On Thu 12 Sep 2019 at 16:00:31 +0200, Vincent Danjean wrote: > Le 11/09/2019 à 20:27, Brian Potkin a écrit :
[Quite a bit of snipping] > > I set up a print queue with these Brother DCP-9020CDW drivers by using > > 'dpkg -i' with the two packages provided. It was automatically named > > DCP9020CDW. How did you get the queue name to be "brother"? > > I manually installed it if I remember correctly (it was a long time > ago). I took the software provided by Brother, but I repackaged it > so that it better respects the FHS. You can find them here : > https://people.debian.org/~vdanjean/debian/pool/main/b/brother-dcp9020cdwlpr/ > https://people.debian.org/~vdanjean/debian/pool/main/b/brother-dcp9020cdwcupswrapper/ > According to the date, I made this in 2014 and it was working since. I actually deleted the queue set up by Brother and created my own, testq. We are on the same page. > > Being "the same" is (I think) part of the issue. It shouldn't have been. > > The server is probed over IPP and CUPS generates a PPD for a discovered > > queue. This is not the same PPD seen from the server. cups-browsed is > > supposed to use the same PPD generator as CUPS (via cups-filters), so > > should produce the same PPD. I also get this disparity when cups version > > 2.2.8-5 is used. > > I'm pretty sure that I used the PPD from Brother (and the ones from Canon > at work) and that they were not automaticcaly discovered. You used the Brother PPD on the client? The whole point of cups-browsed is that it discovers the remote queue (or printer) and *automatically* sets it up > > For the moment, I am inclined to see the issue as concerning cups-filters. > > So, if I understand correctly : > - before, it worked correctly (with all options) because I manually installed > the PPD files on the laptop and cups/cups-browsed do not override them Not because you manually installed the PPD files on the laptop. There has been a change in behaviour in cups-filters/cups-browsed, but I cannot pin it down from the changelogs. Or maybe I am misunderstanding what is going on. Till, would you help out here, please? There should not be any Brother (or Canon) files on the client. The job should be processed completely on the server. This is why I wanted the outputs of 'lpstat -l -e' and 'lpstat -t' (which were not sent). > - now, cups/cups-browsed always (for remote queue) generate a PPD file > using cups-filters, and the generated PPD is not correct (it misses lots > of options available in the remote queue) With 'lpoptions -l -p testq' I get brian@test:~$ lpoptions -l -p testq PageSize/Media Size: 100x150mm 111.76x152.4mm 3.5x5 3.5x5.Borderless 3x5 4x6 4x6.Borderless 5x7 5x7.Borderless 5x8 8x10 8x10.Borderless *A4 A4.Borderless A5 A6 A6.Borderless B5 Env10 EnvA2 EnvC5 EnvC6 EnvChou3 EnvChou4 EnvDL EnvMonarch EnvPersonal Executive ISOB5 Legal Letter Letter.Borderless Postcard Postcard.Borderless Statement Custom.WIDTHxHEIGHT MediaType/Media Type: *Stationery PhotographicGlossy ColorModel/Output Mode: *RGB Gray Gray16 DeviceGray DeviceRGB AdobeRGB Duplex/Duplex: *None DuplexNoTumble DuplexTumble cupsPrintQuality/cupsPrintQuality: Draft *Normal High > If this is correct, what should be done ? > Is there a way to test what cups-filter generate ? For me: ippfind -T 5 gives ipp://,,, for the queue. Then driverless <URI> gives the generated PPD. > Which information is used by cups-filter to generate the PPD ? Whatever it gets from the printer or the queue. [Another snip] > Here are the results from another machine (buster) on the same local net. > If you want the result from the laptop, I can do it later. This is ok (for now). Thanks. Cheers, Brian.
