FPGA image checksum mismatch and cRio image checksum mismatch

[SuperTrooper249]

We had a practice competition yesterday and immediately encountered a fatal error that we couldn't overcome. Our robot runs Java. We placed our robot onto the field, connected to the robot to the Field Management System (FMS) and plugged our driver station laptop into the ethernet. The match started and our robot ran the autonomous routine, but as soon as the game entered teleop mode, the robot wouldn't respond. We checked the output on the netbeans console later, and found two error statements printed. They were something about an "FPGA image checksum mismatch" and a "cRio image checksum mismatch." Prior to these error statements, we saw debugging print statements from autonomous mode, but no more print statements after, suggesting the that the cRio had frozen up.

Here are the pertinent facts:

• This behavior occurred in multiple matches. We power cycled and reload the code.
• We had previously tested our code extensively off the field with no issues.
• After returning to home base, we set up an FMS Lite system to fully emulate the environment that was used at the competition. We put the robot into the same configuration as it was in the competition but it successfully operated in autonomous and teleop mode. We could not reproduce the error.

We're totally stumped, the robot works just fine at home. We have no way of knowing if this problem will arise again at the official competition. Does anyone have any idea what would be causing this?

[KHartley]

Hello Super Trooper,

Do you have the lastest version of the cRIO imaging tool? Your cRIO should be programmed with image v47.

Make sure you are using LabVIEW FRC 2013 to program. You can check this from the getting started window.

Thanks,

Kyle Hartley

[SuperTrooper249]

Yes, used FRC's 2013 cRIO imaging tool to image it with version 47.  Since we use Java, we're using Netbeans, not LabVIEW.  We made sure to update the netbeans plugin repository so it's running the latest version of the WPI Java plugins.

Do you have any idea why such a problem would show up at a practice competition but not be reproducible anywhere else?

[Kevin_F]

To give you a little more information on those errors - Those errors should have no affect on the robot working or not. They are part of the original scheme for the FMS to validate the software installed on the cRIO. After the second year, the FMS stopped updating their expected values. This means that the FPGA image and the "cRIO" (FRC network communications) checksums will always fail on the field. The FTA just ignores those errors on the FMS.

If you can run in FMS Lite and in practice mode in your shop I wouldn't expect any issues at an event.

[SuperTrooper249]

Ok, that's encouraging, but still not a guarantee.  Do you think, then, that something was wrong with the configuration of the FMS Lite system at the practice competition?  Becuase our program croaked immediately after printing those two errors.  But you say that those two errors shouldn't actually stop the robot from functioning?

[Kevin_F]

Those errors definitely shouldn't stop the robot from functioning and I think are probably a red herring in this case.