06-07-2010 11:08 AM
Sorry for no response over the weekend. I don't check this forum or even think about work on the weekends. I only touch my home computer to check my personal email.
How are you determining that your database remains open? Your code looks OK. You don't need 5 second delays, 100 mS is plenty enough. I still insist that problems ocurring only when running at full speed indicates a need for a delay somewhere. Have you tried putting a delay between each and every DB function call? I know this is tedious, but that is the only way to determine where it is needed.
My only other advice is to get rid of that sequence structure and re-write your vi using a state machine. Create a state for each true/false case, and use the next state to jump to the proper case. Maybe with a state machine, your problem may become obvious. Besides, it is makes for a much better vi in terms of readability and scalability.
06-07-2010 11:36 AM
tbob,
I'll try adding delays before each vi call, but even if that works, there's a problem with the vi's that needs fixing.
I know the database connections are open because I can see the connections when logged into MYSQL on a sever we have connected to the test stations. The vi I've attached connects to a unique database which is not accessed at any other time during testing. The name of that database is a clearly seen in the list of connections shown by MYSQL's "show processlist;" And there's one item for every call of the vi - one for each DUT, for each test station.
Thanks for your continued help.