You are not logged in.
Pages: 1
I tested a plate with "Stop Layers", and after "Resume Printing" nanodlp crashed.
Have attached "printer.log".
Offline
Mike,
We have similar issue listed here.
http://www.nanodlp.com/forum/viewtopic.php?id=81
Do you see any similarity with this case?
What kind of hardware / firmware do you use?
Offline
My hardware is Arduino with Ramps and marlin firmware,
Raspberry worked after the crash but quite normal.
However, I have for test reasons also an HDMI-VGA adapter in use to which a monitor is operated.
I will make futher tests without the HDMI Adapter, what I have read could it be a problem.
Could it also be a problem, if the resolution of the monitor differ to the resolution of the layers in the plate ?
Last edited by Xmike (2016-04-22 07:41:10)
Offline
I have the hardware now changed to a true HDMI connection and a monitor with the same resolution as the Layer (1920x1080).
nanoDLP unfortunately crashed again.
In layer 97 I had executed "Stop Print"
after "Resume Print" immediately before the layer should appear,
nanoDLP has crashed.
Raspberry but continues function normally,
I could start nanodlp again with "sudo ./printer" without a restart of RPI
Log:
nanoDLP_2_printer.zip
Last edited by Xmike (2016-04-23 10:34:17)
Offline
Current version is pretty stable, maybe it is a firmware issue. I will upgrade my raspbian to see if I could reproduce the issue.
Are you on raspbian lite or desktop? Which version are you using?
Could you share debug info? Setup page/tools tab/ debug button.
Offline
Im using "Raspberry Pi 2 Model B" with "Full desktop image based"
"Debian Linux 7.8" "Linux 4.1.13-v7+ on armv7l".
In this print-job, I have executed more "Stop Print" and "Resume Print" on different Layers without issues (about 5 times)
Offline
Are the crashes always happen after resume?
If you can use raspbian lite, use it. Raspbian jessie uses systemd which our current setup does not block x which could interrupt our image view backend.
Offline
It does not happen generally after each "Resume Print", but if it crashes,
then after "Resume Print" immediately before the layer should appear
Log:
nanodlp.debug.1153101349.zip
I will change to raspbian lite
Last edited by Xmike (2016-04-23 11:25:26)
Offline
Mike,
Before changing OS, do you have time to checkout patched version?
Offline
Mike,
Before changing OS, do you have time to checkout patched version?
If he doesn't I will. That was my thread you linked to up there.
Offline
Please, reinstall program again using command below. Let me know if it fixes the issue.
(wget http://www.nanodlp.com/nanodlp.debug.tar.gz -O - | tar -C /home/pi -xz);cd /home/pi/printer;sudo ./setup.shOffline
I already have a new installation of Raspbian light on a new
SD Card performed. Currently is running a test print and is
already at Layer 140
Although I already several times "Stop Print" and "Print Resum" have performed,
runs nanodlp without problems.
I have the previous installation on another SD card and can do further tests
Offline
Mike,
I could reproduce the problem, but after couple of thousands of layers.
So it will be really helpful if you could test.
Offline
Hello Shahin,
The issue seems to be solved now.
I have run a testprint with my previous desktop version and "nanodlp Build 1144"
and often operated "Stop Print" and "Resume Print" without any problem.
I'm really glad now.
What I also want to say,
with nanodlp you created a realy outstanding and highly professional
printing software, many thanks for it.
Offline
I've been having the same issue since about build 1131. Currently on 1140. Printer stops printing, Pi doesn't respond to http, but ssh and network are running. This is readily reproduceable for me. Let me know if there is anything you'd like me to test, otherwise I'm going to install 1144 and try that because people are reporting success....
I should note that the same setup was very stable before, so I don't suspect power other anything.
Last edited by mattcaron (2016-04-24 11:51:21)
Offline
Here is the printer.log from when it crashed.
Of note:
runtime: free list of span 0x71e92ac0:
0x118da680 -> 0x118db380 -> 0x10fe2d20 (BAD)
fatal error: free list corrupted
I thought Go was supposed to have automatic memory management so things like this didn't happen...
Offline
Matthew,
Yes, it is pretty annoying, I suspect data race condition. Hopefully it is not memory corruption due to gc.
I have patched above debug version. Could you checkout that? Unfortunately for some unknown reason I could not reproduce error.
Offline
Mike,
Could you try beta version too? I have re-enabled modules which was disable on debug version but changed their data structure.
Matthew,
If you are sure about problem in build 1131, most probable culprit is hardware monitoring module.
Hopefully it's issue has been solved in beta version if not I think we could do better without it.
Offline
Just the same build version with above wget command reinstall ?
or how to install beta ?
Last edited by Xmike (2016-04-24 15:47:22)
Offline
Mike
(wget http://www.nanodlp.com/nanodlp.beta.tar.gz -O - | tar -C /home/pi -xz);cd /home/pi/printer;sudo ./setup.shOffline
I repeated the test print with the beta (Build 1146) and could no more reproduce the issue. The issue is apparently really solved.
Offline
Is it fixed for everyone?
Offline
Pages: 1