Distribute TestStand Application

Hi All,
I have just started to work on TestStand. My work includes various Test Sequences to be controlled and monitored from an User Interface. I had worked on CVI before. Can we make a User Interface similar to CVI in TestStand so that say, if i click a button the corresponding sequence file should be run. My work revolves around the IVI drivers and COM port communication. Please suggest me how to work about this UI and sequence selection.
Coming to the next part, how can we distribute the application developed in TestStand including the UI. Does this require any distribution License from NI?
It would be of great help if someone could sort my confusions out.
Regards,
Siddu

Hi Siddu,
Welcome to using TestStand! You don't have to create a userInterface in CVI. We have created two user Operator Interfaces, one relatively simpler than the other, in CVI that are shipped along with TestStand. They can be found at \OperatorInterfaces\NI. The full featured interface has a lot of functionality than the simple interface. However both these interfaces are completely customizable.
To answer the second question, to distribute your application, you can use the TestStand Deployment Utility. The utility can be found within TestStand, under the tools menu >> Deploy TestStand System. The wizard should walk you through the entire process. You can find more information in the TestStand Reference Manual, Chapter 14.
Also, you would need a TestStand Run-Time License for every computer that you would distribute your application on.
Hope this helps.
SijinK
National Instruments

Similar Messages

Minimum files needed from JRE to distribute Java Application

Hello,
We have developed a Java Desktop Application (A Viewer for projects created in Microsoft Project). The core application is around 4 MB. Now we want to distribute it thru our company website. With each distribution we have to supply JRE so that the user doesnt have to install JRE in his machine manually. Problem is the size of JRE which is around 13 MB. For an application of 4 MB its getting too much for the user to download additional 13 MB of files.
I need the list of files of JRE minimum needed to distribute the application.
thanks
--Ahmed

You are not allowed to distribute a partial set of the JRE, you can only distribute a complete JRE. See the LICENSE file included with any JRE installation.

What files does TestStand add to a computer when a new TestStand application is deployed?

What files (names and directories) are created when a new TestStand application is installed on to a computer? Does TestStand create a new set of files each time that a new application is installed?

Hi,
This depends on how you configure the installer. TestStand 2.0 and 3.0 contain different methods of deploying a TestStand system. Could you tell us what version of TestStand you are using?
Bob

Distributing an application that uses componentworks++

I am new to using MS.
I am going to distribute a Visual C++ application to a remote site, sending the customer a DVD.
The application is an MFC application that uses ComponentWorks++ (from Measurement Studio) and NiDaq.
What do I need to be able to distribute this application? Do I need to license each target for MS? Do I need to separately install Component Works++ on the target?
Thanks,
Menchar

I'm still having some issues ...
We went ahead and used InstallShield 12.0 to create a release file (setup.exe). We're using MS 1.0.1 so we didn't have merge modules, but rather a bunch of .ocx and .dll files in a /redist folder that looks to be the stuff you distribute to get a ComponentWorks++ application deployed. So we included all of those and InstallShield deployed them into the WinNT system folder.
Our app also uses NI-DAQ 6.9, and my understanding is that NI-DAQ must be run on its own installer as provided by NI. So we didn't include any NI-DAQ stuff.
We tried testing our release on a WindowsNT virtual machine (clean machine). We first installed NI-DAQ, then ran our app installer (setup.exe). Install of NI-DAQ and our app appears to go OK, but when we run the application in the virtual machine it fails to start, giving us a popup that says it can't find the CVI rte! We installed the CVI RTE just for giggles but it still died on startup. Does NI-DAQ 6.9 use the CVI runtime engine?
My question now is, can we expect our app to start on the virtual machine? The app is designed to run without any NI-DAQ HW present. I am able to run it on an XP machine without any NI-DAQ HW, but that machine also has ComponentWorks++ 1.0.1 installed, VC++ 6.0 SP 5, and NI-DAQ of course as well (but no NI DAQ HW), and I didn't install the application, I just ran the executable after building it on the XP machine. The app also runs OK on the intended WinNT 4.0 target - but the target has NI-DAQ HW & SW, all of MS 1.0.1 (including CVI), and VC++ 6.0 SP5 installed.
So I can't say I've been able to create a distribution that will install and run on a system that doesn't already have NI-DAQ, CW++ and VC++ installed. I can't find any relevant help on the NI site for these old versions of CW++.
I'm also curious if there's a dependency between CW++ and CW? I.e., from a VC++ application, could it have a dependency on CW (my understanding CW is for VB only).
I see what looks like a VC6 runtime of some sort in the CW /redist folder - is this an extension of the Microsoft VC runtme dll?
Thanks for any help with this, it's funky working with these old versions - I have to support some long-lived systems deployed overseas.
Menchar

