New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[RC Feedback] Feedback on the 1.3.10rc1 Release Candidate #2871

Closed
foosel opened this Issue Nov 6, 2018 · 67 comments

Comments

@foosel
Owner

foosel commented Nov 6, 2018

Please provide general feedback on your experience with the 1.3.10rc1 Release Candidate here. An "All is working fine" is valuable feedback as well, because it tells me that people in fact are testing the RC and just not finding any problems. Thanks :)

If you run into any obvious bugs not yet listed below the following line, please open a new ticket and follow "How to file a bug report".


Currently known bugs

Unreproduced issues

@hashashin

This comment has been minimized.

Contributor

hashashin commented Nov 6, 2018

Hi @foosel, seems like there's a problem with the gcode viewer in this version:
2018-11-06 13:19:26,075 - tornado.access - WARNING - 403 GET /downloads/files/local/honor7A.gcode (2a01:c50e:221e:7d00:90e5:9d8e:5a18:154e) 13.35ms
This happens with any file, reverting to 1.3.9 make it work again.
Salud.

@zeroflow

This comment has been minimized.

zeroflow commented Nov 6, 2018

Raspberry Pi 3 + MKS sBase: Connecting over serial didn't work, reported as issue 2872

@foosel

This comment has been minimized.

Owner

foosel commented Nov 6, 2018

@hashashin thanks, can confirm. Weird, I could still it was working last week and I didn't change the authentication stuff since then, at least not that I can remember. Will take a look.

@zeroflow will get back to you in the ticket.

@ctgreybeard

This comment has been minimized.

ctgreybeard commented Nov 6, 2018

After upgrade the Connect/Disconnect button shows Disconnect even though the printer is not connected.

I ran the upgrade and immediately after the Reload the button says Disconnect even though the printer isn't yet connected.
screenshot 2018-11-06 08 36 05
Clicking the button does not connect to the printer. It seems non-functional.

A reboot of the Pi doesn't bring it back. I don't seem to be able to connect to the printer any more.

@foosel

This comment has been minimized.

Owner

foosel commented Nov 6, 2018

@ctgreybeard what does the Terminal tab say? Also check your powering, I see an undervoltage indicator up there.

edit Also please check the JS error console and provide its contents. Might be that there's some JS error causing mayhem there. I cannot reproduce it locally, but maybe some plugin is causing trouble.

@ctgreybeard

This comment has been minimized.

ctgreybeard commented Nov 6, 2018

The terminal tab is blank and not allowing input. The power indicator isn't there now. I have it powered via a 2.4A adapter and although I haven't checked it in a while I was watching for power problems after upgrading the adapter and I currently get "throttled=0x0" for vcgencmd.

@foosel

This comment has been minimized.

Owner

foosel commented Nov 6, 2018

@ctgreybeard Anything in the JS console?

@ctgreybeard

This comment has been minimized.

ctgreybeard commented Nov 6, 2018

Nothing that I can see. Some view models that don't bind.

@foosel

This comment has been minimized.

Owner

foosel commented Nov 6, 2018

Could you please open a ticket on that and provide all the information that is requested (and check safe mode)? I cannot reproduce anything that sounds what you subscribe here, so I fear I need some logs, screenshots and all that, and that should go into its own ticket. Thanks!

@ctgreybeard

This comment has been minimized.

ctgreybeard commented Nov 6, 2018

Here's the JS console (I think) ... (You might need to guide me, I'm not a JS kind of guy ...)

packed_core.js?9d0e021e:14303 Starting dependency resolution...
packed_plugins.js?4331aa71:7898 [object Object]
packed_core.js?9d0e021e:14413 ... dependency resolution done
packed_core.js?9d0e021e:14716 Initial application setup done, connecting to server...
packed_core.js?9d0e021e:12699 Connected to the server
packed_core.js?9d0e021e:14694 Finalizing application startup
packed_core.js?9d0e021e:14578 Going to bind 42 view models...
packed_core.js?9d0e021e:14631 Did not bind view model TrackingViewModel to target #wizard_plugin_tracking since it does not exist
packed_core.js?9d0e021e:14631 Did not bind view model TelegramViewModel to target #wizard_plugin_telegram since it does not exist
packed_core.js?9d0e021e:14631 Did not bind view model SoftwareUpdateViewModel to target #softwareupdate_confirmation_dialog since it does not exist
packed_core.js?9d0e021e:14631 Did not bind view model SoftwareUpdateViewModel to target #wizard_plugin_softwareupdate since it does not exist
packed_core.js?9d0e021e:2267 User ctgreybeard logged in
packed_core.js?9d0e021e:14670 ... binding done
packed_core.js?9d0e021e:14690 Application startup complete
packed_plugins.js?4331aa71:8144 Testing token 329132562:AAHtqDtIKnBp6idgMcModJ0hdgfhdM4Aaaw
packed_core.js?9d0e021e:10373 ConfiguredLoggers has not changed. Not saving.
@ctgreybeard

This comment has been minimized.

ctgreybeard commented Nov 6, 2018

OK, I'll open a separate ticket a bit later today.

