| Curtis Herrick 2004-02-08, 8:42 am |
| I've seen this error before, and as Larry suggested could be, I've
been using a pool. One simple way that I've seen to avoid the error
that you've been getting is to make sure that you have some
transmission retries configured on your channel (that goes to your
port to your orchestration). That way, if/when you have a pile of
files come in at once, they are "throttled" down to however many
threads that you may have to work with. The ones that come in as you
describe and have a transmission failure do fall into the Retry Queue
in that case and sit there as long as you've configured them to in
your channel retry settings. You will still see an transmission
failure error warnign in your event viewer, but the messages still (as
far as I've seen) get retried properly.
I should warn though, if you do this throttling technique, be wary of
orchestration transactions that can possibly dehydrate. If you have a
pile of files going through and some of them dehydrate, and more come
in and occupy all the allowed threads, you may have issues with
orchestrations not being able to rehydrate properly. (I have a post
on this issue myself right now / looking for help / info).
Hope this helps or is of some use.
- Curtis
|