Jump to content

Cat

Members
  • Posts

    815
  • Joined

  • Last visited

  • Days Won

    15

Everything posted by Cat

  1. Thanks for the response. I hadn't thought of asking about LabVIEW courses. This position is actually more than "just" LabVIEW programming. It's the software part of a team that designs/develops/tests/fields data acquisition systems. So an engineering degree is helpful to being able to see the big picture. Or so the guy who does the hiring believes... I personally fit in the "lots of experience but no certification" slot. That's why I didn't think it would be necessary to require it. Plus it will make it harder to fill the position. If I go that route, having never taken the exam myself, I'm wondering if the group mind here thinks CLD would be good enough, or should it be CLA? Especially considering the person will have to be the "A"rchitect as well as the "D"eveloper.
  2. Subtitle: Cat's continuing attempt to finally retire I retired from the gov't (US Navy) around August of 2019. Since the gov't is the gov't they didn't think about replacing me until after I retired. So being a good little civil servant I came back part time as a contractor to fill the gap. The basic requirements for my replacement were: 5+ years LV work, engineering degree, experience working with large data sets, experience working with large (2500+ vi) code sets, able to work independently (I have always been a team of 1). The gov't finally hired my replacement and they started work in January of 2021. It still being pandemic time, I didn't have a lot of face-to-face (or mask-to-mask) interaction with them, but I got more and more worried when I did. While they had a lot of fundamentals down, there were glaring gaps in their technical knowledge. They were more of a okay beginner programmer than the solid intermediate to expert level that we needed. And they had obviously only ever worked in a team where everything was spoon-fed to them and then someone cleaned up after them. They were let go. This person had 6 years of programming LabVIEW on their resume. But it was obviously 6 years of programming LV badly. Going forward, how do I not repeat this mistake? Should NI certification be required, and if so, is CLD good enough or should it be CLA? Any other suggestions? I really want to retire!! Cat
  3. hooovahh -- what you're talking about is one of our options. Shaun -- we hunted around and those commands don't seem to be available on LinuxRT On top of the "standby" issue, after much back and forth with NI, we discovered that the system drive in the IC isn't removable. And it can't be encrypted. While we probably could have worked around the former, the latter is a show-stopper for our user. So, looks like we're ditching the NI "Industrial Controller" and going to someone else's "Industrial Computer". Plain jane Windows 10, 2 removeable drive slots, all of which will make life much easier. Not to mention we can buy 4 of them for the price of one IC-3173. 🙄 Thanks for the responses, guys!
  4. Hi all (no, I'm not really retired yet, lol), My latest toy is an IC-3173. I need to be able to programmatically (LV, of course) put it into a reduced power state (analogous to Windows standby/hibernate/sleep). And then "wake it up" every hour for 10 minutes. I'm running LinuxRT. Does it even have a concept of standby/hibernate/sleep? Cat
  5. So I'm upgrading from LV2016 to LV2019. Or attempting to, anyway. I have to use the offline version, and it keeps bailing at something like "ni-opc-support", and giving me a very unhelpful, "check your internet connection" message. Hello, this is supposed to be an offline installation?!? My other complaint is that I've installed every version of LV since 2.5 in a "LabVIEW" folder -- no version number added -- and I can't figure out how to make the NI Package Manager let me do that. So I've ended up with 3 or 4 different LabVIEWxxxx directories (depending on how far things get before the install crashes). And if I completely uninstall all NI software, NI Package Manger still thinks it's installed and won't reinstall, but since it's not really there, there's no way to fix it. I'm on my third hard drive copy. Hmm. Sorry, I didn't mean to turn this into a b!tch session. 🙄 My question is whether or not I should uninstall just LabVIEW 2016 (leave everything else on there) before attempting to upgrade? I'm not sure how much I have to take off to get a "clean" enough install, but not too "clean". If that's even the problem. And on a "funny" note, I just saw that LV2019 SP1 was released last week (I've been using 2019 f1). I'm downloading the offline version at the moment. My screaming fast work network is telling me I only have 3 days until the download is complete... Cat
  6. We "upgraded" to Windows 10 version 1903 (from 1803 or 1809, I don't remember) and iperf3 went from ~450 MB/s to ~2 GB/s. Yay! Unfortunately, 1) the IA gods have not deemed 1903 worthy so we're not supposed to even have it installed, 2) there is question of whether this a fix, or something that will disappear in the next version, and 3) ironically, 1903 is causing issues with various types of hardware NICs (not a problem for us -- so far). But, for the moment, we've got something that works. Thanks to all for responses, and thanks to Gribo for suggesting iperf3. It's made it a lot easier to test network throughput. Cat
  7. We could try running iperf, just to confirm it's the loopback adapter and not the code. But as I said, we've run the same code with 2 hardware NICs (10G) connected on the same computer and it works fine.
  8. The point of the loopback adapter is to use TCP to communicates to pass data between two different executables ( one C and one LV) on the same machine. The machine has 2- 1G and 4- 10G physical network adapters. Can you explain more what you mean by "link could be fully local"?
