homerjones1941 schrieb am 30.03.2012 um 08:12 (-0000): > Server: Windows XP with all updates current.Database: Firebird 2.1 > Client 1: Windows 7 - 64 bit Intel I7 - Wireless connection > Client 2: Windows XP 32 bit AMD Dual Core - Wired connection (this > is not the server). > Language: Delphi 2007 - Data access via IBDAC components (devart). > Application: Customer lookup and support data.The program normally has > only one table open using an IBDAC Query component. The data is > displayed one record at a time (not a grid). There are a good number > of other tables, but they are not open when this problem occurs. > Condition/Symptom: Within 15 or 20 seconds of opening the database, > the program freezes for a short period before returning the error: > "... an existing connection was forcibly closed by the remote host."
Sounds like a Microsoft Windows TCP level problem; you probably know that much already. Troubleshooting: Connection Forcibly Closed http://msdn.microsoft.com/en-us/library/ms187005.aspx > This happens while navigating from one record to the next. It also > happens in the same amount of time when sitting idle. Examination > of Firebird.log shows multiple coinsurances of : > MYCOMPUTERNAME (Server) Thu Mar 29 18:11:06 2012 INET/inet_error: send > errno = 10054 > MYCOMPUTERNAME (Server) Thu Mar 29 18:19:03 2012 INET/inet_error: read > errno = 10054 > Assumptions:Since 10054 = "Connection reset by peer", I assume the > origin of both messages is occurring on the server, and they are both > reporting same condition. > Question:If my assumption is correct, is there something I need to do > with Firebird that will prevent/minimize this? Can't think of anything, but I'm not an expert. Rather, try unsetting the SynAttackProtect registry name as suggested by the MSDN article. > I plan to look at Firewall and FileZilla settings, Should I check > anything else? Firewall, maybe; not sure how FileZilla would be related. > More background:I have a server application (also written in Delphi > using Indy) running on the same server in the same domain. It has been > running flawlessly for many months. This same machine, router and > switch has been my web server for over a year. The clients for that other server might just not happen to run into or trigger or detect this error condition. -- Michael Ludwig
