We have tried and searched for a solution but we are running out of time and have to find another way to give our employees access to our mails with IMAP. Somewhere in the topic I’m referring to, someone mentioned following
If we would read the instructions regarding IMAP standards (http://tools.ietf.org/html/rfc3501) than following is very important: On section 2.3.1.1 it says:Bill48105 wrote:chrisso think you missed the point. Perhaps what you tell us is a CLUE in the underlying cause but the problem existed before Outlook 2010.. In fact with 2007 as others reported. So the problem is not 2010. It seems what you point out gives hints & confirms what we've suspected for a while now so it could still be useful in tracking down the root cause. The UID's get out of order or are different in terms of what Outlook thinks they should be. For example during the move to trash outlook might assume the UID remains the same but hmail changes it during the move. Or vice versa. We've never been able to reproduce it consistently enough to track what is happening so maybe what you post will help with that. If we can reproduce the problem we can try to fix it so thanks for the info.
Btw it is entirely possible the issue is a race condition. If you get time you might try one of the latest experimental builds (link in my signature) even if on a test bed. During development of STARTTLS we found a few spots in code where there were race conditions that resulted in corrupt emails and crashes. We have confirmation from a few people that the changes have solved crashes that happen in B9150 etc so you might give it a try. It's possible the threads get out of sync resulting in UID's being out of order.
Bill
We find it hard to believe that no one can simulate the problem while we can do it every time, again and again. It’s just not workable and if this issue cannot be solved, we will have to look for another mailserver.The unique identifier of a message MUST NOT change during the session, and SHOULD NOT change between sessions.
The question is: is someone working on this problem?