@kmanley57

This comment has been minimized.

kmanley57 commented Nov 6, 2018

I tried it on my Taz 6 about 4 hours ago and it would not connect to the printer, but the 3 Lulzbot Minis and a Taz 5 worked fine. So I reverted back for now on the Taz 6.

Setting the Baudrate to AUTO instead of 25000 fixed Connection timeout issue I had this morning.

@foosel

This comment has been minimized.

Owner

foosel commented Nov 6, 2018

@kmanley57 might be #2876, fixed for 1.3.10rc2

@jneilliii

This comment has been minimized.

jneilliii commented Nov 6, 2018

Please blacklist MQTT Publish Plugin version 0.6.0 on this release, it crashed hard. I think I have others but my pi died during testing and can't double-check until I get home to re-build. My suspicion is that the Tasmota-MQTT plugin may also be one causing problems.

foosel added a commit to OctoPrint/plugins.octoprint.org that referenced this issue Nov 6, 2018

@foosel

This comment has been minimized.

Owner

foosel commented Nov 6, 2018

@jneilliii done (well, as soon as the github pages build goes through)

@trendelkamp

This comment has been minimized.

trendelkamp commented Nov 6, 2018

Besides #2876 everything is working fine for me.

@JohnOCFII

This comment has been minimized.

JohnOCFII commented Nov 6, 2018

Other than #2876 and the GCode Viewer issue mentioned above, rc1 seems to be working well with my Prusa i3 MK3 with firmware 3.4.0. Tested Print History Plugin, OctoSlack Plugin, Timelapse, file uploading, and one 23 minute print.

@jneilliii

This comment has been minimized.

jneilliii commented Nov 6, 2018

@jneilliii done (well, as soon as the github pages build goes through)

So I got a new virtual environment up with all my plugins and there actually isn't an issue at all. I guess the errors I was running into were related to sd card corruption after all and not the plugins. Please remove from blacklist if you don't mind.

@ntoff

This comment has been minimized.

Contributor

ntoff commented Nov 7, 2018

Upon reboot I get hit with an anonymous usage enable / disable wizardy thingy. Clicking either enable or disable does.... something? nothing? There's no feedback as to whether or not my clicks are working. I'm not sure if that's intended behaviour but it'd be nice to see a little "tracking disabled" or "tracking enabled" text box or something, anything.

edit: scratch that, doesn't seem like my clicks are doing anything, can't get passed that wizard, watch this space...

edit2: never mind, seems i had something in config.yaml that was making octoprint fubar

softwareupdate:
  _config_version: 6
  check_providers: null <--- this, this right here... dirty little "NoneType is not iterable" error
  checks:
    octoprint:
      checkout_folder: /home/mint/OctoPrint
      prerelease: true
      prerelease_channel: rc/maintenance
  notify_users: false
@ramsesiden

This comment has been minimized.

ramsesiden commented Nov 7, 2018

I installed this version, I have Klipper installed raspberri pi3b+, once installed it couldn't connect anymore to mcu, timeout after timeout, even swapping port or changing port to tmp and usb001, never worked, I had to rollback to .09

foosel added a commit that referenced this issue Nov 7, 2018

SWU: Make sure check providers are in the correct format
Just in case there's some odd data in `config.yaml` as reported in #2871 (comment)

foosel added a commit to OctoPrint/plugins.octoprint.org that referenced this issue Nov 7, 2018

@foosel

This comment has been minimized.

Owner

foosel commented Nov 7, 2018

@ramsesiden probably #2872 - but just to confirm could you provide the terminal output or a serial.log from a successful connection under 1.3.9?

@ntoff hm... the question now is how that got there though... In any case, added a sanity check on that field for 1.3.10rc2.

@jneilliii done!

@ntoff

This comment has been minimized.

Contributor

ntoff commented Nov 7, 2018

@foosel probably when I manually deleted the custombackground plugin after it broke. You know how I love to manually break stuff to see what happens. Just so we're clear, this also broke 1.3.9 (and probably every other version), it wasn't unique to 1.3.10. I only just discovered it after updating to 1.3.10rc1. I'd say it's probably 90% chance it was all my own doing.

