.


    Did my post help? Please use "Vote As Helpful", "Mark as answer" or "Propose as answer". Thank you!

    Thursday, February 14, 2013 1:14 PM
  • I can confirm the issue has been resolved by installing UR6 for Exchange 2010 SP2. The fix was not mentioned in the list of problems fixed in this UR but it still solved my issue.

    After installing UR6 I saw event id 5604 and 5611 from the CI Troubleshooter for a couple of hours. After CI had processed all items in the queue the event disappeared and everything seems working again.


    Did my post help? Please use "Vote As Helpful", "Mark as answer" or "Propose as answer". Thank you!

    Thanks for letting us know the results of your testing Jetze.  One thing to mention though, this fix should be listed in the RU6 documents.  I have it in the highlighted ones in this post

    http://blogs.technet.com/b/rmilne/archive/2013/02/12/exchange-2010-sp2-ru6-released.aspx

    http://support.microsoft.com/kb/2746164 also lists http://support.microsoft.com/kb/2776259 as resolved which the 4999 crash in the Search process.  Is that not the issue you were having?


    Cheers,

    Rhoderick

    Microsoft Senior Exchange PFE

    Blog: http://blogs.technet.com/rmilne  Twitter:   LinkedIn:   Facebook:   XING:

    Note: Posts are provided “AS IS” without warranty of any kind, either expressed or implied, including but not limited to the implied warranties of merchantability and/or fitness for a particular purpose.

    Thursday, February 14, 2013 2:15 PM
  • Hi Rhoderick,

    You're right of course, I totally overlooked that. The issue is described in http://support.microsoft.com/kb/2776259 and the fix is included in the list of UR6.

    Thanks for mentioning.


    Did my post help? Please use "Vote As Helpful", "Mark as answer" or "Propose as answer". Thank you!

    • Proposed as answer by Doug_TX Wednesday, February 27, 2013 9:53 PM
    Friday, February 15, 2013 7:14 AM
  • Ok,

    So RU6-v2 is coming.

    Looking on the bright side of things.

    Perhaps we will get a good install by the anniversary?

    I hope Microsoft starts thinking about releasing the security fixes separate from the bug fixes, soon.

    On the other hand, SP3 will be a possibility.

    I know SP3-Pre is not 2013 compatible, but SP3-RTMv2 is suppose to be.. maybe that would be the next logical upgrade.

    Launching 2013 "greenfield" looks better and better everyday.

    -- John Willis



    • Edited by jwillis84 Thursday, February 28, 2013 9:25 PM
    Thursday, February 28, 2013 9:21 PM