  9. We've set the loopback adapter the same way we've set up hardware NICs (at least as much as is applicable). I'll confirm that QOS and Nagle specifically have been dealt with. The C dev tried fastpath (even tho it was supposedly deprecated a while back) but that didn't help any. The computer has a very fast CPU. Two of them, actually. I guess the implication is that the hardware NICs are doing something the CPUs can't.
  10. Hi all! Long time, no talk to. 🙂 I'm supposedly retired, but then decided to go over to the Dark Side and become a contractor. We'll see how long that lasts... Current issue: A C developer and I are sending data via a MS loopback adapter between a C app and a LV app on the same machine (Windows 10). Past iterations have worked great (after that little bug in LV11 was fixed). The new system, however, needs a much higher continuous data thruput -- somewhere in the neighborhood of 900 MB/s. The loopback adapter is topping out somewhere around 500MB/s. If we switch from internal TCP to connecting two 10G NICs in the same box, the data gets passed with no problem. So it's not the code, and we know how to configure hardware NICs for large data thruput. We've tried everything the web seems to offer, including a lot of things that have been supposedly deprecated. We've tried using localhost, 127.0.0.1, a 192.68 address, and various other IP addy options. If you google "how to speed up a loopback adapter" or any variant of that, we've probably tried it. Does anyone have experience with tweaking a loopback adapter and have any suggestions of things to try? If it matters, the computer is a screaming fast dual Xeon with 128 GB of memory. Cat
  11. I use ini files. If I have complex data types, or large amounts of data required, the ini points to files where that data is stored. Also, LabVIEW annoyingly creates an ini file for every executable, so I figure I might as well use it. Cat
  12. Crud. I replied to this, but it never showed up in the feed. Huh. Anyway, yes, I'm turning this over to one of our C folks to deal with. I'm not sure if I can get the .NET version past our Information Assurance zealots, but a regular old dll I can call with a CIN will be fine.
  13. I just discovered that my latest needs-to-be-done-yesterday project requires communications via an Apache Qpid broker (v0.4 and AMQP 0-10). Since I didn't even know what AMQP was before yesterday, I'm a bit behind the curve on this. I have downloaded LabbitMQ, but I'm assuming that won't work with Apache Qpid? I was hoping that the whole open standard thing meant that different implementations of AMQP should be able to talk with each other, but the following link regarding interoperability between the two seems to say that this often isn't the case. There is a Native AMQP client for LabVIEW on GitHub, but it is missing all the vis (at least from what I can tell), so isn't much help. I posted an "Issue" to the Repository, but don't know when/if I'll ever get a response. Any thoughts/comments/suggestions? Cat
  14. Thanks for the process walk-thru. Unfortunately when I first read it I didn't know enough for it to help. Now it make perfect sense!
  15. Grrr. Yeah, the schedule change messed up my plans there. Hopefully next year I'll be there and can buy you a belated round.
  16. Wow. So after a day or so of flailing around I finally got this working. I think I only ended up making 2 changes to the Example.vi, the rest of the time I was trying to figure out all the networking stuff to make it actually work. Some of that time was spent combing thu this link for little hints. Some was spent figuring out what firewall setting I needed to turn off. Along the way I learned about symbolic links and how to configure an Apache web server. It was fun! Thanks hooovahh! And thanks to everyone else who contributed to this code.
  17. Maybe it's because it's Monday and I haven't recovered from either Cinco de Mayo (tequila) or the Kentucky Derby (bourbon) yet, but this one is twisting my little brain. After years of managing to avoid LabView and the Interwebs, somebody got the great idea of running some of my code via a web browser. Sure! I say, LabVIEW's got stuff to do that! Three hours later and it just ain't working right. I've read all sorts of things like: http://zone.ni.com/reference/en-XX/help/371361M-01/lvconcepts/ws_distributing/ http://zone.ni.com/reference/en-XX/help/371361L-01/lvhowto/enabling_the_web_server_in/ http://zone.ni.com/reference/en-XX/help/371361M-01/lvconcepts/ws_web_server/ http://zone.ni.com/reference/en-XX/help/371361M-01/lvhowto/build_web_service/ etc., etc. They just don't seem to be fully addressing my situation. Unfortunately there doesn't seem to be document that goes thru it all from start to finish. Here's my setup: Computer A: Running a LabVIEW application -- call it TestWeb.exe Computer B: Running Explorer/Chrome/Firefox. No software can be loaded or installed on this computer, including any NI software. Is there something I can do to be able to view/control TestWeb.exe, via a web browser, from Computer B? I've gotten it to work wonderfully -- using source code, but even tho I think I'm doing what the various docs say, it isn't working with an executable. Are there more complete directions somewhere someone can point me to? Cat
  18. I used to do Approach 2, then a user moved an exe to a different computer -- that of course didn't have the dependencies installed for that exe. Now I do Approach 1 which works best if you don't always jump on every LV yearly upgrade and SP. I tend to upgrade every other year, so can go for a long time without needing the "Full" installer. Approach 3 is fine -- unless of course you aren't tied to a network, or don't have the concept of a "server" in your network...
  19. Cat

