Problem with lpd going 'out to lunch' when JetDirect printer is not ready.

[Ken Wolff]

Running OSR505 with the following patches: rs505a, oss497c, oss471c, oss497b.

We serving up the printers on this system to others via lpr/lpd. All
printers are connected to the OpenServer box via JetDirect cards or boxes
and have been added under the HP print manager. The printer interface
we're using is 'laserjet IIIsi'.

The problem we're running into is when a remote print request is received
and the printer is offline (out of paper, jamed, so on). The lpd process
for that printer is running, but nothing ever prints. The print requests
queue up under /usr/spool/lpd. Once we kill the lead lpd process and the
process for that printer and restart lpd, everything works fine (until the
printer goes offline again).

Has anyone run into this problem before? Any suggestions on how to fix it?

Thanks!

--------------------------------------------------------------
Ken Wolff
Phone: 616-957-4949 Ext: 111
FAX: 616-957-1614
--------------------------------------------------------------

[David H. Funte]

Why use lpd? Why not set it up via the HP config scripts.
/usr/lib/hpnp/hpnpcfg sets up jetdirect perfectly, and doesn't use lpd.

Ken Wolff <ke...@cscc.maximus.com> wrote in message
news:5.0.0.25.2.200101...@scogr1.cscc.maximus.com...

[Ken Wolff]

Our OpenServer box uses the HP Print manager. However, we have a bunch of
other NT and Linux boxes that also need to use these printers so they use
lpr/lpd to print through the OpenServer box. We do this, as opposed to
having everyone print directly to the printer, for spool/request
management. There are times we want to prioritize some jobs over others,
or even hold jobs. The way we accomplish this is to have all servers print
through the OpenServer box.

[David H. Funte]

I seem to recall using lpd on one site. I think I had to set the timeout
setting to ZERO. Otherwise it would lose contact. I recall that this has
the effect of keeping the connection open.

[Jeff Liebermann]

On Tue, 9 Jan 2001 16:39:07 -0600, "David H. Funte" <none> wrote:

>I seem to recall using lpd on one site. I think I had to set the timeout
>setting to ZERO. Otherwise it would lose contact. I recall that this has
>the effect of keeping the connection open.

Nope. T'was an added delay. See SCO-559-811 below. This is from:
http://osr5doc.sco.com:457/OSr506/fixlist

----------------------------------------------------------------------

Bug-ID: SCO-244-322

rlp does not conform to rfc1179
rlp must use ports 721-731, inclusive, as the source port. Ours
starts at 1023 (IPPORT_RESERVED - 1) and works down, possibly to
0. Also our lpd accepts any port 0-1023; which it must, to talk
to our old version, but should have option to do the more secure

----------------------------------------------------------------------

Bug-ID: LTD-246-844

Trying to print 20 small files via rlp between OS5.0 hosts fails
results in 15 jobs appearing on remote queue. Need to stop and
restart tcp on lpd server to print remainder.

----------------------------------------------------------------------

Bug-ID: SCO-236-1585

lpd/rlp checks for incorrect fcntl() return value
if (errno == EAGAIN) /* active deamon present
Needs changed to EACCESS. This prevents:
lpd: fcntl: lock: permission denied message in syslog and
may help lpd from skipping jobs.

----------------------------------------------------------------------

Bug-ID: SCO-242-117

lpd does not properly move requests to local print server queue
not all moved into the local queue. The current printing job
must finish before the next job is moved from /usr/spool/lpd to
/usr/spool/lp/temp. This causes delays in printing as well as
erroneous print status from lpstat on the print server.

----------------------------------------------------------------------

Bug-ID: SCO-559-811

Needs doc of etc/default/lpd option PRINTJOB_LAG added 505
seconds to wait between servicing remote job requests
default (if otherwise unconfigured) is 30 secs. 0 to disable.

----------------------------------------------------------------------

--
Jeff Liebermann je...@comix.santa-cruz.ca.us
150 Felker St #D Santa Cruz CA 95060
831-421-6491 pager 831-429-1240 fax
http://www.cruzio.com/~jeffl/sco/ SCO stuff

[Jean-Pierre Radley]

Jeff Liebermann propounded (on Tue, Jan 09, 2001 at 07:53:49PM -0800):

|
| ----------------------------------------------------------------------
|
| Bug-ID: SCO-559-811
|
| Needs doc of etc/default/lpd option PRINTJOB_LAG added 505
| seconds to wait between servicing remote job requests
| default (if otherwise unconfigured) is 30 secs. 0 to disable.

A TA on PRINTJOB_LAG appeared yesterday, number 112162.

--
JP

[DiscGolfer]

Hi Ken,

My suggestion, lose the JetDirect and try RPM from Brooks....
http://www.brooksnet.com (I think). Easy and great support!

All the best,

T

In article <5.0.0.25.2.200101...@scogr1.cscc.maximus.com>,

--
+*+*+*+*+*+*+*+*+*+*+*+*+
PLAY DISC GOLF!
*+*+*+*+*+*+*+*+*+*+*+*+*

Sent via Deja.com
http://www.deja.com/

[Ken Wolff]

JetDirect works just fine with SCO directly, just not the servers printing
remotely to the SCO server. Actually, I don't believe it's a JetDirect
problem, just a problem with lpd where after a printer goes offline for a
while, lpd goes to "sleep" (or something) until it's killed and restarted.

Ken