Bug 196294 - http backend failed error
Summary: http backend failed error
Alias: None
Product: Fedora
Classification: Fedora
Component: cups
Version: 5
Hardware: All
OS: Linux
Target Milestone: ---
Assignee: Tim Waugh
QA Contact:
Depends On:
TreeView+ depends on / blocked
Reported: 2006-06-22 14:18 UTC by Mark Bidewell
Modified: 2007-11-30 22:11 UTC (History)
0 users

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Last Closed: 2006-07-17 17:32:06 UTC
Type: ---

Attachments (Terms of Use)
error log (71.41 KB, text/plain)
2006-06-22 22:28 UTC, Mark Bidewell
no flags Details
PPD I was using gave a foomatic-rip error when I tried using it on the server (26.51 KB, application/octet-stream)
2006-06-23 15:11 UTC, Mark Bidewell
no flags Details

Description Mark Bidewell 2006-06-22 14:18:44 UTC
Description of problem:
I am connecting an FC5 install (cups 1.2.1-1.2) to a FC4 prinst server (1.1.x)
the FC4 system prints fine locally, however, remote printing yields a "http
backend failed error" with an error about not having permissions on
/var/run/cups/certs/0.  SeLinux disabled.

Version-Release number of selected component (if applicable):
cups 1.2.1-1.2
worked fine with 1.1.x

How reproducible:

Steps to Reproduce:
Install printer (Samsung ML-2010)
Try to print
Actual results:
Error above

Expected results:

Additional info:
cups 1.2.x seem to be giving a lot of distros fits.  This particular problem
seems more similar to those experinced by users of Ubuntu Dapper Drake than
anything I have seen on FC5 forums

Comment 1 Tim Waugh 2006-06-22 16:36:00 UTC
Please include the entire error message.  Sounds like you are confusing two
message, one of them harmless.

Comment 2 Mark Bidewell 2006-06-22 22:28:04 UTC
Created attachment 131397 [details]
error log

This is the error log.	The message I described is toward the top of the file

Comment 3 Mark Bidewell 2006-06-22 22:32:53 UTC
Comment on attachment 131397 [details]
error log

The error did not appear in that file.	The error occurs on startup is:
D [22/Jun/2006:18:59:03 -0400] Loading attributes for job 7...
I [22/Jun/2006:18:59:03 -0400] Full reload complete.
I [22/Jun/2006:18:59:03 -0400] Listening to on fd 0...
E [22/Jun/2006:18:59:03 -0400] Unable to set ACLs on root certificate
"/var/run/cups/certs/0" - Operation not supported

Comment 4 Tim Waugh 2006-06-23 08:46:59 UTC
The message in comment #3 is harmless; ignore it.

The problem with the print job is that the final content type is being marked as
printer/Samsung, which the printer claims not to support.

Could you please attach the PPD you are using?  Thanks.

Comment 5 Tim Waugh 2006-06-23 10:55:52 UTC
Actually I don't think I need the PPD -- FINAL_CONTENT_TYPE shouldn't be set to
printer/Samsung regardless of that.

Comment 6 Tim Waugh 2006-06-23 11:08:50 UTC
Reported upstream:

Comment 7 Mark Bidewell 2006-06-23 15:11:45 UTC
Created attachment 131442 [details]
PPD I was using gave a foomatic-rip error when I tried using it on the server

I tried building cups 1.2.1 on the server and the ppd was rejected

Comment 8 Tim Waugh 2006-07-14 10:50:38 UTC
Please try this test update:


Does that fix the problem?

Comment 9 Mark Bidewell 2006-07-15 04:01:10 UTC
With the update the printer prints, however, instead of the document I get a
series of lines begining with PJL

Comment 10 Tim Waugh 2006-07-17 08:29:55 UTC
That sounds like the wrong driver is selected.  What printer model do you have,
and which PPD are you using to drive it?

Comment 11 Mark Bidewell 2006-07-17 14:25:01 UTC
I have a Samsung ML-2010.  Samsung claims linux support, but it does not work on
up-to-date distros and linuxprinting.org doesn't have an exact match.  I have
used ML-4500 and ML-1710 ppds from linuxprinting.org both worked with 1.1.x.  I
have also used an unofficial ML-2010 ppd which I believe was a 1710 ppd modified
to correct margin issues.

Comment 12 Mark Bidewell 2006-07-17 14:27:34 UTC
The ppd is attached in comment #7

Comment 13 Tim Waugh 2006-07-17 17:32:06 UTC
If that PPD worked with CUPS-1.1.x (comment #11) I can only think that it's a
gdi driver change in ghostscript.  Please file a separate bug report against

The original problem (the FINAL_CONTENT_TYPE being wrongly set for backends) has
been fixed for the next update, so I'm closing this bug report.

Note You need to log in before you can comment on or make changes to this bug.