Hi all,
I am currently using a combination of lprng+ifhp to print from a bunch of
Mac
clients to a Linux printserver, connect with a NIC (point to point) to the
plotter. The graphics printed are very huge in size, ranging from 200 to 500
Mb in size. There's no way to reduce the size of the works, as they are used
for light shows (It' a long story...). They get printed on a HP3500CP
plotter
with a Jetdirect (10Mbit only) interface, using the internal lpd server.
IFHP
is needed to convert the binary postscripts from the clients, before being
sent to the plotter. Here's the facts:
-------- -------- ---------
| MACs |------| Linux |----|Plotter|
-------- --------- ---------
ATALK TCP/IP
lp:HP3500CP:\
:lpr_bounce:\
:lp=raw@PLOTTER:\
:af,lf to their dir:\
:if=/usr/lib/filters/ifhp -Tmodel=hp2500cp,tbcp,status@:\
:of=/usr/lib/filters/ifhp -Tmodel=hp2500cp,tbcp,status@:\
I don't know if this is the EXACT printcap, but I clearly remember the if
and
of entries, the lpr_bounce, and the lp entry. If needed, i'll try to get the
exact one.
The work comes from the Mac, and is received in the /tmp directory by the
Linux papd. When the client transfer is over, the file is transferred to the
spool directory, where lpd starts the IFHP filter. This process creates
another huge file of about the same size in the spool directory, which is
what
is really transferred to the plotter. At the end, the job is cleared, and so
is the temporary file.
Here's the first question: is there any way to create the IFHP temp file in
another place, so that the /var/spool/lpd/lp directory can have more space
for
other jobs, instead of having a double huge file? The tempfile=/tmp line
option
is not considered when the filter is actually run.
Second problem: the plotter has started behaving in a strange way: if the
internal queue is enabled, it takes an enormous time to print out the job,
even after the receiving process is complete. This is not the normal
behaviour. If that queue is disabled, the print job starts just after it has
arrived. Is there anything in the ifhp conversion process that can cause
this
behaviour? Any other experience?
Third problem: in a 10Mbit link, the transfer rate is about 180Kbyte/s. Is
it
normal for a Jetdirect card? Do you think that upgrading to 100Mbit should
be
good? Could it be a linux networking problem (the server has two 3com 905
NICs, one speaking Appletalk, one TCP/IP)?
As you can understand, this work has been not "usual", everything is
working,
but performance is still an issue. Any clue, or experience, will be welcome.
The server is a PIII 500Mhz with SCSI Ultra2 drives, adaptec controller,
256Mb
RAM, so hardware should not have performance problems....
Thanx all
________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com
-----------------------------------------------------------------------------
If you need help, send email to [EMAIL PROTECTED] (or lprng-requests
or lprng-digest-requests) with the word 'help' in the body. For the impatient,
to subscribe to a list with name LIST, send mail to [EMAIL PROTECTED]
with: | example:
subscribe LIST <mailaddr> | subscribe lprng-digest [EMAIL PROTECTED]
unsubscribe LIST <mailaddr> | unsubscribe lprng [EMAIL PROTECTED]
If you have major problems, send email to [EMAIL PROTECTED] with the word
LPRNGLIST in the SUBJECT line.
-----------------------------------------------------------------------------