@arminth

This comment has been minimized.

arminth commented Nov 7, 2018

@foosel
Hi Gina,
I can confirm the issue that 1.3.10rc1 can't connect to klipper /tmp/printer anymore. Will try to roll back to 1.3.9 and get you the connect-log.

@arminth

This comment has been minimized.

arminth commented Nov 7, 2018

Rollback done!
Here you are.

Connecting to: /tmp/printerChanging monitoring state from "Offline" to "Opening serial port"Connected to: Serial<id=0x6e2922f0, open=True>(port='/tmp/printer', baudrate=250000, bytesize=8, parity='N', stopbits=1, timeout=10.0, xonxoff=False, rtscts=False, dsrdtr=False), starting monitorChanging monitoring state from "Opening serial port" to "Connecting"Send: N0 M110 N0*125Recv: okChanging monitoring state from "Connecting" to "Operational"

@foosel

This comment has been minimized.

Owner

foosel commented Nov 7, 2018

@arminth thanks! That confirms it then that it's #2872 and already fixed for rc2. Thanks!

The only thing that confuses me that I've definitely seen successful connects to Klipper in the new tracking stats... Only 3 instances, but that's around 3 more than I would expect if Klipper never sends a start message (because then I'd never be able to detect Klipper in the first place). Odd. In any case, will be fixed in rc2.

@foosel

This comment has been minimized.

Owner

foosel commented Nov 8, 2018

That's something covered in the "Heads-up" in the release notes - the plugin needs an update due to a backwards incompatible change in a dependency I needed to update. Best open a ticket on the plugin's repository.

@hamster65

This comment has been minimized.

hamster65 commented Nov 8, 2018

Cool! I have been waiting for the new version :) Thanks!!

The first update attempt did not work.
updatelog.log

The second try finished without errors. After a full reboot, the webpage loads and it seems to be connected, but all controls are disabled. The connection dialogue stays visible, the button reads "Trennen" (disconnect), but I cannot click it. The terminal tab shows no messages.

Could this be related to those messages?
tornado.access - WARNING - 409 POST /api/printer/command (127.0.0.1) 54.01ms

@foosel

This comment has been minimized.

Owner

foosel commented Nov 8, 2018

The first update attempt did not work.

The log appears to be incomplete 🤔

The second try finished without errors. After a full reboot, the webpage loads and it seems to be connected, but all controls are disabled. The connection dialogue stays visible, the button reads "Trennen" (disconnect), but I cannot click it. The terminal tab shows no messages.

Already covered in the list of "known issues" right in the initial post ;) See #2876

@duncanlovett

This comment has been minimized.

duncanlovett commented Nov 8, 2018

Since upgrading, I cannot connect to octoprint any more. I get the message:

The OctoPrint server is currently not running

If you just started up your Raspberry Pi, please wait a couple of seconds, then try to refresh this page.

If the issue persists, please log into your Raspberry Pi via SSH and check the following:

Verify that the process is running: ps -ef | grep -i octoprint | grep -i python should show a python process:
pi@octopi:~ $ ps -ef | grep -i octoprint | grep -i python
pi 1441 1 6 11:12 ? 00:00:15 /home/pi/oprint/bin/python
/home/pi/oprint/bin/octoprint --host=127.0.0.1 --port=5000
If it isn't, the question is why. Take a look into ~/.octoprint/logs/octoprint.log, there might be an error logged in there that helps to determine what's wrong.
You might also want to try if you can restart the server (if no obvious error is visible): sudo service octoprint restart.
If all that doesn't help to trouble shoot the issue, you can seek support on the OctoPrint Community Forum. Please provide your OctoPi and OctoPrint versions as well as your octoprint.log and explain what you already tried and observed as detailed as possible.

@hamster65

This comment has been minimized.

hamster65 commented Nov 8, 2018

The log appears to be incomplete thinking

I copied everything I thought relevant out of my terminal window.

Already covered in the list of "known issues" right in the initial post ;) See #2876

Ups, I indeed missed that. But I think my problem is different: I was able to connect with the OctoClient iPhone App. Its status said "detecting baudrate" for a long time and then changed to "Operational". I see the temperatures and I can home. Serial.log shows communications. But at the same time, the octoprint webinterface still shows the connection dialog and it appears to be offline. When I click on "Trennen", it disconnects but again nothing changes on the webinterface.

@foosel

This comment has been minimized.

Owner

foosel commented Nov 8, 2018

@duncanlovett sorry, but I need more than that. E.g. the results of the steps suggested on that very page you quoted. And the logs. "Doesn't work anymore" isn't going to solve this, especially considering that I'm currently seeing around 380 instances online that enabled the new tracking plugin and where the update apparently worked fine.

