FREE, FAST SHIPPING IN THE U.S.
CUSTOMER SERVICE +1 (949) 613-5838

Linux Issues


  • I am running CAE Linux, which is an Ubuntu-based engineering build. I am also running the latest most up-to-date version of Ubuntu. Running Matter Control on this laptop is problematic. The software vanishes at random without any message or any event logging,

    causing the printer to stop.

    Sometimes I can finish a print, more often than not I waste a significant amount of material. Nothing else is running, just Matter Control. I have four gig of RAM on here, I don't see any signs of memory leak meaning I am not suddenly running out of memory. I suspect that the problem is the INTEL Graphic crap, however the Linux version doesn't seem to accept the force software rendering argument. I tried it and I still get random crashes.

    Right now as it stands I cannot run Matter Control from this laptop which is a setback for us. I was told that MC was sufficiently stable under Linux to be considered reliable. This is not the case.

    First step is overcoming the problem caused by the Intel chip-set. How do I force it like Windows? Second, where is the program logging events? In order to troubleshoot this I need a crash log or event log so I can figure out if this is my build, my laptop, or the software causing this.

    "It is possible to make no wrong moves and still lose, that is not a failing that is LIFE."

  • linux 14
    11
    Posts
    6777
    Views
    Log in to reply


  • @LordRichard

    I can't offer any solid advice, but in passing I seen a comment on a forum saying it is an issue with mono that causes MC to close out spontaneously, your problem was happening to me but I just went back to windows. At least MC "worked" a bit, I couldn't get cura or RH to connecct to my printer at all, I'm too new at Linux to try and fix these kinds of problems, hope this helps or gets you pointed in the right direction.

    Just adding that yes I added myself to the right groups, tty? and another one, forget where I read that, just sayin so that you might be able to go from here.



  • Ok that actually tells me quite a bit, I'll look at "mono" and see what is going on there.

    Linux has left me behind, I was "down with it" back in 2007-2010.... it's really matured, especially Ubuntu.

    If I had some kind of output (error messages or the like) I could begin working on it in my spare time. Hopefully someone comes along that is more up-to-date than I am.

    "It is possible to make no wrong moves and still lose, that is not a failing that is LIFE."


  • The first troubleshooting step you should take on Linux is to simply run MatterControl from the terminal.

    $ MatterControl
    

    Any errors or exceptions it throws when it crashes will be shown there. For more detailed error logging, you can also do

    $ MONO_LOG_LEVEL=debug MatterControl > log.txt

    The issue you are having is probably this one, and not related to Intel graphics. The Intel graphics drivers on Linux are much more solid than on Windows and you are seeing different symptoms. If this is the case, the issue is actually a bug in one of the underlying libraries that MatterControl uses (OpenTK). This has made it very difficult for us to deal with. We have been investigating it, but I'm afraid that our dev team has higher priorities at the moment.

    What browser are you using? Maybe @jesstify can help with your forum issues.



  • Ok thanks for your reply!

    First, I popped the hood and forced Mono to update, I had some really out-of-date components that installed with CAE Linux. Turns out that Mono doesn't update as part of the normal cycle because it uses it's own repositories... which had to be enabled manually. I was able to get three entire prints out without a crash. Aside from the usual ABS-warp-in-one-spot issues the parts came out excellent. THAT problem is caused by uneven heat coming up from the heat-bed... I will be troubleshooting that once I get the software working. This is less of a priority because I normally run PLA, seldom run ABS parts larger than a half-dollar. When I do, I get warpage in the exact same spot every time regardless of orientation.

    Anyway we'll see. Thank you for the logging commands. I'll run a couple of test prints this evening after work and see how we do.


    "It is possible to make no wrong moves and still lose, that is not a failing that is LIFE."


