I've developed a C++ application for Solaris that reads and writes from a remote Microsoft SQL Server 7.
It utilizes FreeTDS and unixODBC to communicate with the SQL Server.
On my customer's development system and network, the application worked without issue.
However, after moving their SQL Server machine onto their production network and installing my application on their production system, they are now encountering problems.
After some investigation, I've found that my application is able to perform reads but encounters a "communication link failure" error upon write.
The customer has connected to the SQL Server using a laptop with an ODBC client and is able to read and write without a problem.
The customer claims that both networks use an identical switch with identical configuration, no firewall is involved.
This leads me to believe there is differing network configuration between the development and production Solaris machines, or there is in fact some intricate difference between the network set ups.
I'm quite stumped at this point. Any insight would be appreciated.
Bear in mind that since this is a production system, I do not have free reign.