Labview Crashes

[Martin_Kroner]

I use a large Labview VI with lots of subvis and Vis running in parallel doing data aquisition (daqmx, visa, ...). Since a few weeks, the Vis crashes after hours or days of running with i.e. the following log message:

####
#Date: Di, 16. Sep 2014 09:31:33
#OSName: Windows 7 Enterprise Service Pack 1
#OSVers: 6.1
#OSBuild: 7601
#AppName: LabVIEW
#Version: 13.0.1f2 64-bit
#AppKind: FDS
#AppModDate: 04/02/2014 13:00 GMT
#LabVIEW Base Address: 0x0000000140000000

starting LabVIEW Execution System 2 Thread 0 , capacity: 24 at [3493697493.54960490, (09:31:33.549604893 2014:09:16)]
starting LabVIEW Execution System 2 Thread 1 , capacity: 24 at [3493697493.54960490, (09:31:33.549604893 2014:09:16)]
starting LabVIEW Execution System 2 Thread 2 , capacity: 24 at [3493697493.54960490, (09:31:33.549604893 2014:09:16)]
starting LabVIEW Execution System 2 Thread 3 , capacity: 24 at [3493697493.54960490, (09:31:33.549604893 2014:09:16)]
starting LabVIEW Execution System 2 Thread 4 , capacity: 24 at [3493697493.54960490, (09:31:33.549604893 2014:09:16)]
starting LabVIEW Execution System 2 Thread 5 , capacity: 24 at [3493697493.54960490, (09:31:33.549604893 2014:09:16)]
starting LabVIEW Execution System 2 Thread 6 , capacity: 24 at [3493697493.54960490, (09:31:33.549604893 2014:09:16)]
starting LabVIEW Execution System 2 Thread 7 , capacity: 24 at [3493697493.54960490, (09:31:33.549604893 2014:09:16)]
Thread consumption suspected: 1 Try starting 2 threads
starting LabVIEW Execution System 2 Thread 8 , capacity: 24 at [3493698812.61505130, (09:53:32.615051270 2014:09:16)]
starting LabVIEW Execution System 2 Thread 9 , capacity: 24 at [3493698812.61505130, (09:53:32.615051270 2014:09:16)]
Thread consumption suspected: 1 Try starting 1 threads
starting LabVIEW Execution System 2 Thread 10 , capacity: 24 at [3493699331.02170230, (10:02:11.021702290 2014:09:16)]
Thread consumption suspected: 1 Try starting 1 threads
starting LabVIEW Execution System 2 Thread 11 , capacity: 24 at [3493702067.81423810, (10:47:47.814238072 2014:09:16)]
Thread consumption suspected: 1 Try starting 1 threads
starting LabVIEW Execution System 2 Thread 12 , capacity: 24 at [3493708191.86751370, (12:29:51.867513657 2014:09:16)]

<DEBUG_OUTPUT>
16.09.2014 16:49:19.082
DAbort 0xD3A61F82: Attempted to assign event to unknown loop.
c:\builds\penguin\labview\components\LVManager\trunk\13.0\source\eventq.cpp(169) : DAbort 0xD3A61F82: Attempted to assign event to unknown loop.
minidump id: 22ae8d9c-6c19-4f38-bbf1-a59401ca6c01
$Id: //labview/components/LVManager/trunk/13.0/source/eventq.cpp#6 $

</DEBUG_OUTPUT>
0x0000000140349BD9 - LabVIEW <unknown> + 0
0x0000000180016685 - mgcore_SH_13_0 <unknown> + 0
0x00000001800171F0 - mgcore_SH_13_0 <unknown> + 0
0x0000000141F205E6 - LabVIEW <unknown> + 0
0x0000000141F20824 - LabVIEW <unknown> + 0
0x0000000141F20A7A - LabVIEW <unknown> + 0
0x0000000141EEE539 - LabVIEW <unknown> + 0
0x0000000180039B39 - mgcore_SH_13_0 <unknown> + 0
0x000000018003A431 - mgcore_SH_13_0 <unknown> + 0
0x0000000141F20AFC - LabVIEW <unknown> + 0
0x0000000141FC5474 - LabVIEW <unknown> + 0
0x0000000141F1FD72 - LabVIEW <unknown> + 0
0x0000000140058287 - LabVIEW <unknown> + 0
0x000000014005881A - LabVIEW <unknown> + 0
0x00000001421872DB - LabVIEW <unknown> + 0
0x0000000076E259ED - kernel32 <unknown> + 0
0x0000000076F5C541 - ntdll <unknown> + 0

has anyone seen something like this? Do you have an idea what can cause such a behaviour or can you suggest a good ways of debugging this?

