We have a large built LabVIEW 2014 application running under Linux RedHat 6.4 (on a 12-core machine) which runs for up to a day, and then freezes at a seemingly random time. What I mean by "freeze" is that the front panel goes completely white, the CPU load drops from about 40% to 2% for the LabVIEW process (using the "top" command in a command window), writing to periodic log files stops, and TCP/UDP communications stop. The front panel window can be dragged around, and the rest of the machine appears to be unaffected. I don't see any evidence of a memory leak. We see this behavior on different machines.
The application makes heavy use of TCP and UDP communications, but we still see the freezing even after significantly reducing that usage. DAQmx is not used at all.
Checking on the NI website, I see that on Linux development systems, there is a dump log under <user>/natinst/LabVIEW Data/lvfailurelog, but our machine just has the RunTime engine and is running a built application.
Is there a way to get a dump log for a Linux built application?
Any other suggestions on how to diagnose the root cause of the freezing? Another approach that we will be looking at is checking the stability with older versions of the application.