Re: Some print jobs stalling

["Pete Peterson" <petersonp genrad com>]

> From: John Derbyshire <john wtsys net>
> To: Redhat Install List <redhat-install-list redhat com>
> Subject: Some print jobs stalling
> 
> Under Red Hat 7.3 I have an HP LaserJet 6L attached locally and am using
> the ljet4 driver. I have printed a number of Postscript documents
> successfully by simply using lpr xxx.ps and allowing the default RedHat
> magic filter to choose the right processing. However, printing the
> 'iptables' man page from Emacs, for example, using either "Print Buffer"
> or "Postscript Print Buffer (B/W)" both result in a stalled print (after
> five or six pages but at seemingly unpredictable points). Sample lpq
> output follows:
> 
> # lpq
> Printer: hplj6l xxxxxx
>  Queue: 1 printable job
>  Server: pid 13933 active
>  Unspooler: pid 13934 active
>  Status: IF filter 'mf_wrapper' filter msg - 'Loading
> NimbusMonL-BoldObli font from
> /usr/share/fonts/default/Type1/n022024l.pfb... 3876308 2279841 1743000
> 392295 0 done.' at 11:06:27.560
>  Rank   Owner/ID                  Class Job Files                 Size
> Time
> stalled(627sec) root xxxxxx+932   A   932 (STDIN)              71879
> 11:06:26
> done   Usr1 xxxxxx+813            A   813 smbprn.000152.yI5gEq 51729
> 10:50:55
> # 
> 
> Doing 'man iptables | lpr' prints the man page without problems. Do I
> have to set up printing specially in Emacs? 
> 
> Many thanks in advance for any ideas.
> 
> -John
> 

It should NOT require special setup in Emacs.

I'm not sure how much help this is, but postscript-print-buffer after
M-x man iptables works for me in limbo2 with a REMOTE printer with no
special configuration in Emacs.   If I get a chance later, I'll reboot
and try on valhalla, but valhalla and limbo2 are both using Emacs 21.2.
Emacs shouldn't know/care how the printer is connected --- the OS print
subsystem should take care of those details.  Emacs just uses the 
operating system's print commands.

In my case, the printer is a networked Lexmark 'Optra N' that I'm accessing
through a remote lpd queue on a Solaris machine; the Linux machine
specifies "Postscript printer", since 'Optra N' isn't one of the listed
printers.

You might try making Emacs print its Postscript output to a file, to see
if that gives you any clues.  I'm not sure how to do that *directly*, but
you CAN do this:

  Make a file "/tmp/printit" containing "cat >/tmp/test.ps" and make it
  executable 

  In Emacs, set the variable "ps-print-command" to "/tmp/printit" and
  try "printing" the "man iptables" buffer with "postscript print buffer".

You should end up with a Postscript file in /tmp/test.ps.  You can sanity
check this file manually and then try printing it to your printer.  

   If the Postscript file appears incomplete, you should probably look for
   an Emacs problem (not likely).

   If the file looks ok and prints on the printer, there's something
   weird happening with Emacs' communication with the print subsystem.

   If the file looks OK, but the print job fails, try looking at the
   file with a Postscript viewer (e.g. gv) to see if that gives you
   any clues.

It's possible that there's some problem between the Postscript that Emacs
generates and your print driver or the printer itself.  Is that a 
real Postscript printer or is the driver doing the Postscript
interpreting?  I've never had any problems with postscript-print-buffer
in Emacs on a wide variety of machines on several versions of Unix/Linux
and even on Windoze.

-- 
    pete peterson
    Teradyne, Inc.; 7 Technology Park Drive; Westford, MA 01886-0033
    +1-978-589-7478 (Office); +1-978-589-2088 (FAX);
    pete peterson teradyne com or petersonp genrad com