    Wait. What?

    Well crud. I was actually starting to schedule some stuff around NIWeek -- in August. I've been telling the Big Boss if they send me, I might consider not retiring next year. But May would be hard to work in. I have to agree about the heat, tho. Austin in August is, umm, toasty.
  20. Cat

    Wait. What?

    Am I the last person on Earth to realize that NIWeek is moving to May in 2017?!? Clueless Cat
  21. A tangential issue when using Ethernet conversion dongles... We use the USB to Ethernet variety. A problem I ran into recently was when we had to communicate with other NICs that used jumbo frames, and the dongles we had didn't support that. Just something to keep in mind. (I found a USB to Ethernet dongle from Anker that does jumbo frames) Cat
  22. According to the calendar, I'm eligible to retire in 1 year, 1 month, and 1 day. I'm doing the code cleanup for my (so far hypothetical) replacement. Or I just retire, and come back the next day as a contractor for twice what I'm making now. :-)
  23. Ah, the good ole days... I only use the S&V toolkit, which I purchased long before OOP was a gleam in NI's eye, and never had a need to upgrade. Other than that, I use very little code that I didn't write (instrument drivers and OpenG being the major exceptions). This means, of course, that a lot easier ways to do things have probably been developed over the years, and I'm just too unaware/lazy/stubborn to switch over. hooovahh, I think you've convinced me to at least build a few of my exe's with that 8.x box unchecked and see how horrible it's going to be to "fix" it. Or I can just leave it as an exercise for my replacement...
  24. Yup, all my executables are in 8.x layout. I will admit the one time this has been an issue was when I used two hardware drivers that were in lvlibs, and they both had an "Init.vi". I just renamed the two vis and went on happily. I haven't seen a need to use lvlibs. I am currently on a team of one, and I only distribute exes, not source code, so the public/private aspect of lvlibs hasn't been necessary. I organize via directories. I guess I gave up on LabVIEW libraries back when one of my llbs went over 1.44MB and I couldn't fit it on one floppy anymore. :-)
  25. Here's one most of you probably haven't thought about for a few years. I built an application for someone else, and when the exe was run on their computer, it started complaining about missing vis. I realized this probably meant the "Use LabVIEW 8.x file layout" button got unchecked somehow, so I fixed that and all was fine. Which started me thinking... Other than the issue LV 8.x and earlier builds have with vis with the same name, is there any technical reason to NOT use the LabVIEW 8.x file layout when making an executable? I don't use LVOOP, and think it's Bad Programming to have two vis with the same name in the same build (either they have slightly different functions, and therefore have different names, or it's the same vi living in my code reuse tree, or maybe it should be a polymorphic vi, etc). I'm going thru a big code cleanup push and am wondering if this is something worth the effort of "fixing" in my 2500+ vi and 25+ exe library.
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.