Cannot print on new printer because "Waste Toner Box (waste-toner) seems to be empty."

[addiks]

I have bought a new Color-Laser Printer (Kyocera ECOSYS PA2101cx), but my System is refusing to print on it because it thinks that one of the cartridges is empty, one named "Waste Toner Box" (waste-toner). But this is not a normal toner-cartridge, it is a waste-box for used toner, it is perfectly fine for it to be empty to be able to print. In fact if the waste-toner would be full it would be an error. The printer itself is able to print test-pages (using the on-device display).

CUPS seems to misinterpret the printer-status for this waste-toner cartridge and refuses to print on a perfectly working printer.

There might be better solutions, but one simple 95%-quickfix should be to just ignore cartridges with the word "waste" in them. If you need more information from me (to f.e. better identify waste-cartridges) then please tell me how I could debug this, I'm happy to help.

To Reproduce
Steps to reproduce the behavior:

1. Buy a Kyocera ECOSYS PA2101cx
2. Try to print anything
3. See error

Expected behavior
CUPS should be able to correctly interpret the completely fine printer-status that the waste-box is empty and print my pages.

Screenshots

System Information:

• OS: Tuxedo OS 12 (based on Ubuntu 24.04)
• KDE: 6.5.2 (Frameworks 6.19.0)
• Linux: 6.14.0-116036-tuxedo
• CUPS: 2.4.7-1.2ubuntu7.9 (cups-filters: 2.0.0-0ubuntu4.1)

Additional context
ChatGPT told me that cups-filters is responsible for interpreting the printer-status, which is why i created this issue here. If that was wrong, please point me to wherever this issue should be placed instead.

$ lpstat -l -p
Drucker Kyocera_ECOSYS_PA2101cx druckt jetzt Kyocera_ECOSYS_PA2101cx-31.  Aktiviert seit Di 30 Dez 2025 15:10:17 CET
        cfFilterExternal (ipp): Sleeping...  
        Geladenes Formblatt:
        Inhaltstypen: beliebig
        Druckertypen: unbekannt
        Beschreibung: Kyocera_ECOSYS_PA2101cx
        Alarme: cups-waiting-for-job-completed
        Ort: 
        Verbindung: direkt
        Schnittstelle: /etc/cups/ppd/Kyocera_ECOSYS_PA2101cx.ppd
        Bei Fehlern: kein Alarm
        Nach einem Fehler: fortfahren
        Erlaubte Benutzer:
                (alle)
        Erlaubte Formblätter:
                (keine)
        Banner erforderlich
        Zeichensatz-Set:
                (keine)
        Voreingestellte Gradation:
        Voreingestellte Seitengröße:
        Voreingestellte Porteinstellungen:

[addiks]

Excerpt from /var/log/cups/error_log:

2025-12-30.kyocera-printer-error.log

Relevant lines (as far as I can tell):

D [30/Dec/2025:16:11:02 +0100] [Job 34] cfFilterExternal (ipp): marker-colors 1setOf nameWithoutLanguage #00FFFF,#FF00FF,#FFFF00,#000000,none
D [30/Dec/2025:16:11:02 +0100] [Job 34] cfFilterExternal (ipp): marker-high-levels 1setOf integer 100,100,100,100,95
D [30/Dec/2025:16:11:02 +0100] [Job 34] cfFilterExternal (ipp): marker-levels 1setOf integer 92,92,92,95,0
D [30/Dec/2025:16:11:02 +0100] [Job 34] cfFilterExternal (ipp): marker-low-levels 1setOf integer 3,3,3,3,0
D [30/Dec/2025:16:11:02 +0100] [Job 34] cfFilterExternal (ipp): marker-names 1setOf nameWithoutLanguage Cyan TK-5490CS,Magenta TK-5490MS,Yellow TK-5490YS,Black TK-5490KS,Waste Toner Box
D [30/Dec/2025:16:11:02 +0100] [Job 34] cfFilterExternal (ipp): marker-types 1setOf keyword toner,toner,toner,toner,waste-toner

At least in this case, the waste toner is reported with a marker-color none instead of an actual color. Maybe this indicator (together with the name containing "waste"?) can be used to apply a different restriction for these cartridges?
The toner is explicitely marked with the marker-type 'waste-toner' (opposed to all others: 'toner', this should be enough to identify these cartridges and apply a different logic to them

Instead of marker-level must be above marker-low-level for printing to be allowed it should be marker-level must be below marker-high-level for printing to be allowed for waste cartridges.

