blbates@AERO4.LARC.NASA.GOV ("Brent L. Bates TAD/TAB ms294 x42854") (07/08/89)
1. I don't know what the differences are between lpr and lp.
I couldn't find anything to discribe how to use lpr, so
I haven't used it. I only use lp. I would suggest you
use lp only too. It gives you more options and more printers.
2. I don't think lpr uses the interface files. Like I said above,
I think you should forget about lpr and just use lp.
3. Is the times off by hours or minutes? If it is hours it could
be that the sh shell environment variable TZ is not set correctly.
This sets the time zone. If it is minutes, it could be the
difference between when you send it and when it finally gets printed.
Those are the only things I can think of.
It seems to me that when I changed the interface files, I had to also
shut the scheduler off and on. Maybe even change the model file instead,
and redo lpadmin again to set the interface. It seemed to behave as if
there are two copies of the interface files one in /usr/spool/lp/interface
and another someplace else that I couldn't find. It has been a while
since I messed with all that stuff, so I am not sure. I wished SGI had
used BSD lpr instead of System V. The best OS is BSD with some System V
extensions. System V is such a pain.
Well I hope that is some help. Sorry I haven't been any more helpfull.
--
Brent L. Bates
NASA-Langley Research Center
M.S. 294
Hampton, Virginia 23665-5225
(804) 864-2854
E-mail: blbates@aero4.larc.nasa.gov or blbates@aero2.larc.nasa.govjweldon@renegade.sgi.com (Jack Weldon) (07/11/89)
In article <8907080016.AA08798@aero4.larc.nasa.gov>, blbates@AERO4.LARC.NASA.GOV ("Brent L. Bates TAD/TAB ms294 x42854") writes: > 1. I don't know what the differences are between lpr and lp. > > 2. I don't think lpr uses the interface files. Like I said above, > I think you should forget about lpr and just use lp. > > 3. Is the times off by hours or minutes? If it is hours it could > > I wished SGI had used BSD lpr instead of System V. The best OS is BSD with > some System V extensions. System V is such a pain. The IRIS is SYS V with BSD enhancements...enough said. Many of our customers LIKE the SYS V lp system--It has much more flexibility than the BSD system as far as multiple printers on a single queue point, but alas, BSD users want their printcaps. The lpr source code I sent out to people who asked me for it was to allow SYS V users (this was originally written for AIX) to print to BSD-based printservers. Since I cannot distribute any of OUR source code, this seemed like an opportune time to send the lpr.src to others who might not know such a solution was available. This would allow a simple request to be generated for the lpd listening on the other end. Another way to do it is to setup a printer using mknetpr, and then hack around and make the script do a "rsh <> lpr" instead of rsh <> lp". A trivial issue. I could say the same about BSD being a pain to implement...Please tell me how you can get a Sun to print to a printer on a SYS V machine...? (Other than using "cat <> | rsh IRIS lp") Again, the issues here is not which is better--we will see what happens in ATT 4.0. In answer to your questions: 1) lpr really shouldn't be on the 3000's as far as I'm concerned--it's not on the 4D's, unless the owner actually ported the real BSD-lp*. 2) BSD-lpr uses /etc/printcap, whereas the lp on the 3000 and 4D machines use the interface files. 3) Since you posted a response to a question that we didn't have the pleasure of seeing, I can't touch this one. Why don't you try including a quick blurb about the question so the rest of us without the original posting can see what you're talking about. 4) "The best OS is BSD with SYS V enhancements". I'm not too sure about this. 5) In response to your other posting, "what does a psuedo lpr/lpq do? Well, I guess you already modified your printer scripts to do it for you. This is an actual port of a portion of the BSD-lpr code. I guess you don't need it. Doesn't sound like you want it either. Hope this explains the issues at hand for you and other users--sure has been popular--I have gotten at least 20 requests so far. I'm sure the printer tool folks within SGI are looking at new ways to make things like this available in the future--not for me to decide, though. Jack Weldon Communications Engineer SGI Geometry Hotline -- Cheers, Jack "Elvis is alive and well and living in a trailer park in Muncie, Indiana" [National Enquirer]
blbates@AERO4.LARC.NASA.GOV ("Brent L. Bates TAD/TAB ms294 x42854") (07/11/89)
Like I said I prefer it the other way, BSD with SYS V enhancements.
I really don't see the SYS V lp as being more flexible than the BSD.
I haven't seen anything on the SYS V lp that can't be done with BSD.
We have a Gould that is BSD with SYS V extensions and it has several
printing devices contected to it directly and we can send output to
other computers printing devices without any problem. I find the BSD
printcap easier to use than the SYS V interfaces.
I used mknetpr and then changed it so I could print files on our
Gould. It took a little while, but I finally got it to work. I would
have been easier of the 3130 had been BSD.
"how can you get a Sun to print to a printer on a SYS V machine...?"
Well the problem lies in the SYS V machine, not the BSD machine.
I guess I though he had sent a copy to info-iris. On point 3,
"3. Files printed using lpr have a date and time on them - but the time is
not the same as the system time! .... "
It seems that TZ wasn't set correctly for the filter and the times were
off by 3 hours.
Thanks for the info.
--
Brent L. Bates
NASA-Langley Research Center
M.S. 294
Hampton, Virginia 23665-5225
(804) 864-2854
E-mail: blbates@aero4.larc.nasa.gov or blbates@aero2.larc.nasa.gov