LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Not able to access PV which are deployed in RT using EPICS IO Server

Solved!
Go to solution
Hi, We are having NIPXIe 8133 and 1 windows vista PC (32 bit) and I have installed LabVIEW 2012(32 bit) and required modules. I also have one another RedHat Linux based PC (RHEL 6.3 Santiago) and all 3 PCs are connected in same network. I want to develop an application in which I want to deploy some Process variables in RT and want to access this PVs in Red Hat Linux PC. To develop this application, I make RT as EPICS Server and one Red Hat Linux PC as EPICS Client. I have deployed some Process variables using EPICS Server in RT. Now I want to access value of those PVs in Red hat Linux PC and in this PC EPICS base is already installed but I am not able access PVs which are deployed in RT. Please help me to solve this issue. Regards, Ishan
0 Kudos
Message 1 of 5
(2,809 Views)

Ishan, I'm confused already....

 

Which OS is running as the EPICS server? Are we talking the LabVIEW based EPICS I/O Server?

 

What is your knowledge of EPICS? What is your knowledge of LabVIEW?

 

What are your data types? What does any of your code look like?

 

If the RedHat system is just running as EPICS, then I'd head to that community for help with that if you're new to EPICS, and start with making sure that base is compiling and you can do a straightforward caget of something other than the LabVIEW.

 

If you know your EPICS, then hopefully you've already verified your base comilation and run a softIOC up and done a caget/caput on it (if that makes no sense, you really need to learn a little more about EPICS before trying to do much more with it, if it makes sense, great this is just a LabVIEW problem).

 

The most common error I've seen for the LabVIEW/EPICS interface not talking is getting the name of the PV wrong, it's worth pulling up the Distributed System Manager, and opening up the various hosts to find the EPICS server and then checking the variable names (and that the server you expect to be running is actually running).

 

Just to give you an idea, I'm using LabVIEW 2012 32-bit with the DSC module when I do use a LabVIEW EPICS server, my EPICS base is running on Windows (just to be different - my workplace never does things in a normal way), so the basic interactions I'm familiar with. I don't have any RT systems to play with however, so can't comment on those and EPICS and LabVIEW really.

0 Kudos
Message 2 of 5
(2,792 Views)

Hi KathrynB,

I have deployed EPICS Server in NI PXIe RT(Pharlap RT). I have already developed a project in which I made the LabVIEW as EPICS Server and acquired PV data in Red Hat Linux in CSS. I am very familiar with EPICS. Coming back to my current issue, I have already checked PV status in NI Distributed system manager(DSM). If I change the value of any PV which are deployed in RT then it's value is being updated in DSM means EPICS Server is deployed properly and under EPICS Server, I can see the exact name of PV. So I execute the "caget 192.168.1.100:RTLibrary:Amp" command in the PC from which I deployed the EPICS Server in RT and it shows the channel connect time out error. I am able to see the PV under EPICS Server in DSM but caget command gives the channel connect time error.

 

How is it possible? If I run the same command in RedHat Linux then also I get the same channel connect time out error. If I deploy EPICS Server in LabVIEW in Windows Vista PC and run caget command then it execute sucessfully and shows current value of that PV but if i deploy EPICS Server in PXIe RT and run caget command in windows vista PC, it shows an error.

0 Kudos
Message 3 of 5
(2,788 Views)

Thanks for the extra info, it makes it so much easier to talk to the same language when you know the terms of reference.

 

When you deploy the server to a real time/other server with LabVIEW there are a few things you need to make sure are set, that are counter-intuitive as I recall. I don't use any RT, so I'm trying to remember back to a random tutorial. It may be that the Big Physics team will see this and get there before I can find my notes, but there are certain settings for the network shared variable, as well as needing to make sure that the EPICS server is included in your deployed items, as it isn't automatically, even if you are using it, which is the most likely candidate for waht you are seeing. I can't remember where in the settings, but there is something that needs enabling or checking to make this work on the RT system, or that needs installing so do at least feel reassured that there is something missing as standard.

0 Kudos
Message 4 of 5
(2,781 Views)
Solution
Accepted by topic author IshanPatel

 

I haven't assigned Default Gateway in any of my system and that was the only problem in my system. Once I assign Defaullt Gateway in all systems, I can access all PV which are deployed in RT from other 2 machines connected in network .

0 Kudos
Message 5 of 5
(2,736 Views)