Ok, so I’ve finally sucumbed to the Twitter phenomenon and I’m now twittering, or is that tweeting? Eitherway, you can follow me over at http://twitter.com/nickheppleston if you are so inclined…
Fast, Efficient Message Archiving for BizTalk with the BizTalk Message Archiving Pipeline Component. Download a Free 14-Day Trial!
A partner in our message exchange system, has configured their application to abort sending messages to us, if 10 message attempts fail (5743 events) within a 5 minute period. What is unfortunate about this, is they have Biztalk 2004, and if allowed, their Biztalk system would have retried sending the message(s) in 5 minutes, and likely submitted the message. I estimate 99% of the time, we’re running into network time-outs that last for less than a minute, such as in the case of our clustered SQL server back-end failing over (30 sec). Worse still, for every message attempt that fails they send an email message to somewhere in the range of 30 or so people.
I have been insisting that their system should accommodate a network delay of 30 sec (we are going througha WAN after all). Do you have any good links of information which could outline what a well written application should accommodate as far as sending messages across a WAN like we are?