A few users of Exchange 2010/2013 may be experiencing a very strange issue that causes mail flow to cease up at seemingly random times during the day; if you can confirm that this is not backpressure, it is a known bug. Unfortunately, this bug has not yet been addressed by Microsoft, so many users have resorted to either simply coping with the issue (mail eventually starts flowing again, and all mail that is sent during the interval of downtime are still processed when the Microsoft Exchange Transport service stops hanging).
Check out these links to find out if you are having the symptoms that the fix will address:
The fix that I wrote uses Cygwin, and is launched by task scheduler every seven minutes (well, task scheduler actually launches a batch script that then calls “bash script.sh”, but we’ll get to that a bit later).
Starting with the setup first, get the script here. Next, if you don’t already have it installed, download and install Cygwin with the extra package “nc” or “netcat” (just select it from the installer). Make sure you add the Cygwin bin folder to the system path. Now write a batch script with the following:
bash.exe /path/to/script.sh exit 0
Now add the batch script to task scheduler, and set it to run every 5 minutes. Enjoy working mail flow without the need for manual MSExchangeTransport service restarts.
More on how the script works a bit later; I just want to publish the post as quickly as possible in order to help anyone who is experiencing this issue.