  • Unhandled Exception:
    System.NullReferenceException: Object reference not set to an instance of an object
    at System.Windows.Forms.XEventQueue+PaintQueue.Dequeue () <0x412f0920 + 0x000ed> in <filename unknown>:0
    at System.Windows.Forms.XplatUIX11.GetMessage (System.Object queue_id, System.Windows.Forms.MSG& msg, IntPtr handle, Int32 wFilterMin, Int32 wFilterMax) <0x412e3280 + 0x003f7> in <filename unknown>:0
    at System.Windows.Forms.XplatUI.GetMessage (System.Object queue_id, System.Windows.Forms.MSG& msg, IntPtr hWnd, Int32 wFilterMin, Int32 wFilterMax) <0x412e3230 + 0x00047> in <filename unknown>:0
    at System.Windows.Forms.Application.RunLoop (Boolean Modal, System.Windows.Forms.ApplicationContext context) <0x412e1340 + 0x00e4b> in <filename unknown>:0
    at System.Windows.Forms.Application.Run (System.Windows.Forms.ApplicationContext context) <0x412e1290 + 0x00057> in <filename unknown>:0
    at System.Windows.Forms.Application.Run (System.Windows.Forms.Form mainForm) <0x412c0fc0 + 0x00033> in <filename unknown>:0
    at MatterHackers.Agg.UI.WidgetForWindowsFormsAbstract.Run () <0x412cd5d0 + 0x0001f> in <filename unknown>:0
    at MatterHackers.Agg.UI.WidgetForWindowsFormsOpenGL.Run () <0x412cd5b0 + 0x0000f> in <filename unknown>:0
    at MatterHackers.Agg.SystemWindowCreator_WindowsForms.ShowSystemWindow (MatterHackers.Agg.UI.SystemWindow systemWindow) <0x412035e0 + 0x00391> in <filename unknown>:0
    at MatterHackers.Agg.UI.SystemWindow.ShowAsSystemWindow () <0x41203570 + 0x0002a> in <filename unknown>:0
    at MatterHackers.MatterControl.MatterControlApplication.get_Instance () <0x41086460 + 0x0004b> in <filename unknown>:0
    at MatterHackers.MatterControl.MatterControlApplication.Main () <0x4101dea0 + 0x0020f> in <filename unknown>:0
    [ERROR] FATAL UNHANDLED EXCEPTION: System.NullReferenceException: Object reference not set to an instance of an object
    at System.Windows.Forms.XEventQueue+PaintQueue.Dequeue () <0x412f0920 + 0x000ed> in <filename unknown>:0
    at System.Windows.Forms.XplatUIX11.GetMessage (System.Object queue_id, System.Windows.Forms.MSG& msg, IntPtr handle, Int32 wFilterMin, Int32 wFilterMax) <0x412e3280 + 0x003f7> in <filename unknown>:0
    at System.Windows.Forms.XplatUI.GetMessage (System.Object queue_id, System.Windows.Forms.MSG& msg, IntPtr hWnd, Int32 wFilterMin, Int32 wFilterMax) <0x412e3230 + 0x00047> in <filename unknown>:0
    at System.Windows.Forms.Application.RunLoop (Boolean Modal, System.Windows.Forms.ApplicationContext context) <0x412e1340 + 0x00e4b> in <filename unknown>:0
    at System.Windows.Forms.Application.Run (System.Windows.Forms.ApplicationContext context) <0x412e1290 + 0x00057> in <filename unknown>:0
    at System.Windows.Forms.Application.Run (System.Windows.Forms.Form mainForm) <0x412c0fc0 + 0x00033> in <filename unknown>:0
    at MatterHackers.Agg.UI.WidgetForWindowsFormsAbstract.Run () <0x412cd5d0 + 0x0001f> in <filename unknown>:0
    at MatterHackers.Agg.UI.WidgetForWindowsFormsOpenGL.Run () <0x412cd5b0 + 0x0000f> in <filename unknown>:0
    at MatterHackers.Agg.SystemWindowCreator_WindowsForms.ShowSystemWindow (MatterHackers.Agg.UI.SystemWindow systemWindow) <0x412035e0 + 0x00391> in <filename unknown>:0
    at MatterHackers.Agg.UI.SystemWindow.ShowAsSystemWindow () <0x41203570 + 0x0002a> in <filename unknown>:0
    at MatterHackers.MatterControl.MatterControlApplication.get_Instance () <0x41086460 + 0x0004b> in <filename unknown>:0
    at MatterHackers.MatterControl.MatterControlApplication.Main () <0x4101dea0 + 0x0020f> in <filename unknown>:0

    "It is possible to make no wrong moves and still lose, that is not a failing that is LIFE."


  • Happens at random, usually during a longer print.

    I can get smaller prints to work just fine. I can get smaller prints to work off of the SD card. I cannot however get anything over about 10K to print from either the SD card or Matter Control on Linux.

    Worked JUST GREAT under Windows, never had a lick of trouble.

    Now... the problem is we are without a Windows computer for at LEAST six weeks possibly longer. We are counting on our Robo R1 to continue to fulfill orders during this time. If I don't get the orders out we don't get paid and so on so forth.

    Where I am having a major issue here, I was told flat out that Matter Control was fully developed and deployed on Linux, that it worked flawlessly and that there would be zero issues running under Ubuntu. Now I am finding out this is not the case? The program is not in fact in a stable configuration and there is no plan currently to work through the issues? Am I understanding this correctly?

    If this is in fact the case this is a huge blow to our business. I do not have enough Matter Control Touch tablets to keep everything running, and those are just as buggy, just as prone to "blowing up" mid-print for no detectable reason. ESPECIALLY on longer prints.

    We need a solution. We do not have the cash-flow to fix our Windows computer at the moment, we are stuck using Linux.

    I am open to suggestions that do not involve spending money that we simply do not have, or involve pirating software.

    "It is possible to make no wrong moves and still lose, that is not a failing that is LIFE."


  • I'm very sorry that you were told that the Linux port was that stable. It is still very much experimental at the moment.

    In our testing, this problem has come up approximately once a month on a machine with heavy use. Unfortunately, it will take some time to deal with it. I'd still like to do everything we can to help you get printing. Printing from SD is usually very reliable. What kind of issues were you having with it? If we can get that sorted out, I think it would be your best option. Also, you can always just generate your gcode in MatterControl, then use some other host software for running the printer, like Pronterface.



  • I monitor all my prints in the Terminal as they run.

    Basically I sliced from Matter Control, copied the G-code to the SD-card, put that in the printer and loaded it. I start printing, get 3-4 layers in, and the printer stops. Scroll back through terminal it's stuck in a loop UNKNOWN COMMAND (numbers) ""

    Only does that from the SD card, and only from the Linux version. Windows it worked great, we used it on large parts.

    For the sake of troubleshooting I did test it with three different cards and multiple files. All sliced from Matter Control, all loaded onto the SD card... loaded into the printer, then launched from Matter Control.

    I'm wondering if the slicing itself is bad, and if that isn't what is causing these blow-ups? If the slicing engine is generating gibberish-code that would cause all kinds of exceptions, loops and failures. I am going to try getting Cura up and running on here so I can slice some files as a test, then feed them to Matter Control and see if that makes any difference.





    "It is possible to make no wrong moves and still lose, that is not a failing that is LIFE."


  • It would help if we knew what unknown command it was referring to. Could you export and post your print log?


linux 14
11
Posts
6777
Views
Log in to reply