Labivew 8.2.1 is locking up

[Matt361]

Hi Everyone,

I was wondering if anyone has seen a simular issue to what I'm seeing now. I have a fairly large app thats running on Windows Server 2003 (I've also ran it on XP and had the same issue). Periodicly the app locks up. It will go for days on end without any problem and then it starts locking up two or 3 times in a 24 hour period. We were orginally running the program on XP but transfered it to a server (both hardware and software upgrade) because we though that a more stable machine would help. Typically the CPU ussage is around 30-40% on XP but once in a while it would go up to 90-100% for a few hours to a couple of days.

The app monitors a station and must run without locking up. It currently is talking to a RSlogic OPC Server, Kepware OPC Server, Modbus and a cRio. I have 12 or 15 IO Servers to connect to the different servers (OPC and Modbus). I'm also using Win911 to aleart the operators of alarms, which is connected to NI's OPC server. I'm loging all the data to a citradel database, which is the root cause to the extreme CPU ussage, but I'm not sure about the Locking up. Most of my code is fairly simple thanks to the DSC package and shared varibles so I don't think it's a problem with that.

Has anyone seen anything simular to my problem? Any posible ideas I can try? I'm at the point were I'm starting to think that labview (and the DSC package) just can't handle everything I'm trying to do. I have 4 other stations set up the same way running Wonderware on a XP machinie without any problems.

Thanks in advance for any comments on this problem

Matt

[Micham]

Matt,

You should use the Windows Task Manager program, go to the Processes tab and sort the CPU column. Then you can see which process is consuming the CPU. Then you can contact the supplier of this process/program and check with him for the reason.

Another approach may be to close one process after the other and see when the CPU returns to normal.

Regards,

Mike

[Matt361]

Micham wrote:

You should use the Windows Task Manager program, go to the Processes tab and sort the CPU column. Then you can see which process is consuming the CPU. Then you can contact the supplier of this process/program and check with him for the reason.

Another approach may be to close one process after the other and see when the CPU returns to normal.

Thanks for the idea Micham,

The task manager shows the nicitdl5.exe and taskmgr.exe using up most of the CPU usage. If I stop the services (Labview crashes of course) the cpu usage on the XP box was done under 10%. I haven't tried this on the Server yet, I will next week once I'm back onsite. I have detached the database a couple of times in MAX and recreated a new one. That seems to help the problem for a few weeks.

To support my programming skills... the same application works fine on another computer, but this test computer doesn't have the modbus, kepware or Win911 installed (so its not working as hard). Basically I was using the this test unit to run the cRIO and the RSlogix interface while testing.

Any advice would be greatly appreciated!

Matt

[Micham]

Matt,

I suggest that you should close the following process one by one and see if the CPU percentage is going down after each close.

Modbus
Kepware
Win911

If the CPU is going down after you close one of the above processes, then you know which of the processes is the "main suspect".

Mike