none
Outlook 2010 Connected or Not

    Dotaz

  • Hello,

    we have a few users that reporting that after working with Outlook 2010 for a while Outlook suddenly stops sending and/or receiving e-mails. In the case of sending the e-mail sits in the Outbox.

    The status bar says connected but when you click "Send All" the error message is that Outlook needs to be online and connected.

    There is no other error message or any other indicator of what is going on.

    The only two solutions are to close Outlook and restart it or click on "Work Offline" and then reconnect immediately.

    Users

    I would appreciate any help on this problem.

    Regards

    Sascha

    22. února 2012 17:28

Odpovědi

  • The hardest kind of problem to figure out, the random problem. :-(

    You might need to look at network stuff, NIC drivers, switches etc. If the software layer is good, you need to look further down. Very hard if you can't repro the problem consistently.

    Good luck.

    16. března 2012 16:34

Všechny reakce

  • Is there a load balancer being used? If so, and if you bypass the load balancer (use a hosts file) do the isssues go away?
    22. února 2012 22:11
  • Hello,

    no load balancer is used. The CAS is published using TMG.

    Some of the affected users reported that the problem occurs around the same time (between 5pm and 6pm). However I am not able to confirm if this is accurate for everyone or just the impression of a few users.

    Regards

    Sascha

    23. února 2012 6:16
  • So did you configure TMG by following our whitepaper? Is it using IP affinity? Can you check the publishing rule to see if any of the settings have been changed from the default? Or put another way, try creating a new rule, following our guidance, and see if that works better? It sounds like a timeout somewhere, but it could be hard to track down, working by process of elimination by re-doing things might be a good way to find the bit that is causing you trouble.
    23. února 2012 17:14
  • Hello,

    I tried to find out more information and common configuration between the affected systems and users but no luck yet. I will try over the weekend going through the White Paper step by step but the entire setup is rather simple at the moment. One TMG, one CAS, two MBX servers, two Hubs.

    And I would say only about 5% of the users are affected.

    Regards

    Sascha

    2. března 2012 13:09
  • If there's only one CAS it's not a load balancing issue. It sounds more like some kind of network timeout issue. Are there any other network security type devices in play? Something that could be affecting a timeout? Or any changes made to TMG to timeout sessions? If so, or if you don't know, re-configure TMG to the defaults and go from there following the guidance.
    2. března 2012 16:44
  • If there's only one CAS it's not a load balancing issue. It sounds more like some kind of network timeout issue. Are there any other network security type devices in play? Something that could be affecting a timeout? Or any changes made to TMG to timeout sessions? If so, or if you don't know, re-configure TMG to the defaults and go from there following the guidance.

    No security devices in play here. The timeouts have not been change in TMG. I recreated all the policies, but the problem still occurs. Unfortunately I am still not able to reproduce the problem consisently. it appears to be totally random.

    16. března 2012 12:34
  • The hardest kind of problem to figure out, the random problem. :-(

    You might need to look at network stuff, NIC drivers, switches etc. If the software layer is good, you need to look further down. Very hard if you can't repro the problem consistently.

    Good luck.

    16. března 2012 16:34
  • Hello,

    we set up all publishing rules on a separate TMG and now it seems to work.

    Thank you for your support.

    27. března 2012 6:50