Hide Forgot
Created attachment 560934 [details] troubleshoot.txt Description of problem: We have an HP Color Laserjet 5500DN, accessible over the network. In previous versions of Fedora, in particular in Fedora 14, this printer worked fine. In Fedora 16, the system insists the printer is black & white, which of course it is not. Version-Release number of selected component (if applicable): cups-1.5.0-22.fc16.i686 hplip-common-3.11.12-1.fc16.i686 hplip-libs-3.11.12-1.fc16.i686 foomatic-filters-4.0.8-4.fc16.i686 How reproducible: Always Steps to Reproduce: 1. Open Activities, Applications, Other, Printing 2. Select "Add printer", "Network printer" 3. Choose any protocol you like - lpd, ipp, HP JetDirect 4. Select "HP", "Color LaserJet 5500" from the driver list (CUPS+Gutenprint simplified) 5. Try to print in color The issue seems to be that the PPD in /etc/cups/ppd has: *ColorDevice: False This is, rather obviously, wrong. It's a colour device, as the name indicates. Actual results: Printing is black & white; all printer UI elements only offer grayscale Expected results: Colour printing Additional info: This worked in Fedora 14; I will attach a "troubleshoot.txt"
This printer works if I use the PPD from here: http://www.cups.org/ppd.php?L645+I0+T+Q5500 I've also noticed a 2nd issue - with the "built in" PPD, the test page right hand edge goes off the paper. With the PPD from cups.org (above) this doesn't happen. So it looks like the PPD for this printer is bad in F16. As mentioned, this printer worked fine in F14.
What does this command say?: /usr/lib/cups/backend/snmp 5500dn-nsg.cc.ic.ac.uk
That command prints nothing. Running: $ CUPS_DEBUG_LEVEL=2 /usr/lib/cups/backend/snmp 5500dn-nsg.cc.ic.ac.uk DEBUG: Scanning for devices in "public" via "5500dn-nsg.cc.ic.ac.uk"... DEBUG: Sending get request to 155.198.51.40... DEBUG: OUT Hex Dump (43 bytes): DEBUG: OUT 0000: 30 29 02 01 00 04 06 70 75 62 6c 69 63 a0 1c 02 0).....public... DEBUG: OUT 0010: 01 01 02 01 00 02 01 00 30 11 30 0f 06 0b 2b 06 ........0.0...+. DEBUG: OUT 0020: 01 02 01 19 03 02 01 02 01 05 00 ........... DEBUG: OUT Message: DEBUG: OUT SEQUENCE 41 bytes DEBUG: OUT INTEGER 1 bytes 0 DEBUG: OUT OCTET STRING 6 bytes "public" DEBUG: OUT Get-Request-PDU 28 bytes DEBUG: OUT INTEGER 1 bytes 1 DEBUG: OUT INTEGER 1 bytes 0 DEBUG: OUT INTEGER 1 bytes 0 DEBUG: OUT SEQUENCE 17 bytes DEBUG: OUT SEQUENCE 15 bytes DEBUG: OUT OID 11 bytes .1.3.6.1.2.1.25.3.2.1.2.1 DEBUG: OUT NULL VALUE 0 bytes DEBUG: 2.003 Scan complete! ...seems to suggest it's not responding to SNMP, which is what I'd expect for this printer.
Thanks. How about if you run this command?: /usr/lib/cups/backend/dnssd I'm trying to get it to tell us its IEEE 1284 Device ID.
This gives me: # /usr/lib/cups/backend/dnssd DEBUG: Found "hp color LaserJet 5500 (0001E69F45A8)._ipp._tcplocal"... DEBUG: Found "hp color LaserJet 5500 (0001E69F45A8)._pdl-datastream._tcplocal"... DEBUG: Found "hp color LaserJet 5500 (0001E69F45A8)._printer._tcplocal"... network dnssd://hp%20color%20LaserJet%205500%20(0001E69F45A8)._printer._tcp.local/ "hp hp color LaserJet 5500" "hp color LaserJet 5500 (0001E69F45A8)" "MFG:hp;MDL:color LaserJet 5500;FZY:1;" "" network dnssd://hp%20color%20LaserJet%205500%20(0001E69F45A8)._pdl-datastream._tcp.local/ "hp hp color LaserJet 5500" "hp color LaserJet 5500 (0001E69F45A8)" "MFG:hp;MDL:color LaserJet 5500;FZY:1;" "" network dnssd://hp%20color%20LaserJet%205500%20(0001E69F45A8)._ipp._tcp.local/ "hp hp color LaserJet 5500" "hp color LaserJet 5500 (0001E69F45A8)" "MFG:hp;MDL:color LaserJet 5500;FZY:1;" "" A couple of notes: Firstly, it is not impossible that one of my colleagues has helpfully (sigh) re-shared the printer, and these DNS-SD results are from their machine; how can I tell which IP address these correspond to? Second, I added the printer by known hostname, not by DNS-SD, but I assume this doesn't matter.
Those aren't CUPS DNS-SD replies. The "FZY:1;" tells me that CUPS is just pretending that there is a Device ID though, when in fact it's been made up. Never mind, I can see from the model name that gutenprint is not the recommended driver for this printer, and it is only being used because no other drivers are installed. The page boundary issue you mentioned in comment #1 sounds like it might be to do with locale settings: your system locale is set to en_US but your user locale is en_GB. Please adjust your system locale to match the user one. To do this in GNOME, visit System Settings, then Region and Language, then switch to the System tab. After you've done that, please restart cups (or else just reboot). Now, to try the recommended driver for this model: could you please install the "hpijs" package and then change the driver for the queue? The one you want is: HP Color LaserJet 5500 v3010.107 Postscript Does that give better results?
> Those aren't CUPS DNS-SD replies No. Hopefully, the printer is advertising itself via DNS-SD. Re-sharing it via CUPS is just confusing (especially when ~30 computers on the LAN do it, all with the same name). Anyway - the hpijs driver seems to generate much better output (though I'm still not convinced the page border is quite centred compared to previous versions, that could be a psychological issue at this end ;o) (I did fix the locale, but not via the GNOME tool, which seemed to just ignore my clicking of the "Copy settings" button. Install-time locale for Fedora LiveCD seems to be a bit broken generally, but that's off-topic for this discussion) That driver was also chosen by default when I selected "Change" for "Make and Model". This particular F16 system was installed from a Live CD. Is it possible that under previous versions, the "hpijs" package was installed on the Live CD by default, but is absent in F16 Live CD, hence the sub-optimal driver?
(In reply to comment #8) > > Those aren't CUPS DNS-SD replies > > No. Hopefully, the printer is advertising itself via DNS-SD. Re-sharing it via > CUPS is just confusing (especially when ~30 computers on the LAN do it, all > with the same name). The as-shipped configuration for CUPS in Fedora does not cause queues to be announced. This printer is advertising itself via DNS-SD but it is not including a Device ID in its information. There isn't really much we can do in the face of that. :-/ > This particular F16 system was installed from a Live CD. Is it possible that > under previous versions, the "hpijs" package was installed on the Live CD by > default, but is absent in F16 Live CD, hence the sub-optimal driver? Yes. The idea with leaving out printer drivers from the Live CD was that in most cases the correct driver is automatically installed as needed.