@hamster65 Please don't copy-paste a log from the output in your SSH client, download the full log file from the server ;) You can do that also through the UI of OctoPrint itself, when it's running, otherwise use an SCP client (e.g. WinSCP under Windows).

But I think my problem is different: I was able to connect with the OctoClient iPhone App

Doesn't sound like a different issue at all since the problem lies with the web interface, not the backend. Any API clients like e.g. such an app won't be affected.

@hamster65

This comment has been minimized.

hamster65 commented Nov 8, 2018

That explains it, I guess. Logs attached.
logs.zip

@foosel

This comment has been minimized.

Owner

foosel commented Nov 8, 2018

@hamster65 Thanks.

So... funny thing... this error there on the first attempt, I have code in OctoPrint to detect it. However:

2018-11-08 16:48:37,391 ! OSError: [Errno 2] Datei oder Verzeichnis nicht gefunden: '/home/pi/oprint/local/lib/python2.7/site-packages/sarge-0.1.4-py2.7.egg'

I did not expect German locale in that error message, so the mechanism in place to detect this error and work around it doesn't take hold ;) That's not a new issue in 1.3.10 though, and I usually keep the fixes in newer RCs to regressions. I'll have to think a) how to detect this kind of error reliably without relying on localized language and b) whether I dare to put a potential fix into rc2 or not.

In any case, funny bug.

@arminth

This comment has been minimized.

arminth commented Nov 8, 2018

@arminth thanks! That confirms it then that it's #2872 and already fixed for rc2. Thanks!

The only thing that confuses me that I've definitely seen successful connects to Klipper in the new tracking stats... Only 3 instances, but that's around 3 more than I would expect if Klipper never sends a start message (because then I'd never be able to detect Klipper in the first place). Odd. In any case, will be fixed in rc2.

Gina,

maybe it's because we are using an MKS GEN 1.4 running Klipper?
I just received a FYSETC F6 board. If I find the time over the weekend, I will install Klipper, do a test with it and let you know.
Tschüssi!
Armin

@arminth

This comment has been minimized.

arminth commented Nov 8, 2018

Couldn't wait and quickly wired it up on the desktop!
Same issue with a Fysetc F6 and Klipper. Connection timeout.
Grüßle
Armin

@GhostlyCrowd

This comment has been minimized.

GhostlyCrowd commented Nov 8, 2018

I am getting this for some reason.

T0 reported as invalid, reverting to T0

T0 exists in my firmware and i only have one

@foosel

This comment has been minimized.

Owner

foosel commented Nov 9, 2018

Anyone running into any kind of issues, I can't look into them if you don't open a proper bug report for them. That means logs (octoprint.log, JS console output and if the printer is involved also serial.log), steps to reproduce, your hardware, all that. In short, a proper fully filled out bug report. Just throwing "X doesn't work" at me won't allow me to look into that, and I'm starting to feel like a broken record having to constantly reiterate this 😕 I want to fix anything that is observed in this RC, that is the whole point of putting it out in the first place, but you need to help me with that and not have me pull every tiny bit of information out of your nose...

@ThaliaFromPrussia

This comment has been minimized.

ThaliaFromPrussia commented Nov 9, 2018

@ThaliaFromPrussia I can't reproduce the reload prompt on backup download - can you reproduce it reliably? Otherwise my money would be on some fluke...

backup2
It did it again and also after a full restart. I wanted to download the log file and it did it again.
In safe mode it didn't.
octoprint.log

@foosel

This comment has been minimized.

Owner

foosel commented Nov 9, 2018

In safe mode it didn't.

Then it's one of your installed third party plugins - if an issue can't be reproduced in safe mode but can when not in safe mode, in 99% of cases a third party plugin is to blame (that's btw also why the bug template asks whether an issue reproduces in safe mode - if it doesn't I usually can't fix it since it's caused by a third party plugin and the author of THAT needs to fix it ;))

@benlye

This comment has been minimized.

Contributor

benlye commented Nov 9, 2018

Looks like I need to update the firmware updater plugin due to the Sarge change. Seems simple enough, but I guess the best thing for compatibility is to use octoprint.util.commandline.CommandlineCaller and it looks like that class has no documentation. Is there an example somewhere I could take a look at?

@foosel

This comment has been minimized.

