locked
Event ID: 26 "Connection to the Microsoft Exchange Server has been restored."

    Question

  • Recently our organization has upgraded from an ATT MPLS network to an ATT fiber using Sonicwall NSA 3500 and NSA 2400 firewalls.  We have site to site VPN tunnels using HUB and Spoke topology.  Within hours from the time we switched over to the new Fiber/Sonicwall network some users start getting the Event ID: 26 "Connection to the Microsoft Exchange Server has been lost. Outlook will restore the connection when possible." following instantly within less than a millisecond with another Event ID: 26 :Connection to the Microsoft Exchange Server has been restored."

    Our enviroment is as follows:

    Exchange Server:

    Windows 2003 R2 x86 (SP2)
    Exchange 2003 (version 6.5 Build 7638.2: SP2)

    Client Computer

    Windows XP Pro SP3
    Office 2003 (11.8330.8333) SP3
    McAfee V 8.7.0i Patch 4

    Exchange server configured to a 172.168.60.x IP with a subnet of 255.255.224.0
    Client is configured on a 192.168.80.x with a subnet of 255.255.255.0 subnet

    We ran every possible network test, regular ping between the workstations and the exchange server, ping with packets in the size of 1500 between the desktop and the server, no errors, no delays, our ping time is less than 16ms and no packet loss or lost packets.  It is important to mention that we get on an average a message every hour therefore 8 messages over the course of the day.  It is also important to mention that we have many other client server applications on the same computer as well network drives and none of them observe any sign of network disconnects.  The user productivity is not hindered but it is annoying.

    Another important fact this error message only comes in Cached mode, the moment I unchecked the Cached Mode the error message goes away.

    I have searched the net in and out, spoke to Sonicwall, told Sonicwall about some other instances wherby users changed the MTU on their routers and that fixed the problem, Sonicwall disagrees in this scenario and don't believe the MTU has anything to do with it.  It is also important to mention that out of each remote office out of 25 users only 6-8 users will experience this issue.  This is so weird and frustrating.

    Anyone has any idea?

     

    Thank you in Advance.

     

    OZ

    Thursday, June 09, 2011 5:20 AM

Answers

  • I have found the solution.

     

    The problem was residing within the Sonicwall routers access rules.  The access rules can be set to have a certain Connection Inactivity Timeout (minutes), which by default is set to 15 min. Therefore if a user did not send or receive an e mail for at least 15 min when they try to send an e mail or receive an e mail after the time out period they would get the disconnect from exchange server warning and immediately after that Outlook will reconnect.

    Initially I changed the timeout to 60 minutes and I reduced the amount of errors by 99%, I changed it to 480 minutes and now I have 0 errors.  It is important to mention that the only Access Rules that I changed are for LAN to VPN and VPN to LAN.  Doing so does not increase the vulnerability for hackers to hijack the connection as it is internally and via the VPN IPSEC tunnel.

     

    Thank you all for your help!

     

    OZ


    Thank You Sincerely, OZ
    • Marked as answer by WizardsofOZ Wednesday, June 29, 2011 1:40 AM
    Wednesday, June 29, 2011 1:39 AM

All replies

  • Did you checked it?
    http://support.microsoft.com/kb/890352 
    Gulab | MCITP: Exchange 2010-2007 | Skype: Gulab.Mallah | Blog: www.ExchangeRanger.blogspot.com
    Thursday, June 09, 2011 7:19 AM
  • Dear Gulab,

     

    The Event ID 50 and 26 that you describe are errors that are related to the Microsoft Server and they are completely different from the Event ID: 26 that I get on the Microsoft XP Pro desktop.

     

    Thank you for your help!

     

    OZ

    Thursday, June 09, 2011 5:31 PM
  • Hi,

     

    What is the issue you are encountering? Does Outlook lost connection for just one time and then the connection got restored immediately? Or the Outlook connection status is not stable after you changed the network connection, and prompt lost connection repeatedly ever one hour?  

     

    On your Exchange server, is there any event log related to the problematic user account, or around the time issue occurs? If so, please paste the exact error message and the event ID number.

     

    How does the problematic user connect to your office network? Wireless or cable? Is RPC over HTTP enabled or not?

     

    -Fiona


    Please remember to click “Mark as Answer” on the post that helps you, and to click “Unmark as Answer” if a marked post does not actually answer your question. This can be beneficial to other community members reading the thread.



    Friday, June 10, 2011 4:30 AM
    Moderator
  • I have found the solution.

     

    The problem was residing within the Sonicwall routers access rules.  The access rules can be set to have a certain Connection Inactivity Timeout (minutes), which by default is set to 15 min. Therefore if a user did not send or receive an e mail for at least 15 min when they try to send an e mail or receive an e mail after the time out period they would get the disconnect from exchange server warning and immediately after that Outlook will reconnect.

    Initially I changed the timeout to 60 minutes and I reduced the amount of errors by 99%, I changed it to 480 minutes and now I have 0 errors.  It is important to mention that the only Access Rules that I changed are for LAN to VPN and VPN to LAN.  Doing so does not increase the vulnerability for hackers to hijack the connection as it is internally and via the VPN IPSEC tunnel.

     

    Thank you all for your help!

     

    OZ


    Thank You Sincerely, OZ
    • Marked as answer by WizardsofOZ Wednesday, June 29, 2011 1:40 AM
    Wednesday, June 29, 2011 1:39 AM