ReportGen in TestStand Public Directory not overriding TestStand Application Directory

I copied the reportgen_html.seq from C:\Program Files\National Instruments\TestStand 4.2.1\Components\Models\TestStandModels\ to
C:\Documents and Settings\All Users\Documents\National Instruments\TestStand 4.2.1\Components\Models\TestStandModels and customized it, but find that it will only run if I rename the original in the TestStand application directory.
Is there something else I need to do to get the new report generator to be called correctly?
Thanks!

You need to copy the main model files (Sequential, Batch, or Parallel) to that directory too. The reportgen is being called from those models, and it's finding them in the NI directory, then uses a relative path to find reportgen. I would recommend copying everything from the NI Models directory to the Public directory to get everything to work.
Allen P

Why does Labview hang when switching from one TestStand application to another?

I have several 2013 TestStand applications that use a number of LabView 2013 VIs on a test computer. When the operator closes one TestStand program and opens another often Labview does not close properly and so when TestStand tries to load the LabView VIs it hangs or just runs and every test that is affected by LabView fails. The only potential issues that I know about are that two of the applications were created in LabView 2009 and then updated to 2013. Then two other TestStand applications that were created in 2013 use the function "UnloadAllModules".
Are there any known issues with LabView 2013 that cause this or any ideas of what I might be doing wrong?
Solved!
Go to Solution.

Hello Michael,
I did find my issue. In my case a VI was not getting closed properly in cleanup due to the previous command not passing the handler out correctly. Thus this VI would remain open and when I ran another TestStand Sequence, when it tried to load that LabView VI it would just hang. But the result of this error could be seen when I closed TestStand and LabView stayed open (since it was frozen) or when done running the sequence and closing it. Then if you tried to pop-up LabView it would not show the starting splash screen because it had froze.
I found the problem by turning on the show VI front panel option in TestStand for all of the VI calls. The VI that stayed open was the problem VI.
Thanks everyone who responded for your help!
Kurtus

Distributing an Application that calls Visual Studio built DLLs

I have a LabVIEW application that calls a Visual Studio 2005 built DLL, that then calls functions from another Visual Studio built DLL. When trying to distribute this application it will execute perfectly well on all machines with Visual Studio installed, but on any machine that doesn't have Visual Studio installed, despite installing the Visual Studio Redistributable Package on the machine, I get the error message on attempting to run the application:- "Missing External Function 4575Interface.dllname of function):C in (sub VI that uses function).vi" for the 10+ VIs that use functions called from that dll.
Any suggestions?

Hi,
You will need to manually add all of the DLLs you
are calling into your LabVIEW project. Check in the folder
"Dependencies" in your Project Explorer what external DLLs the VIs are
using, then drag and drop those DLLs into your project. LabVIEW will
not automatically include these external DLLs when you specify a build
specification unless you explicitly tell it to do so. When you
distribute the application using an executable, always explicitly
select those DLLs as "Always Included". Finally, register these DLLs on
the target machine by using Windows command: regsvr32.exe. I would recommend you to write a Windows batch file *.bat to ease the deployment process.
Another
point to note is, the DLLs you will find in the folder "Dependencies"
are the DLLs LabVIEW automatically recognized through what you are
calling in your program. However if any of these DLLs is linking some
other DLLs, LabVIEW will not be able to detect them on the build, so
you will need to add the DLLs into your project manually for
distribution.
Hope this helps,
Best regards,
Mark M.
Applications Engineer
National Instruments UK & Ireland