[ValdikSS]

Your printer reports that it successfully printed zero pages and completed the job.

I wonder if this is an URF issue.
Please try this: OpenPrinting/cups-filters#660 (comment)

To be able to say something more precise, please do this: OpenPrinting/cups-filters#660 (comment)

[tillkamppeter]

Note that there are 2 types of consumables and IPP supports this fact:

• Those which come full and need to get replaced when empty, like ink or toner
• Those which come empty and need to get replaced when full, like a waste toner box

I do not know how in IPP it is told which of the 2 senses applies for each consumable (marker) item, looks like it is just the marker-type name, toner or waste toner. How does that work, @michaelrsweet ?

It seems that somewhere the waste toner box is misinterpreted and the new, empty box considered used up.

[tillkamppeter]

As the filters do not check or handle consumables, but this is actually done by the ipp backend and by CUPS itself I am moving this to CUPS ...

[tillkamppeter]

Could you run the command

driverless

It lists URIs of printers in your network. Find the one of your Kyocera.

Now run

ipptool -tv URI get-printer-attributes.test > attrs .txt

replace URI by the URI you have found in the first step and attache attrs.txt to this bug report.

[michaelrsweet]

@tillkamppeter Printer supplies are reported in two ways.

The first uses the CUPS "marker-xxx" attributes where the "marker-low-levels" and "marker-high-levels" values tell you whether a supply is filled or consumed.

The second uses the IPP "printer-supply" attribute, where the "class" property in the string value is 'supplyThatIsConsumed' or 'receptacleThatIsFilled'.

WRT the printer working or not working, supply levels are informational and do not prevent printing from occurring.

[tillkamppeter]

So @addiks the job should print independent of any report about the levels of the consumables. The mis-interpretation here is probably happening in the GUI (KDE) and is not more then displaying the waste toner level in a misleading way. This is a bug of KDE and should be reported to there.

The jobs not being printed is a problem independent of this one.

[szotsaki]

I can confirm that there're two different bugs: one with KDE display, the other is printer not printing.

In my case the latter was because firewalld didn't allow through mDNS (Avahi) traffic. You can verify it: if the following command returns empty, then Avahi discovery doesn't work and CUPS cannot find the printer:

avahi-browse --all --ignore-local --resolve --terminate

This is signified by this CUPS error in its web interface:

Unable to locate printer 'XYZ.local'

The solution, in my case, was to assign the local network to the home firewalld zone where mdns is enabled by default (and with it, making it active):

firewall-cmd --permanent --zone=home --add-source=192.168.1.0/24  
firewall-cmd --reload

After that, printing started to work.

Let me report these two bugs to KDE and firewalld next week.

[tillkamppeter]

@michaelrsweet should this perhaps be put somewhere into the documentation of CUPS? Nowadays where all printer discovery (even including USB, thanks to IPP-over-USB) is going via mDNS/DNS-SD.

[michaelrsweet]

@tillkamppeter If there is something generic/common we can document, absolutely.

[szotsaki]

I filed the following two bugs against their respective projects:

• 514525: KDE erroneously indicates a printer issue about empty waste toner box
• Add 192.168.0.0/16 to home network by default firewalld/firewalld#1534

[tillkamppeter]

I am currently chatting with KDE developer Mike Noe (in KDE's Matrix) and explaining to him the meanings of the attributes so that he knows how to fix the bug.

@michaelrsweet thanks for the explanations and the links to the documentation.

[tillkamppeter]

Mike Noe has posted Merge Requests for KDE, one fix for right this bug of receptacles not supported for Plasma 6.6 (which is after Feature Freeze) and a more sophisticated fix which also fixes that consumable warnings happened at hard-coded 3% instead of respecting the threshold reported by the printer for Plasma 6.7.

Plasma 6.6:
https://invent.kde.org/plasma/print-manager/-/merge_requests/298

Plasma 6.7:
https://invent.kde.org/plasma/print-manager/-/commit/c146e39ec9e73afb0916f6be332d405917b4bfa5

So the Kool Desktop Environment will be fixed soon ...