Time-outs with NI DSM monitoring cRIO but not with other applications

[tyk007]

Hi there.

A colleague of mine is having a problem overseas that he has communicated to me. He has a cRIO-9063 running a small real-time application. It is publishing a number of network shared variables (approx 50) and they are monitoring a measure of these with a small LabVIEW executable. Only a small number (20) are monitored as the remainder are purely debugging information to assist with remote support (aka me).

We use Distributed System Manager to look at all these network variables. For some reason, every 10 seconds or so DSM displays the variables as "Connecting" and all the variables values in the column do not change - however if he looks at the variable value on the detailed display on the right (where one sees trends etc.) each value still appears to be updating however only if the focus is removed from the variable (ie moved to another variable) and then replaced - we have a counter for example that can clearly be seen increasing using this method. After the time period all the values start updating again in the table. Interestingly the small LabVIEW application does not show any interruption at all with the monitored data - the heartbeat we added still appears to function.

We did not see this quirk when this was being setup in the development environment (locally) prior to being put into manufacturing. Does this sound familiar to anyone?

[tyk007]

I forgot critical information - this is NI DSM 16.0 32-bit and the LabVIEW application the same version and bitness (albeit with a f2 patch).

[alyssa.h]

Hey tyk007,

I would like to clarify a few things:

1. You can still see the cRIO in DSM, just not the Shared Variables sometimes?

2. You said this is happening around every 10 seconds or so. What else is your customer's computer doing? What's the CPU usage, Memory usage, etc... look like?

3. Does pressing F5 to refresh help? 

Thank you,

Alyssa H.

National Instruments

[tyk007]

Thanks for the reply! To answer your questions with what I know at the moment:

1. The cRIO still seems to be present; although if we attempt to look at this directly (eg. cpu usage) then that occasional drops too.
2. Apparently DSM is the only application running at the time (unless we use that additional application we talked about) and my engineer says that the CPU (dual core 2 GHz) is running around 10% during this time.
3. We have not tried the Refresh, I'll ask him to give that a go.

We have improved the situation with DSM by completely disabling the Windows firewall for that adapter. Now the disconnects occur less frequently (every minute or so) though its not clear to me why. Perhaps the variable connectivity occurs over more than one port for redundancy? Either way the little Labview application still appears to have no ill effects.

[alyssa.h]

It sounds like you may be having network issues then and that it is getting bogged down by something. Do you have a lot going over the network? Are you using the same network to access the internet? 

Regards,

Alyssa H.

National Instruments

[tyk007]

It is a closed network with a total of 4 devices. In this situation, only three of those devices are active (the cRIO, PC and a PLC from another supplier doing some custom process control). When the problem is occurring the PC is only running DSM and/or the Labview application.

I talked to the engineer last night and he confirms that the F5 refresh doesn't seem to improve the situation - in fact if it appears to be reading ok, using F5 actually seems to cause the interruption to occur.

Our next step is for him to have a look at the connections and traffic throughput on the PC to see if there are any flags there. Will get back to you with what he finds. Thanks for your help so far.