Why is the Windows NI Variable Engine Service unexpectedly terminated ?

[Hammerson]

I am using a NI USB I/O card model 6525.

The operating system is Windows XP Pro SP2
Without warning, something shuts the software service off, and the connection to the I/O card is lost.

The Windows "Event Viewer" just says "The NI Variable Engine service terminated unexpectedly".

The log also states that "Source: Service Control Manager" and that the "Event ID: 7034".

A little later in time, when rebooting the PC to try to regain communication to the card, the Event Viewer logs Errors.

The text for these errors says:

Err 1: "viceTable.cpp:294 - Unrecognized device 3923:71DF"

Err 2: "[usb6xxxk][[usb6xxxLoader]] Fatal: .\source\De..."

For both of these errors "Source: NIPALK" and "Event ID: 12288".

Any ideas what I should be looking at ?

Is it hardware or software ?

Thanks a lot.

H

[Brad_K]

Hi Hammerson,

The two usb6xxxk/DeviceTable.cpp events can be safely ignored. These were due to an overly strict error check, which was later removed in NI-DAQmx 8.7. They are also unrelated to the NI Variable Engine service event.

What versions of NI-DAQmx and LabVIEW do you have installed? By any chance, are you using Zone Alarm firewall software?

Brad

[Hammerson]

Thank you for reply.

I have NI-DAQ 8.6.

There is no version of LabView installed.

I am using the USB 6525 card alone.

I am reading/writing I/Os by using OPC tags.

There's MacAfee antivirus on this computer, but no firewall software.

Any thoughts why NI Variable Engine service shuts down on occasion ?

[Brad_K]

Hi Hammerson,

The information that you are using OPC with the USB-6525 is very helpful, because the NI Variable Engine service is where DAQmx tasks execute when you do OPC with DAQmx. (If you weren't using OPC or shared variables, disabling the NI Variable Engine service could be a reasonable workaround.)

Does the crash seem to be correlated with system uptime or the amount of time your application has been running? (E.g. it crashes right after boot, or it only crashes after running for 3 days?) Does it seem to be related to specific operations performed by your application? Or is completely random?

Since you are using NI-DAQmx 8.6 (not 8.6.1 or later), there is a chance that you are encountering the issue described in Product Notice: Software Problem May Corrupt Configuration in MAX (NI-DAQmx). Even if you're not, NI-DAQmx 8.6 is from 2007, and you might be running into a bug that has already been fixed in the three years since NI-DAQmx 8.6 released. Upgrading to the latest version of NI-DAQmx would not be a bad idea.

If it doesn't take too long to reproduce the crash, it would also be nice if you could post a "crash dump" file, because that would identify which part of the code crashed. Windows XP includes a program called drwtsn32 that can be configured to save a minidump when an application crashes. (If you have a debugger such as Microsoft Visual Studio you can also use that.) However, if this is not feasible or you can't get it to work, don't worry about it.

Brad

[Hammerson]

Brad.

The first crash happened about 4 months ago.

Since then, we have seen about three more crashes.

This is not very frequent, and to complicate things even more: This system is in the field in a client's manufacturing facility.

His system is in use 24/7, so I can't really do anything that will potentially interrupt his production.

I have not been able to find any repeatable actions that may have caused the crashes.

An interesting fact is that we installed this system late last year, and it operated for a long time without any isses at all.

I will take a look at (I do have remote access to this PC) what I can do to collect a "crash dump".

Thanks again !

H