Real-time system booting into safe mode

[cirrusio]

Hello all,

I am having an incredibly frustrating problem.  My RT PXI chassis has decided that something has gone wrong.  I am not sure what that could be, but now it boots into safemode attempting to reboot three times.  I have never been able to retrieve the error log via MAX (I get time outs), but if I go into the physical drive via ftp and poke around in the errorlog.txt, I find the most recent error:

10/29/2015 03:25:09.153 0 -2147220694 000 Incompatible schema detected, but no specific DD can be identified as the source.
Please refer to "c:\ni-rt\system\mxsSchemaError.log" file for more information.
CmxsNeoStorage::ValidateDBStructuralIntegrity()

If I go to the file it refers to, I find the following error:

[10/29/2015 03:25:09] *** COLLISION DETECTED *** between object with NeoID:0 of class Database_Header located at 0 (256) and free-list 1 (28344)
[10/29/2015 03:25:09] Incompatible schema detected, but no specific DD can be identified as the source.

Anyone have any thoughts on this?  I have tried uninstalling all of the software and then reinstalling.  Everything seems to be ok until I install DAQmx drivers and there dependencies.  But, I don't think this is directly related to DAQmx.  I think it is ties up with one of the dependencies.  Any input is helpful (although I want to avoid having to format as I don't have a recent copy of the MAX configuration for the system).

Cheers, Matt

[cirrusio]

Looks like a corrupt databse :(...

Likely will have to reboot the configuration.

[Bob_Schor · Accepted solution]

Have you changed LabVIEW versions recently (say, from 2014 to 2015)?  You probably need to check/update the software on the PXI.

If all else fails, it is not all that difficult to "start over" with a reformatted hard drive on the PXI and install a fresh RT system, particularly if you have a controller that doesn't boot from a floppy (don't ask -- we have lots of 8176's).

Bob Schor

[cirrusio]

Nope.  This seemed to happen magically (as all of these problems seem to arise).  This is the second time in a year this seems to have happened and I think that I must have something wrong with my network settings on my development computer because I have been unable to import a copy of the MAX configuration file.  Soooo, given this, starting over is an abominable option.  My most recent backup of the file is stale and so I am going to have to rebuild the configuration file from my notes.  This may be a horrible lesson learned about using the MAX configuration file.  Ugh...

[cirrusio]

Bob,

I marked your answer as the correct one, but I wouldn't say this is a particularly great one.  This exposed a major weakness in using the MAX API to configure tasks.  I will probably move my configuration into the code itself (as well as a custom INI file for this) so that I don't have to deal with this in the future.  

Thanks for taking a look at this.

Cheers, Matt

[Bob_Schor]

Matt,

     Sorry to hear your system is being flaky.  Have you tried creating and saving your Tasks in LabVIEW Project?  When I first introduce students to DAQ with LabVIEW (using a USB-6009), we "play" with MAX and I show them how to create and save a Task (in MAX) that they can use to store the details of the configuration.  But I discovered when configuring my Real Time Tasks on a PXI, this seems to associate the Tasks with the PXI, not with the Project and the code, so I switched to specifying the Tasks in Project.  Now I can develop on one machine, save the Project in Subversion, Update on my "target" machine, and the Task is Ready to Roll, even though the PC and PXI are different.

Bob Schor