Cos Naming & WLS Naming Service problem in distributed Tuxedo application

Hi,
we have been starting our investigation into using the distributed application feature of Tuxedo (multiple machines running under 1 Tuxedo domain which make up the entire Tuxedo Application).
when trying to access WLS EJB object references from Tuxedo, we hit a road block.
we have been using the standard method of searching through Cos Naming for a bind context which refers to WLS Naming Service. (this method is documented online and in dev2dev samples) Under non distributed application setup (1 Tux domain with 1 machine as in all examples), everything works fine and I can see into 1 WLS server's JNDI tree. But when in distributed application setup where we want to have two (for example) machines and we also want two WLS managed server with each running on their own machine, it seems that there are some unexpected problems:
1. - only one set of GWADM/GWTDOMAIN is running
- each WLS managed server has its own WTC server
- each WTC server has its own LOCAL ACCESS POINTs
- each WTC server uses the single REMOTE ACCESS POINT
(since only one GWTDOMAIN is running in this scenario)
--> problem: only one of the managed server's JNDI tree can be traversed into, and only one of the WTC link works for Cos Naming -> WLS Naming Service resolution
2. - each machine is running their set of GWADM/GWTDOMAIN
- in dmconfig, two local Tux domain-ids are defined, for the
two groups of GW
- WTC server setup is identical to above; except
each of the WTC is pointing to separate REMOTE ACCESS POINTs
- when DOMAINID in ubbconfig is not defined, the same Cos Naming to WLS Naming Service search fails.
- when DOMAINID in ubbconfig is defined, the same Cos naming to WLS Naming Service search fails with an CORBA:INTERNAL error.
Does anyone know how to get Cos Naming -> WLS Naming Service working properly under a distributed Tuxedo application setup and also with multiple sets of GWADM/GWTDOMAIN?
Thanks.
Edited by a_tam at 04/17/2008 11:34 AM

Hi,
we have been starting our investigation into using the distributed application feature of Tuxedo (multiple machines running under 1 Tuxedo domain which make up the entire Tuxedo Application).
when trying to access WLS EJB object references from Tuxedo, we hit a road block.
we have been using the standard method of searching through Cos Naming for a bind context which refers to WLS Naming Service. (this method is documented online and in dev2dev samples) Under non distributed application setup (1 Tux domain with 1 machine as in all examples), everything works fine and I can see into 1 WLS server's JNDI tree. But when in distributed application setup where we want to have two (for example) machines and we also want two WLS managed server with each running on their own machine, it seems that there are some unexpected problems:
1. - only one set of GWADM/GWTDOMAIN is running
- each WLS managed server has its own WTC server
- each WTC server has its own LOCAL ACCESS POINTs
- each WTC server uses the single REMOTE ACCESS POINT
(since only one GWTDOMAIN is running in this scenario)
--> problem: only one of the managed server's JNDI tree can be traversed into, and only one of the WTC link works for Cos Naming -> WLS Naming Service resolution
2. - each machine is running their set of GWADM/GWTDOMAIN
- in dmconfig, two local Tux domain-ids are defined, for the
two groups of GW
- WTC server setup is identical to above; except
each of the WTC is pointing to separate REMOTE ACCESS POINTs
- when DOMAINID in ubbconfig is not defined, the same Cos Naming to WLS Naming Service search fails.
- when DOMAINID in ubbconfig is defined, the same Cos naming to WLS Naming Service search fails with an CORBA:INTERNAL error.
Does anyone know how to get Cos Naming -> WLS Naming Service working properly under a distributed Tuxedo application setup and also with multiple sets of GWADM/GWTDOMAIN?
Thanks.
Edited by a_tam at 04/17/2008 11:34 AM

Distributed Oracle Application implementation

Dear all
I am EBS technical consultant and I am working on SOA and have distributed Oracle Application implementation, Please consider the scenario and suggest me and idea.
Assume
1. We have 2 Oracle Application R12 (EBS-1, EBS-2) installed at 2 different remote locations, network disconnection may be broken in this network.
2. User always work on (EBS-1 installed at Head office) and only work on EBS-2 (installed local) in case network connection is lost to connect with EBS-1.
EBS-2 is a replica of EBS-1 when installed first time.
Scenario
Last inventory Item ID is 5444 at EBS-1
Last inventory Item ID is 5444 at EBS-2
user created new item at EBS-1
Last inventory Item ID is 5445 at EBS-1
Last inventory Item ID is 5444 at EBS-2 (No new item created)
Now User can not access EBS-1 and now start working on local EBS-2 and take following steps.
create new item at and now inventory item id is 5445. and create misc receipt to this newly created item.
Now EBS-2 installed at head office is available and now transaction made at EBS-2 need to be transferred at head office, and what about inventory item id, 5445 is already created at EBS-1 how and i have to import item and misc receipt, inventory item will be different how can i load misc receipt transaction.
Please help.
1. Can I solve this issue by using BPEL process. ? or
2. this type of integration is not possible in Oracle E Business suite.
3. Oracle Application will not work in Distributed environment.
please suggest me.
Thanks
Aamir

