views:

166

answers:

2

I have been getting these intermittent ColdFusion Database connection reset errors and was wondering if anyone had experience with this and had a particular solution that worked?

Here is the error:

Error Executing Database Query.[Macromedia][SQLServer JDBC Driver]A problem occurred when attempting to contact the server (Server returned: Connection reset). Please ensure that the server parameters passed to the driver are correct and that the server is running. Also ensure that the maximum number of connections have not been exceeded for this server.

This doesn't happen with any particular query, the code breaks in different queries every time, returning a SQLState error 08s01. These query's logic are fine, no logic errors etc. I checked the network logs and there were no database server connection refusals at the time of the error. Once the first error occurs, it keeps happening for no more than a minute or so at random times of the day, every few days. I've googled this thing and so far anyone that has had this issue was only on CF6 or 7, which the fixes coldFusion put out are only for CF6 or 7.

Server configuration wise:

  • The ColdFusion server is version 8
  • The database server is SQL Server 2005 Standard
  • The database connections allowed setting is set to unlimited on both SQL Server and ColdFusion

Any help would be greatly appreciated, Thanks!

A: 

I'd personally try to update JDBC driver ( http://sourceforge.net/projects/jtds/ ) and add Validation Query if not already existing. I used once tool which was constantly checking TCP connections between servers, detecting even milisecond gap, check this tool http://www.colasoft.com/ping_tool/ .

zarko.susnjar
Thank you for the feedback. There definitely appears to be a network issue. I analyzed the TCP connections and it retrieved a TCP packet with a reset flag. I found that the error occurs almost a quarter after 4PM and 12 AM (give or take 10 minutes). While i was observing the packet analyzer, my remote connection between the server and my computer was broken right when the error occured, so that tells it has to be a network issue. What could cause this? I also noticed that during non-error times, a most of the network packets are not passing CHECKSUM, could this play a roll?
Gavin
Dunno really, I assume there are other scheduled tasks on those machines or network architecture. 8 hours gap reminds me at our company network when at working time start/end whole network slows down for 10-15 minutes :)
zarko.susnjar
A: 

Have you already resolved the issue?

I'm having the same problem on CF8 VPS, and after doing some research online I found this post: http://www.mail-archive.com/[email protected]/msg302264.html It says it's related to some features on the network card and something on the SP2 for Win2k3 that messes it up. Don't know if that applies to you but just in case.

oarevalo