Thanks, best wishes,

Martin

[Norbert_B]

Martin,

since this issue is "new" to your application: What did you change in your code a few weeks ago?

If you made no changes to the application itself: What did you change on the system a few weeks ago?

Remember that there could be auto-updates of software on the system (e.g. Windows hotfixes, Anti-virus-software,...) which can make significant changes without notifying you.

Norbert

[Martin_Kroner]

Dear Norbert,

thanks for the reply! 

Unfortunately, I do constantly small changes to the code which I do not allways track carefully! I know, very bad programming style. Also I upgraded the Labview version to 2013 recently.

Apart from Labview there are of course all the windows updates and I also regularly update hardware drivers. 

I monitored the runtime behaviour of the VI using the windows tools and it does not look like there are any memory problems.

Best regards,

Martin

[Norbert_B]

Do you use timed structures? If so, do you use different (read: unique) priority levels for each timed structure?

Norbert

[billko]

---

@Martin_Kroner wrote:

Dear Norbert,

thanks for the reply! 

Unfortunately, I do constantly small changes to the code which I do not allways track carefully! I know, very bad programming style. Also I upgraded the Labview version to 2013 recently.

Apart from Labview there are of course all the windows updates and I also regularly update hardware drivers. 

I monitored the runtime behaviour of the VI using the windows tools and it does not look like there are any memory problems.

Best regards,

Martin

---

Please tell me this is an engineering tool and not for production.  With all those incremental changes and no regression testing, who knows what's going on now?  Hopefully you can correlate your problems with a change to the coding environment - and in the meantime learn that "winging it" is a bad thing if you want dependable code for anything more than a simple engineering tool.  (Really, an upgrade to the development platform with no testing???)

[Martin_Kroner]

I am very sorry, I should have been more precise. I am not an engineer. I am not developing any commercial product! I am an experimental physicist and I use Labview to controll my experiement and to do data aquisition. Since my experimental setup changes once in a while I also have to adapt the code. Since I am the only person using this code I do not to carefully keep track of the changes.

I did upgrade to the lates labview version because I wanted to use some of its new features. I was not aware of any changes I had to do in my code. I always thought if there are no errors, it will run... Well apparently not!

[Norbert_B]

As the code obviously derives from an older LV version: Did you run a mass compile on the VIs? Did you get any warnings/errors?

Did you change your computer as well to e.g. a more powerful (read: more cores) machine?

Norbert

[Martin_Kroner]

Dear Norbert,

thanks for your help! I am not using timed structures. My code is essentially a large event structure in wich certain events are created (either by the user or by the code). I use the timeout event (generated every 5ms) to execute code that I want to be executed all the time (i.e. data loging). 

Martin

[billko]

---

@Martin_Kroner wrote:

I am very sorry, I should have been more precise. I am not an engineer. I am not developing any commercial product! I am an experimental physicist and I use Labview to controll my experiement and to do data aquisition. Since my experimental setup changes once in a while I also have to adapt the code. Since I am the only person using this code I do not to carefully keep track of the changes.

I did upgrade to the lates labview version because I wanted to use some of its new features. I was not aware of any changes I had to do in my code. I always thought if there are no errors, it will run... Well apparently not!

---

I'm sorry, too!  The vast majority of users who post here are either engineers or "engineers-in-training".  I couldn't figure out why you would do this in a production environment and I wanted to try to catch an engineer before they strayed too far down the dark path.  😉

[Martin_Kroner]

I am to slow, sorry..

I did not compile any of my code. It runs in the "development mode" of labview.

I did indeed also change the computer, but this was maybe 6 months before the code became unstable.