Dear expert !
I have a similar question but the scenario is somewhat different.
The are two separate organizations and both have implemented EBS (Vision 1 and 2). they both are under third organization (vision A). for numerous business processes, both vision 1 & 2 get their approvals from vision A. Moreover, inventory and master data is required to be sync in all three.
vision A
^ ^
v v
vision 1 vision 2
In some instances, a sub organization from vision 1 can be placed under vision 2, thus requiring movement of its relevant data from vision 1 to 2.
the scenario implies that all three organizations are running an EBS instance which is a legacy for the other.
can we achieve this integration?
Please do let me know if I am unable to explain the use case.
best regards

UCS-B for distributed computing application

Hi,
What will be the best server configuration UCS-B for under centOs for a distributed computing application through a clustering tool like SGE or LSF
This software is a biological application and the customers want 400 core in each blade.
I need just a first assessment.
Regards

What's wrong with that code is that it's not posted using code tags.
What's wrong with your thread here http://forum.java.sun.com/thread.jspa?threadID=5280844 ?

Are there requirements for deterministic data messaging for distributed LabVIEW applications across an Ethernet/IP network?

Our company, RTI, just entered the Alliance program at NI recently. We are curious if power users of LabVIEW have requirements for running distributed, networked multiple LabVIEW applications in a real-time, deterministic manner? We are not attempting to make this a sales call, but if we should move forward with development of VI's that would allow distributed LabVIEW applications. We have already scoped out the effort but would desire guidance on VI layout, descriptions and ways to give a "LabVIEW" experience and not a disjointed VI that looks unprofessional. We are not sure if demand exists for this requirement. We would like to talk with
and investigate what LabVIEW users want. We have attached a brochure on this real-time data messaging technology for your review. Maybe there is no need and we apologize for taking up NI Developer Zone bandwidth and your valuable time. We look forward to your responses.
Attachments:
NDDSProductBrief3-022.pdf ‏299 KB

We are developing applications which have a distributed layout. E. g. we have 2 or more test places which have no user interface and 1 place where all configuration and monitoring and data storage takes place. We use TCP/IP for the inter computer communication. For this purpose we have developed a library which has a similar interface and usage as the queue VIs from LV 6.0.2. We support automatic reconnect on TCP/IP level and an acknowledged transfer so the application programmer can choose to do some temporary storage on the local machine if the server part is not running for sometime. We have not seen any market for that so we do not plan to sell it.
From a lecture at the NI VIP days in Munich in 2001 personal from Siemens A&D stated to
get a real time like behaviour over ethernet with TCP/IP you have to take care that the load will not exceed 30% of the bandwidth.
For more information you can contact me at [email protected]
Waldemar
Waldemar
Using 7.1.1, 8.5.1, 8.6.1, 2009 on XP and RT
Don't forget to give Kudos to good answers and/or questions

Updating JRE distributed with application

Hi guys,
I have an application that I distribute with the JRE. I have created an update routine, so I can update my application over the internet. This works great... but all I update is my own application.
I would like to update the JRE if this has been updated as well. This is not updated by the Java Web Start service if this is installed on the computer either.
The application consists of two applications. First a launcher application, that will check my website for available updates and also validates user permissions. What this application does is that it downloads zip file containing all the necessary jar files etc, and unpacks them into the application directory to overwrite the existing installation, and after that executes the Java executable to run the application.
The launcher application would be running on the JRE installation that I want to upgrade. How would the application react to upgrading its own JRE?!? Can it be done?
The initial install is done using a Wise Installer.
/jp

I am pretty sure that wise itself can be configured/constructed to do updates. If java is involved at all it would only be to tell wise what version is currently installed.

Distributed TestStand vs. Sequence Editor

I am having problems with TestExec that I created a distribution for. The executable that I created and distributed, will not run, for failing to find or open a DLL. However, when I run the same sequence using the Sequence Editor that I also included in the distribution, it works fine. I have checked and it appears that all setting are the same on both the Executable and the Sequence Editor, can anyone explaion this.

