Hello, I have recently taken over responsibility for an Exchange 2003 server in an education environment. They have been experiencing a problem with ActiveSync for several months. ActiveSync functionality sporatically stops and the following errors occur in the Application log: Server ActiveSync Error Number 3007: Exchange mailbox Server...response timeout...Exchange ActiveSync did not respond in a timely manner. Server ActiveSync Error Number 3014: The Exchange mailbox Server...has reached its timeout threshold. The mailbox server will be protected from new requests for [60] seconds. I can't find a pattern to when this occurs, but it happens mostly in the early hours of the morning, around 2 or 3am. A reboot of the server fixes the issue and mail starts syncing to phones again. OWA and general mail delivery are not affected. I have run the Exchange Best Practices Analyzer and the only issues were backups (which are on my list to fix), out of date NIC drivers, some minor DNS issues, file optimization, and IMAP4 fast retrieval. A consultant did change one of the session timeout parameters in the registry to fix the issue. It seemed to help for a few weeks, but now the issue is back again. Any suggestions on where to look next? Thanks for any help you can offer!

Is there any related error traces in the IIS log when the issue reoccurs? If there’s anti-virus software on the BE server, such symptom may appear (KB 827615) Please describe the exchange topology, is there any third party device between the mobile and exchange server, or between FE and BE (If you have a FE server)? Please check the server performance based on the data from the articles below: Optimizing Storage for Exchange 2003 Troubleshooting Microsoft Exchange Server PerformanceJames Luo TechNet Subscriber Support (http://technet.microsoft.com/en-us/subscriptions/ms788697.aspx) If you have any feedback on our support, please contact tngfb@microsoft.com

Run this as well and report back the results. https://www.testexchangeconnectivity.com/Default.aspxMark Morowczynski|MCT| MCSE 2003:Messaging, Security|MCITP:EMA 2K7,EDA Win 7,ES,SA,EA|MCTS:Windows Mobile Admin|Security+|http://almostdailytech.com

I found the following error repeating in httperr at the time the issue occurs: 443 HTTP/1.1 POST /Microsoft-Server-ActiveSync?User=xxxx&DeviceId=xxxxx&DeviceType=iPhone&Cmd=Ping - 1 Connection_Dropped ExchangeApplicationPool The only entries I find in the W3SVC logs are normal, like GET, SEARCH, PUT, SUBSCRIBE, UNSUBSCRIBE, POST, POLL, PROPFIND for Microsoft-server-ActiveSync. We are running Trend Micro Worry-free Business Advanced on the server. We haven't experienced the issue since the morning of 6/27, but the definition files have been updating. Does that point to AV? Unfortunately, whoever designed this put all the Exchange functionality on the same server. It's running the mailbox, OWA, etc. It is also running some extraneous web sites, AD, and DNS. Could this design be part of the problem? Thanks!

Do I need to run this when we are experiencing the problem? Thanks!

Please describe the exchange topology Are the mobile phones all Iphone? Do you have any mobile that’s running windows mobile system with the same symptom? “If response processing takes longer than the time-out value that is used by the Http.sys driver, the Http.sys driver destroys the original connection object. Then, the Http.sys driver writes a "Connection_Dropped" event message” --------Refer to <A "Connection_Dropped" event message is logged in the Httperr.log file on a Windows Server 2003-based computer that is running IIS 6.0> For now, the issue still points to timeout or packet loss. Please run the command below from internet to ActiveSync server Ping -l 1472 -f “ActiveSync server” Please increase the keep-alive time-out value on the servers, make the value consistent across all exchange servers (KB 813827) If you notice an immediate CPU spike for w3wp.exe process during the issue, please apply the Hotfix 941439 Possible cause: Apple shortened the request timeout from 4 mins to 30 sec, which means the device will only wait for 30 seconds for a request to be filled before re-sending the request. As you can imagine some users have large mailboxes that take a long time for ActiveSync to fulfill the request. This means the servers constantly resubmit the request to the mailbox if it’s not fulfilled within 30 seconds Workaround: Apply has published an Apple KB article (TS3398) to increase the timeout Resources: How To Troubleshoot Server ActiveSyncJames Luo TechNet Subscriber Support (http://technet.microsoft.com/en-us/subscriptions/ms788697.aspx) If you have any feedback on our support, please contact tngfb@microsoft.com