Report forwarded to debian-bugs-dist@lists.debian.org, papowell@sdsu.edu, Sven Rudolph <sr1@inf.tu-dresden.de>:
Bug#10386; Package lprng.
debian-bugs-dist@lists.debian.orgpapowell@sdsu.eduSven Rudolph <sr1@inf.tu-dresden.de>
Sorry, this message was lost when this bug report was restored from a backup.
Acknowledgement sent to Bradley Bosch <brad@lachman.com>:
New bug report received and forwarded. Copy sent to papowell@sdsu.edu, Sven Rudolph <sr1@inf.tu-dresden.de>.
Bradley Bosch <brad@lachman.com>
Sorry, this message was lost when this bug report was restored from a backup.
Received: (at submit) by bugs.debian.org; 6 Jun 1997 05:45:09 +0000
Received: (qmail 31518 invoked from network); 6 Jun 1997 05:45:05 -0000
Received: from waldo-2.lachman.com (HELO waldo) (192.35.54.251)
by master.debian.org with SMTP; 6 Jun 1997 05:45:04 -0000
Received: by waldo
id m0wZroe-001eIHC
(Debian Smail-3.2 1996-Jul-4 #2); Fri, 6 Jun 1997 00:44:00 -0500 (CDT)
Message-Id:
Date: Fri, 6 Jun 1997 00:44:00 -0500 (CDT)
From: Bradley Bosch
To: submit@bugs.debian.org
Subject: Lprng flushes serial ports too soon, truncating output
Return-Receipt-To: brad@lachman.com
X-Mailer: VM 6.28 under Emacs 19.34.1
Bcc:
X-Debian-CC: papowell@sdsu.edu
Mime-Version: 1.0 (generated by tm-edit 7.105)
Content-Type: text/plain; charset=US-ASCII
Package: lprng
Version: 3.2.1-1
I had trouble with my serial connected HP Deskjet 500c printer not
printing the last several lines of text, or the last inch or so of
postscript. I tracked this down to the serial port buffers being
flushed by lpd just after the last write to the tty and before the
printer has a chance to drain the buffers.
I switched from the old lpr and apsfilter (which had lots of bugs) to
lprng and magicfilter when I upgraded to Debian 1.3 and I still run
kernel version 2.0.28. After fixing this serious bug and correcting
some other minor bugs by editing the /etc/lpd.conf file, I find this
combination works quite well. Most of the lpd.conf file changes
required have already been reported as debian bugs, but I also noticed
that "default_remote_host=localhost" is required if your network
interface is unavailable (unplugged PCMCIA card for example).
Here is a patch which will fix the problem. Note that we still flush
the tty in Print_abort() if the filters don't exit normally. (which,
if I understand the comments in the change log, was the reason for
adding this code).
--- pr_support.c.old Thu Jun 5 22:48:49 1997
+++ pr_support.c Thu Jun 5 22:49:05 1997
@@ -224,7 +224,6 @@
void Print_close(int timeout)
{
DEBUG3("Print_close: closing printers");
- Print_flush();
/* shut them all down first */
Close_filter( &Device_fd_info, timeout, "Print_close" );
Close_filter( &OF_fd_info, timeout, "Print_close" );
For completeness, here is the printcap entry:
ps|lp
:rw:sh:
:lp=/dev/ttyS0
:sd=/usr/spool/lpd/djetc500
:sy=19200 -echo -crmod -raw -oddp -evenp pass8 cbreak -ixon crtscts
:af=acct:lf=log:ps=status
:fx=flpv
:if=/usr/local/lib/magicfilter/dj500-filter
Thanks,
--Brad Bosch
brad@lachman.com