LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Different results when I use the debug trace mode

Hi,

I'm using LabView to configure test subject equipment via SNMP messaging. During the process and before each reading I need to re-calibrate my spectrum analyzer for things like center frequency and channel bandwidth. To do this I have a function which send out the SNMP get request to an SNMP server.

Currently I have the same vi appearing in a sequence loop a couple of times. When I run the program in debug mode everything is good. The first step in the sequence gets me a valid center frequency and the second delivers a valid channel bandwidth. However if I turn off the little light bulb (debug tool) I get a valid center frequency but the channel bandwidth returned to the calling vi is the center
frequency.

I've looked at the messages comming back from my SNMP server and they are correct. I've tried setting booleans which would hopefully ensure the correct message is delivered but no.... I even tried timers all over the place to slow things down.

I'm running LabView 5.1 on an NT box with a P3 500 processor.

This problem may have something to do with the same vi being called in the same sequence loop more than once??

Any help would be apreciated.

Regards
Mike Gaskin
0 Kudos
Message 1 of 5
(2,523 Views)
You could make it reentrant. And make sure that they execute in right order.

"Michael Gaskin" wrote in message
news:381882F3.7626F697@fox.nstn.ca...
> Hi,
>
> I'm using LabView to configure test subject equipment via SNMP messaging.
During the process and before each reading I need to re-calibrate my
spectrum analyzer for things like center frequency and channel bandwidth.
To do this I have a function which send out the SNMP get request to an SNMP
server.
>
> Currently I have the same vi appearing in a sequence loop a couple of
times. When I run the program in debug mode everything is good. The first
step in the sequence gets me a valid center frequency and the second
delivers a valid channel bandwidth. However if I turn off the little light
bulb (d
ebug tool) I get a valid center frequency but the channel bandwidth
returned to the calling vi is the center frequency.
>
> I've looked at the messages comming back from my SNMP server and they are
correct. I've tried setting booleans which would hopefully ensure the
correct message is delivered but no.... I even tried timers all over the
place to slow things down.
>
> I'm running LabView 5.1 on an NT box with a P3 500 processor.
>
> This problem may have something to do with the same vi being called in
the same sequence loop more than once??
>
> Any help would be apreciated.
>
> Regards
> Mike Gaskin
>
0 Kudos
Message 2 of 5
(2,523 Views)
I tried that and then went on to actually rtfm and I now believe that I am
having problems running parallel threads. I have tried making my TCP vi
run in "other_1" execution mode but have not had any luck so far. It looks
very much like I'm going to have to figgure out the semiphore.

Suggetions?
0 Kudos
Message 3 of 5
(2,523 Views)
Also, watch out if you're using local variables. Sometimes they can execute out of order than if you were just running the program normally.

-Dobbs
0 Kudos
Message 4 of 5
(2,523 Views)
Hi,

I have similar problem when I tried to read Freq and Power level. I discovered that you have to make sure that the first send/receive is completed then you can do the second send/receive. It has something to do with synchronization. To do that you have to creat a vi that tell the previous task is completed and ready for the next task. This can be done by just setting up a simple respond message . It will look like this: Send --- wait for resp message---message receive . Now you can put the whole thing in the loop. If this is not help, I think you can put those steps into sequence (it's only a couple of them).
0 Kudos
Message 5 of 5
(2,523 Views)