|
From: Ivan V. <iv...@gm...> - 2009-09-16 18:44:44
|
2009/9/16 Laurent Danesi <ld...@sm...>: > Hi Ivan, > > I'm preparing a new QuickFixJ release containing a new feature called forceResync which seems exactly what you are trying to do. > I think that I will integrate most of the changes in the trunk soon (end of this week). Hi, Thank you for your response, but in parallel I've talked to the counterparty people and they have suggested a different solution that, while I don't really know enough to claim is perfectly correct, at least appears to work. They've asked us to add ResetSeqNumFlag to the login message (in toAdmin) and I'll see if the application survives the next 24h of continuous operation. They've also asked us to reduce the reconnect interval to 120s (was 1s). I'm curious to learn if this approach has some nontrivial consequences. We are concerned about not losing any potential data records. > What are your time constraints? If that doesn't work, we would need to have an acceptable solution within two weeks. |