Spoolman Error

[claudermilk]

I keep getting this error from Spoolman when running jobs: "The amount of filament required to print the selected file is unknown. Do you wish to continue?" A firmware restart resolves it, but that's getting really old now. I am seeing this on two printers and both have been updated to the current revisions of Klipper, Moonraker, Fluidd, and Mainsail. Spoolman is an independent install on a different Pi server and I just updated that to the current version. This has been popping up since Klipper (or is it Moonraker) added the integration. I'd like to open a bug report, but I have no idea which GitHub project to open it on. Klipper? Moonraker? Fluidd/Mainsail? Spoolman? I have a feeling whichever I do I'll get pointed at one of the others, so any ideas which would be the best place for reporting this error?

 

[Spice_King]

I keep getting this error from Spoolman when running jobs: "The amount of filament required to print the selected file is unknown. Do you wish to continue?" A firmware restart resolves it, but that's getting really old now. I am seeing this on two printers and both have been updated to the current revisions of Klipper, Moonraker, Fluidd, and Mainsail. Spoolman is an independent install on a different Pi server and I just updated that to the current version. This has been popping up since Klipper (or is it Moonraker) added the integration. I'd like to open a bug report, but I have no idea which GitHub project to open it on. Klipper? Moonraker? Fluidd/Mainsail? Spoolman? I have a feeling whichever I do I'll get pointed at one of the others, so any ideas which would be the best place for reporting this error?

Not Klipper nor Spoolman at least. Klipper has no clue about Spoolman (past there is stuff you can cram into macros for talking to Moonraker about what the current spool is), and Spoolman has no clue about anything in Klipper land.

First stop is to check if your slicer and Moonraker are playing nice. If you don't see "Filament Usage", or it's wrong, the problem is somewhere between the two.


If the slicer and Moonraker agree, the bug is probably in Mainsail with a slim chance it's an edge case in Moonraker causing Mainsail to trip up, though personally, I've not had issues with it and Spoolman anyway.

As for why doing a firmware restart helps when Klipper it's self is a non-factor, it triggers somethings in Moonraker when Klipper becomes ready again and also makes Mainsail reload the connection.

 

[claudermilk]

Most of the time it works. I do have the usage value (I primarily use SuperSlicer, and am transitioning to Orca). For example a common scenario I have is I just printed a plate with no problem, I clear the parts and tell the printer to go again, then I get this error. I do firmware_restart and it's happy again. I'm also mainly using Fluidd for my UI. I also just completely reinstalled everything (from Debian on up) on one printer and it's still doing it. Ironically that reinstall was triggered by hitting the update button so I could get this integration with Spoolman.

So yeah, sounds like it's probably Moonraker or Fluidd for some reason not getting the filament use from the gcode file. Hmm, I'll have to remember to try just reloading the web page next time I see this error.

 

[claudermilk]

Aha. I am now pretty sure it's Moonraker. Got the error again. I tried a web page refresh, no dice. Changed the spool, no dice. Firmware restart got it going again. Now I look right in the moonraker logs, and sure enough there's errors. I'm seeing a bunch of JSON-RPC request errors with file metadata not available. I think I'm pretty safe reporting a bug to the Moonraker project now.

 

[StevenB]

Is it the Pi running spoolman going into a low power sleep state? You request data from it, it error but wakes the Pi and by the time the restart happens it's awake and accepting connections?

 

[claudermilk]

Nope. It's a standalone server that has it's own spec wall wart power supply. It's also my Plex server. Looking at the error message, it looks to me like Moonraker is losing track of the gcode file somehow. Spoolman is working fine; I can go to its web interface, I can print from another printer, I can change active spools.