cRIO communiciton fell

[reuven]

Hello all CD community!
yesterday , just 1.5 hr's from stopping with the build we had a strange thing happening with our robot,
our team of programmers burned a code successfully on the Crio, and the code ran good for like 5 min's
after that we lost communication with the RIO,
the interesting part is, that this is not a router problem, iv'e manged to PING both the router itself and the camera but not the RIO.
after turning on and off the robot we had communication again, but the same problem occurred again,
i have to add that the same code ran successfully all day long before we burning the code (building and running as start-up).

can anyone think what the problem is? and more then that how can i solve it (i don't have this RIO with me cause it was shipped with the robot, but i do have a second RIO which i use for the time before the regional)

thanks in advance

[Tottanka]

NO IDEA.

[Mark McLeod]

Check the memory % for the cRIO displayed on the Driver Station Charts tab.
A memory leak could eventually cause a crash.

NetConsole might give you useful information, if errors are generated prior to the code crash.

[mesamb1]

While I would go as far as saying memory leaks are rare in Labview, I would follow marks advice, in the console look for a watchdog error. It would sound to me like your watch dog isnt gettting fed, he flashes an error on the driverstation and the driver is so busy watching the robot that they miss it. If that is the case, I would disable the watchdog and see if that fixes it

[Greg McKaskle]

If this happens again, please cable to the cRIO, bypassing the DLink wireless, and see what state it is in. Use the DS to inspect the cRIO CPU usage,memory, error messages, etc. This should help isolate whether the comms is failing on the cRIO, or if the radio is not letting traffic through to the cRIO.

Greg McKaskle