*BSD News Article 19854


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!elroy.jpl.nasa.gov!swrinde!cs.utexas.edu!uunet!ddsw1!lll-winken.llnl.gov!fnnews.fnal.gov!cdfsga.fnal.gov!dejan
From: dejan@cdfsga.fnal.gov (Dejan Vucinic)
Newsgroups: comp.os.386bsd.questions
Subject: date+socket+getpgrp()+tip+disklabel+ctwm...
Date: 23 Aug 1993 00:37:02 GMT
Organization: Fermi National Accelerator Laboratory, Batavia IL
Lines: 82
Distribution: world
Message-ID: <2593fe$f32@fnnews.fnal.gov>
NNTP-Posting-Host: cdfsga.fnal.gov


Hi All,

Here I am back to the *BSD land after quite a while, setting up 
NetBSD 0.8 on a brand new messdossless system (lucky me :),
and it's time to look into the net.wisdom...

1. Isn't that funny date function fixed yet? I remember having
   problems with it in 0.1 days. When I link /etc/localtime 
   correctly, recompile kernel with correct timezone, and set
   time as root, it sets the current time during that session,
   but after reboot the time goes exactly 24 hours and 47 minutes
   into future and stays that way no matter how many times I reboot.
   Of course, the problem is that date sets the wrong time in the
   system clock, so at boot it picks that time and doesn't touch it.
   If I set the system clock from BIOS it reads the correct time and
   works fine ever after (unless there is different timezone set in
   kernel and link).
   The Nemeth/Snyder/Seebass say: "Like most time-sensitive systems,
   UNIX keeps track of time using Greenwich Mean Time." So where's
   the screwup? My system time certainly doesn't look like GMT, 
   neither does the time read from system clock at boot. Can anyone
   explain this? (And along the same line, why does BSD have TWO
   timezone references, one in the kernel and other in the filesystem,
   in the first place?)
   Not a big problem, but tends to annoy make. :)

2. I recompiled the kernel with option INET turned ON and got about
   two screenfuls of "_something_: socket: protocol not supported."
   at boot. Was I absent for too long? :) Missed something?

3. Ah that tip again. But this time it's beyond me. I get 
   "/dev/com1: Permission denied
    link down"
   and the permission is NOT denied. Yes I did recompile tip. It's 
   all the same. Com port is okay, mouse in X works fine.
   Subquestion: anyone use _dialback_ slip? Kinda cuts down phone
   bills and they're happy to provide it, cheaper for them than
   insecure dial-in ;) (those .gov addresses...).

4. My disk didn't experience a byte of DOS (what a wonderful life
   for a disk), but it obviously has caching enabled, and probably
   translation. What is the canonic way of finding out those things
   and setting the correct parameters? Disk is Seagate, 3???, 249Mb,
   1001 cyl, 34 sec, 15 heads. Nothing came with it.

5. Disklabel and newfs: I try to make a new floppy and disklabel says:
   overwriting disk with DOS partition table? (n): y
   ioctl: DIOCSDINFO: Device not configured.
   ioctl: DIOCWLABEL: Device not configured.
   fd0: hard error (ST0 40 ST1 1 ST2 0 ST3 0 cyl 0 hd 1 sec 2)
   and a bunch of same lines with hd 1, 5, 1, 5, 9, 9, 13 and 13.
   When I do cat /usr/mdec/bootfd and fdboot on /dev/rfd0a I get
   only the hard error lines.
   When I do a newfs on the resulting disk I get 
   write error: 2879
   wtfs: Input/ouput error
   What am I doing wrong?

6. Is there a Lucid 19.6 (or any other) port for 0.8? I am getting 
   segmentation faults when first trying to run temacs. (I did have 
   to poke quite a bit to make it compile, which brings me to:

7. What is the status of setpgrp() and getpgrp()? getpgrp() is still
   with an (int) argument in man page, but prototype sez (void). Who's 
   right? Does it act like sysV getpgrp now? 

8. I need a ctwm binary, mine just hangs there when I run it. I remember
   some discussion about yylineno from some days ago, but that doesn't 
   seem to be the problem.

9. If you managed to read the article up to this point here's a
   big thank you. If you have a solution for some of those problems 
   here's a HUGE thank you. There's obviously a LOT of work in NetBSD
   compared to 0.1, in the last couple of days I threw rocks at it,
   did two compilations at the same time under X in 8Mb RAM, it 
   yelled and screamed, but NEVER ever crashed in any way. Hey, that's
   something! We've got a real system here. Congratulations and thanks
   to all you girls and boys who brought it to this stage.

                       Best Regards,
                       Dejan