Centos 5.6 and Cups 2.4.6

[Scott Hallenbeck]

> Scott Hallenbeck wrote:
>
> >> Scott Hallenbeck wrote:
> >>
> >> >>
> >> >> --Boundary_(ID_QpIPheQUAuyplDZmG8ES+g)
> >> >> Content-type: text/plain; CHARSET=US-ASCII
> >> >> Content-transfer-encoding: 7BIT
> >> >>
> >> >> On Jun 14, 2011, at 9:36 AM, Scott Hallenbeck wrote:
> >> >> > ...
> >> >> > D [14/Jun/2011:12:33:36 -0400] Print-Job client-error-bad-request:
> >> >> > Unsupported character set "iso-8859-1"! ...
> >> >> > What I am thinking is that the real issue here is: Print-Job
> >> >> > client-error-bad-request: Unsupported character set "iso-8859-1"!
> >> >> >
> >> >> > Since it seems to print 'local' jobs just fine, I can't help but
> >> >> > wonder if it's the print jobs coming in that are causing this.
> >> >> >
> >> >> > Any idea's on how I can fix this?
> >> >>
> >> >>
> >> >> Your clients are submitting print jobs with an unsupported character
> >> >> set. CUPS 1.4.x only supports "utf-8" (Unicode).
> >> >>
> >> >>
> ________________________________________________________________________
> >> >> Michael Sweet, Senior Printing System Engineer, PWG Chair
> >> >>
> >> >>
> >> >
> >> >
> >> > What I don't understand is that these files are all pdf type, and when
> >> > I push the same file through the printers from the command line they
> >> > print. If it were the files being sent, wouldn't it fail at the command
> >> > line as well?
> >> >
> >> > Thanks,
> >> > Scott
> >>
> >> Might it be that the remote system still has a CUPS 1.1.x running?
> >>
> >> Helge
> >>
> >
> > Thank you for your reply!
> >
> > And yes, the remote servers are running 1.1.22rc1 associated with CentOS
> > 4.5. Unfortunately I am not yet privy to the print path from the
> > application servers to the print server but I am told that if cups is shut
> > down on the apps server that everything stops printing.
> >
> > I thought of this too, but the only thing that doesn't make sense then is
> > that we have a CUPS server running cups 1.3.7 (centOS 5.5) and it works
> > fine with everything except our HP p3005 servers, which is the main reason
> > for trying to upgrade to 1.4.6.
> >
> > So as it stands, I need to have a version of cups that is more recent than
> > 1.1.22rc1 but can print to HP p3005 printers without jamming them up and
> > without updating their firmware.  CUPS 1.1.22rc1 works fine with these,
> > and the local prints I send from CUPS 1.4.6 works fine as well, it's just
> > CUPS 1.3.7 that doesn't.
> >
> > I am fairly new to CUPS in a large environment like this and really lost
> > on how to proceed from here.
> >
> > Scott
>
> Well, then, on all your boxes which are runnning cups 1.1.22 (or some other
> 1.1.x release), find all files named cups_xx (where xx is a language
> identifier like en, en_US, de, etc), presumably in the /usr/share/locale
> directory tree, and replace the first line of all these files (which will
> read iso-8859-15 or windows1251 or the like) by utf-8. And don't forget to
> restart cups on these machines.
>
> I suspect, your 1.3.7 has been a patched version, as you told it has been
> working with the remote machines.
>
> Helge
>

Thank you again for your reply!
When I researched this originally I looked on the cups server that's printing (the one using CUPS 1.4.6) and there was no cups_en file in /usr/share/locale/en, so I created one and put as the first line 'utf-8'.  But I never did check the application servers sending the files.

After your post here, I went and checked cups_en on the application servers and sure enough the first line was 'iso-8859-1'

Unfortunately I do not have a real test environment setup here where I can simply change that and test it, so I am yet unsure if this is what's causing the problem, but I highly suspect it is.

I am told there is some kind of cups java client running on the application servers where the print job is submitted, and that without cups running on the apps servers, nothing prints, so I have every reason to believe at this point that fixing the cups_en file will resolve our issue with cups 1.4.6.

I appreciate everyone who has responded to this thread, and I will update the thread once I have had an opportunity to test this.

Thank you,
Scott