Hi greggj_00,
Did you follow the directions in this article to include your search directories and configuration in your deployment? I'm assuming this is a DLL which is called by by one of the steps in your sequence. Where is this DLL placed on the file system in relation to the sequence file? Also could you post some more info on the error like snowpunter suggested?
Justin D
Applications Engineer
National Instruments
http://www.ni.com/support/

Extract error while distributing Desktop Application

Hello,
we have a problem with extracting a Desktop Application on a special
location. The same TED Dist is working on a few other locations (different
channels). Also other Distribution work on the specific location. we've
already deleted all temporary files on the ZFS volume on the subscriber and
distributer, rebuilded the distribution. We've also deleted the specific
channel and recreated it, but this didn't solve the problem anyway.
the error is as following:
Info 2006/04/12 08:42:02 AM Creating new application failed, Subscriber
Tree Name= OrgTree, Subscriber DN = Subscriber_WOHNW01.ZFS.WIN01.PD, error
message: Unknown meaning for error number -1; Please call a Novell support
provider Also please make sure you set up a working context in the
Subscriber's properties. was sent successfully to winnw05.intra.pdag.ch
Error 2006/04/12 08:42:02 AM Extract
ZFS:\SUB\org~desk_pda.deskapp.zfs.win01.pd\1142589 211390 stopped with
error: Unknown meaning for error number -1; Please call a Novell support
provider Also please make sure you set up a working context in the
Subscriber's properties
we are running ZENworks 6.5 SP2 Server Management on all servers, but the
problem was already there without sp2.
thanks in advance for any help
jimmy ott

Are you sure the working context is set?
Does the working context show up if you look at the ZSM config through
remote webconsole?
Rom
<[email protected]> wrote in message
news:MB1%f.4628$[email protected]..
> Hello,
>
> we have a problem with extracting a Desktop Application on a special
> location. The same TED Dist is working on a few other locations (different
> channels). Also other Distribution work on the specific location. we've
> already deleted all temporary files on the ZFS volume on the subscriber
> and
> distributer, rebuilded the distribution. We've also deleted the specific
> channel and recreated it, but this didn't solve the problem anyway.
>
> the error is as following:
> Info 2006/04/12 08:42:02 AM Creating new application failed, Subscriber
> Tree Name= OrgTree, Subscriber DN = Subscriber_WOHNW01.ZFS.WIN01.PD, error
> message: Unknown meaning for error number -1; Please call a Novell support
> provider Also please make sure you set up a working context in the
> Subscriber's properties. was sent successfully to winnw05.intra.pdag.ch
> Error 2006/04/12 08:42:02 AM Extract
> ZFS:\SUB\org~desk_pda.deskapp.zfs.win01.pd\1142589 211390 stopped with
> error: Unknown meaning for error number -1; Please call a Novell support
> provider Also please make sure you set up a working context in the
> Subscriber's properties
>
> we are running ZENworks 6.5 SP2 Server Management on all servers, but the
> problem was already there without sp2.
>
> thanks in advance for any help
> jimmy ott

Distributed Swing Applications??

I'm not sure if this belongs in the swing or the RMI sections, but since I think it has a little more focus with GUI development I posted here - sorry if it is off-topic!! Here is my problem:
I am interested in building a desktop application to house registered java applications on remote machines. I am interested in pulling in developed and existing applications that are to be run remotely, and tieing them together under one central view. The best software analogy I can come up with would be an xserver/xclient relationship between java applications.
The functionalities I need are:
local views of each distributed application, encapsulated inside one large (local) java desktop.
Distributed performance... half of the applications are developed with JNI interfacing and ARE machine dependant.
The ability to stop and restart server applications on-the-fly, in order to patch and update server functionality.
Minimal rearchitecture of existing java applications.
I have been bouncing around ideas, but there are only a few that really seem to stick out.
One solution would involve rearchitecting existing applications to decouple the view and the event handlers. So, my Model-View-Controller has local views and controllers, but remote models to maintain remote state (and computational performance). So then, each event handler should update the GUI correctly, but also update remote state information on my server.
I guess another solution would be to intercept these applications at the X-level and re-assign their graphics contexts inside my java application. I haven't pursued this option yet, because I figured there was a better pure-java solution.
My current test solutions have been implemented utilizing RMI, and I think that is probably the way to go, but I am open to opinions. While rearchitecting these applications to exist within a client/server relationship is allowable, it is not preferred. Thanks for all the help!!

Why don't you just open up like five SSH consoles, or
remote desktop connections, and be done with it?
Ok, do you have motivation for being against
client/server... as you could build a server that
acts as a controller for your swing GUI. The server
would talk to the apps, and coordinate the views in
the swing GUI.I don't want to just use consoles and x-forwarding, because the end result should be a desktop that will be "easy-to-use" for people that are relatively unfamiliar with the application suite.
I agree with the second choice; and think that I'll probably end up going that route. Is another good implementation besides a MVC architecture, that doesn't require me to rearchitect existing apps??
Thanks!

Distribute TestStand Application

Similar Messages

Maybe you are looking for