Owner

foosel commented Nov 9, 2018

@benlye oh, thanks for the heads-up, that lack of documentation should be fixed. In the meantime take a look at the software update plugin (search for CommandLineCaller and caller), it's using the class for running update scripts, the PipCaller subclass and its use by the plugin manager plugin for installing/uninstalling plugins (search for LocalPipCaller and self._pip_caller), or the plugin dev cli that uses it for some stuff (search for CommandLineCaller and self.caller).

It has the advantage that a lot of the boilerplate needed to not only run commands but actually do something with their output (e.g. forwarding it to the frontend) is already taken care of. Downside is that if you need to run something non-blocking, you'll have to wrap it in your own thread (I might add another parameter to work around that though). Otherwise it just forwards any additional keyword args to sarge which in turn forwards stuff to the internally used Popen.

@EddyMI3d

This comment has been minimized.

EddyMI3d commented Nov 9, 2018

@ThaliaFromPrussia: I had that issue too and found out that in my case it is this: https://github.com/fabianonline/OctoPrint-Telegram

@ThaliaFromPrussia

This comment has been minimized.

ThaliaFromPrussia commented Nov 10, 2018

@EddyMI3d Thank you. I thought about it as well because it's no longer working every time.

@duncanlovett

This comment has been minimized.

duncanlovett commented Nov 10, 2018

It's ok I just flashed the sd card back to the previous version and all works fine again. I'll try the newer version next week when I have more time/

Cheers

d> @duncanlovett sorry, but I need more than that. E.g. the results of the steps suggested on that very page you quoted. And the logs. "Doesn't work anymore" isn't going to solve this, especially considering that I'm currently seeing around 380 instances online that enabled the new tracking plugin and where the update apparently worked fine.

@tedder

This comment has been minimized.

Contributor

tedder commented Nov 10, 2018

Has anyone been able to reproduce the "can't get past the 'enable or disable anonymous usage' page" issue? I don't see a ticket, so I assume not.

@tech-rat

This comment has been minimized.

tech-rat commented Nov 11, 2018

Upgrade smooth on Raspberry Pi 3 and Prusa MK3
Have the error where GCODE viewer won't load.
My Octoprint complains on each print that movement was detected but a G90/91 command was not detected however my output from slic3r has G90 near the top so I assume its a bug.

Thanks
Tech-Rat

@foosel

This comment has been minimized.

Owner

foosel commented Nov 12, 2018

My Octoprint complains on each print that movement was detected but a G90/91 command was not detected however my output from slic3r has G90 near the top so I assume its a bug.

That doesn't sound like anything that core OctoPrint does, sounds more like Octolapse.

@racenviper

This comment has been minimized.

racenviper commented Nov 12, 2018

RC has been working fine except for gcode viewer, this does not affect me as most of my prints exceed the max size for the viewer.

@tech-rat

This comment has been minimized.

tech-rat commented Nov 12, 2018

My Octoprint complains on each print that movement was detected but a G90/91 command was not detected however my output from slic3r has G90 near the top so I assume its a bug.

That doesn't sound like anything that core OctoPrint does, sounds more like Octolapse.

Yes, sorry Octolapse. I can activate it and when I click print it gives the error I mentioned.
Takes the snapshots and renders when fine.

Tech-Rat

@FormerLurker

This comment has been minimized.

FormerLurker commented Nov 12, 2018

@tech-rat, can you post an issue here please? Be sure to include your gcode file, the octolapse version, and your settings.json file. Thanks!

@foosel

This comment has been minimized.

Owner

foosel commented Nov 12, 2018

@EddyMI3d @ThaliaFromPrussia so can I consider "Reload prompt on backup download" to be a side effect of the Telegram plugin?

At everyone, I want to second the question from @tedder here:

Has anyone been able to reproduce the "can't get past the 'enable or disable anonymous usage' page" issue?

I have had four "reports" on this - and I'm using quotation marks here since three of those reports where more or less drive-by "it doesn't work"s so far without any kind of reproduction steps, logs and sometimes even basic human decency, and the fourth got fixed accidentally because it was thought to be caused by a broken install that was then solved before information could be collected. So right now I'm in the dark about this and would be grateful for any reproduction steps, JS console output and just collaboration really from anyone running into this ;)

@JohnOCFII

This comment has been minimized.

JohnOCFII commented Nov 12, 2018

At everyone, I want to second the question from @tedder here:

Has anyone been able to reproduce the "can't get past the 'enable or disable anonymous usage' page" issue?

It works fine for me. Using Safari 12.0.1 on MacOS High Sierra (10.14.1) if that matters. I enabled anonymous user tracking and continued with the rest of my OctoPrint setup.

@foosel

This comment has been minimized.

Owner

foosel commented Nov 13, 2018

1.3.10rc2 just got pushed out, new feedback ticket is here.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment