Re: cups not sharing printers with other bullseye machines
On Sat 8 Apr 2023, at 22:12, gene heskett <gheskett@shentel.net> wrote:
> On 4/8/23 02:40, Gareth Evans wrote:
>> On Sat 8 Apr 2023, at 03:20, gene heskett <gheskett@shentel.net> wrote:
>>> Greetings all;
>>>
>>> Where do I turn on cups debugging so I'll see every bit of traffic
>>> addressed to cups from my local 192.168/xx.yy network?
>>>
>>> The problem is: other buster machines on this local network can see and
>>> use the two brother printers just as if the printer was local to that
>>> buster machine.
>>>
>>> But no bullseye, debian or armbian can see anything at
>>> localhost:631/printers except the search screen when there are no printers.
>>>
>>> These printers are marked as shared in this bullseye machines /etc/cups
>>> files.
>>>
>>> There's a new roadblock someplace, I've asked about before. I'd like to
>>> find it.
>>>
>>> Buster machines can, other bullseye machines can't.
>>>
>>> Thank you.
>>>
>>> Cheers, Gene Heskett.
>>> --
>>> "There are four boxes to be used in defense of liberty:
>>> soap, ballot, jury, and ammo. Please use in that order."
>>> -Ed Howdershelt (Author, 1940)
>>> If we desire respect for the law, we must first make the law respectable.
>>> - Louis D. Brandeis
>>> Genes Web page <http://geneslinuxbox.net:6309/>
>>
>> Hi Gene,
>>
>> Not sure if CUPS debugging may be helpful, see eg.
>>
>> https://sysadminera.com/2020/09/10/linux-how-to-enable-and-capture-cups-debugging-logs/
>>
> The most intelligent output I can get from the error_log on one of the
> armbian bullseye machines is a garbled attempt to open a pipe (I think)
> from that log a snippet is attached. Tail end of a cups restart. Looks
> like something in the name resolution is totally fubar to me.
>
> But, I can send ff to the exact entry in client.conf, and it can see all
> the shared printers here on this machine. But cups on that machine can't.
>
>> But first, I seem to recall you removed avahi and cups-browsed from Bullseye machines. Is that correct? Do the Buster machines have either or both of those installed?
>
> avahi and cups-browsed have both been re-installed on that armbian machine.
> And its networking continues to work thru a reboot, something I could
> not do at the original install.
> Then I check this machines error_log which is flooded with thousands of
> lines of this from my attempts to get a printer list on the armbian machine:
>
> E [08/Apr/2023:12:25:16 -0400] [Client 18] Returning IPP
> server-error-version-not-supported for CUPS-Get-Printers (no URI) from
> 192.168.71.12.
> E [08/Apr/2023:12:25:16 -0400] [Client 18] Returning IPP
> server-error-version-not-supported for CUPS-Get-Default (no URI) from
> 192.168.71.12.
> E [08/Apr/2023:12:25:16 -0400] [Client 18] Returning IPP
> server-error-version-not-supported for CUPS-Get-Printers (no URI) from
> 192.168.71.12.
> E [08/Apr/2023:12:25:16 -0400] [Client 18] Returning IPP
> server-error-version-not-supported for CUPS-Get-Default (no URI) from
> 192.168.71.12.
> E [08/Apr/2023:12:25:16 -0400] [Client 18] Returning IPP
> server-error-version-not-supported for CUPS-Get-Printers (no URI) from
> 192.168.71.12.
>
> So the armbian machine is trying, and its this machine that is rejecting
> its attempts. That's progress ;o)>
>
> Whats next?
>
> Thanks Gareth, take care and stay well.
>
> Cheers, Gene Heskett.
> --
> "There are four boxes to be used in defense of liberty:
> soap, ballot, jury, and ammo. Please use in that order."
> -Ed Howdershelt (Author, 1940)
> If we desire respect for the law, we must first make the law respectable.
> - Louis D. Brandeis
> Genes Web page <http://geneslinuxbox.net:6309/>
>
> Attachments:
> * cups-error-log
> server-error-version-not-supported for CUPS-Get-Printers (no URI) from
> 192.168.71.12.
The only reference to this error I could find was this
https://bbs.archlinux.org/viewtopic.php?id=168485
which (a long time ago) seems to have been resolved by using matching cups versions.
I find my Brother MFC-L2740DW is only detected on Bullseye if both devices use the same wifi band (the printer supports only 2.4GHz). I'm sure this was never a thing in Buster, but I may have switched bands at some point and forgotten about having done that. I imagine it could conceivably be due to a dodgy ISP router firmware update too.
Are either of the printers you want access to from Bullseye currently shared via Buster CUPS, or Bullseye CUPS, or just via own wifi?
HL-L2320D seems to be USB-only. MFCJ6920DW seems to support both airprint but this may be susceptible to the wifi band thing.
If shared via Buster CUPS, does Bullseye (with cups-browsed installed) fail to detect these queues?
Is one or more firewalls involved? Does it make any difference if this/these are temporarily disabled?
cups-browsed here doesn't interfere with Brother drivers, but only "applies" them to printers specifically added with those drivers selected. As I can't reproduce the behaviour you report, are you by any chance expecting airprint/driverless printers already detected to use the Brother drivers after the drivers have been installed?
I have a couple of hunches to test so I'm trying to set up a Buster VM with bridged networking so accessible from "outside". The bridged networking part is proving more challenging than several sets of instructions would appear to indicate. I think I shall abandon virt-manager and try qemu with -netdev tap or macvtap, but this will have to wait a couple of days. If anyone can recommend approaches or good, up-to-date documentation, that would be appreciated.
So I hope to be able to dig a bit further on this.
Best wishes,
Gareth
Reply to: