From:     Digestifier <Linux-Admin-Request@senator-bedfellow.mit.edu>
To:       Linux-Admin@senator-bedfellow.mit.edu
Reply-To: Linux-Admin@senator-bedfellow.mit.edu
Date:     Sun, 7 Nov 93 04:27:01 EST
Subject:  Linux-Admin Digest #144

Linux-Admin Digest #144, Volume #1                Sun, 7 Nov 93 04:27:01 EST

Contents:
  Re: Net2? in pl13p with errors (Michael O'Reilly)
  Re: Does any *human* know how NYS works? (Michael Griffith)
  Finger and Last problems (root)
  Re: Finger and Last problems (Mike Horwath)
  Windows NT and Lilo (Martin Sckopke)
  Re: Windows NT and Lilo (Crassly Commercial Quantum Mechanic)
  what should hostname report? (John Paul Morrison)
  mv feature? (Stephen Harris)
  syslogd that handles /proc/kmsg ? (Chris Metcalf)
  Re: Finger and Last problems (Robert Bauer)
  Re: Finger and Last problems (Robert Bauer)

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

From: oreillym@tartarus.uwa.edu.au (Michael O'Reilly)
Subject: Re: Net2? in pl13p with errors
Date: 6 Nov 1993 03:30:08 GMT

Volkmar Eich (volkmar@gatekeeper.rhein.de) wrote:
: Hi,

: As I got to know in pl13p (including net2debugged) this missing feature
: might be implemented. Now I got the following error message:

: <6>*** tcp.c:tcp_data bug acked < copied

: Can somebody help me. I know that this occurs when I should receive SMTP-
: Packets and try to connect to my newsserver.


linux/net/inet/tcp.c:
line 2521, change...
        skb2 !=(struct sk_buff *) sk->rqueue->next;
to
        skb2 !=(struct sk_buff *) sk->rqueue;


That fixed it for me.. 

: Thanks for your assitance.

: Volkmar

Michael.

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

From: grif@ucrengr.ucr.edu (Michael Griffith)
Subject: Re: Does any *human* know how NYS works?
Date: 6 Nov 1993 08:44:03 GMT

In article <9311052304.AA00978@fearless.saintjoe.edu>,
Brian Capouch <brianc@fearless.saintjoe.EDU> wrote:
>
>I'm trying as trying to get the NYS "yp" interface hacked into my linux box.
>
>I've gotten the code, and ypcat, etc. seem to work fine.  However, I'm sure 
>that somehow the login program must have to know about using the yp 
>maps instead of the contents of the passwd and shadow files. 
>
>Mail to the NYS developers has gone unanswered.  No one knows for certain how 
>to tap into the NYS developers' channel (at least, no one who I've contacted)
>and unless I can find some kind of documentation about how to use this stuff, 
>it's not going to be of much use.  There's essentially no documentation 
>included with the distribution.  I have version 11. 
>
>I'd be much obliged for any pointers anyone out there could give me? 

As far as I know, it is not finished.  The interface into the C
library needs some simple (yet time-consuming) code added to it.  If
you are interested, send mail to:

        Peter Eriksson
        pen@lysator.liu.se

--
                                                Michael A. Griffith
                                                grif@cs.ucr.edu






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

From: root@plisgyn.demon.co.uk (root)
Subject: Finger and Last problems
Date: Sat, 6 Nov 1993 10:55:48 +0000

 When using finger on its own it shows logged in users o/k but if you finger
 a user (who is not logged in) it says they have never logged in. Last gives
 times for reboot ( these are highly suspect ) but every user has crash in
 place of its second time. I believe that the problem is that the programs
 that write to wtmp use a format incompatible with those that read it. Can
 anyone help with informationon correct versions to use to get correct 
 operation of finger and last. I am running pl11 (SLS1.03 upgraded) with
 shadow passwds.

                            Phil Jones. 

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

From: root@jacobs.mn.org (Mike Horwath)
Subject: Re: Finger and Last problems
Date: 6 Nov 1993 16:13:30 GMT

root (root@plisgyn.demon.co.uk) wrote:
:  When using finger on its own it shows logged in users o/k but if you finger
:  a user (who is not logged in) it says they have never logged in. Last gives
:  times for reboot ( these are highly suspect ) but every user has crash in
:  place of its second time. I believe that the problem is that the programs
:  that write to wtmp use a format incompatible with those that read it. Can
:  anyone help with informationon correct versions to use to get correct 
:  operation of finger and last. I am running pl11 (SLS1.03 upgraded) with
:  shadow passwds.

:                             Phil Jones. 

I don't know if this is in a FAQ yet or not, but it should be and hopefully
fixed in the distributions floating around.

As root:  rm /usr/log/wtmp ; ln -sf /etc/wtmp /usr/log/wtmp

Enjoy
--
Mike Horwath    IRC: Drechsau   BBS: Drechsau   LIFE: lover
root@jacobs.mn.org  drechsau@jacobs.mn.org
Jacob's Ladder  612-588-0201  UUCP, UseNet, Linux files, BBS

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

From: ecco@ipx2.rz.uni-mannheim.de (Martin Sckopke)
Subject: Windows NT and Lilo
Date: 6 Nov 1993 21:39:31 GMT

Hi all,

does anybody use Lilo and Windows NT on one machine ? I installed NT a
few days ago and it installs a new bootselector. What happens if I
overwrite this selector with Lilo ? Is it possible to boot NT from Lilo ?
Thanx for any hints

--
* ----------------------------------------------------------------------- *
* Martin Sckopke (ecco@ixp2.rz.uni-mannheim.de or ecco@rniwh.rni.sub.org  *
* ----------------------------------------------------------------------- *

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

From: morphy@cco.caltech.edu (Crassly Commercial Quantum Mechanic)
Subject: Re: Windows NT and Lilo
Date: 6 Nov 1993 22:58:31 GMT

ecco@ipx2.rz.uni-mannheim.de (Martin Sckopke) writes:

>Hi all,

>does anybody use Lilo and Windows NT on one machine ? I installed NT a
>few days ago and it installs a new bootselector. What happens if I
>overwrite this selector with Lilo ? Is it possible to boot NT from Lilo ?
>Thanx for any hints

I use lilo with NT. I installed lilo after NT and everything has worked
just fine as far as booting is concerned. Now if I could only get SLIP
working with this SLS distribution....
Jones
-- 
Jones M Murphy Jr                                
Assistant Vice President, New Products
AIG Financial Products
100 Nyala Farm, Westport, CT06880                      (800) 248-SWAP

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

From: jmorriso@rflab.ee.ubc.ca (John Paul Morrison)
Subject: what should hostname report?
Date: 7 Nov 1993 00:43:03 GMT

the newer net-2 hostname commands has a flag -f which sets the
full hostname like this:
hostname -f foo

hostname
foo.bar.com

depending on what your domain is. I notice SunOS hostname always shows
the domain. The Linux makefile assumes no domain name, uname will show
foo.bar.com.bar.com if you build a new kernel. A few other programs
get a little confused by it.

So what's the correct way? (I'm 100% certain myself that hostname should
report the fully qualified domain name, since that IS your host's name.
But I'm just asking here to shame people into changing.)

-- 
___________________________________________________________________________
 John Paul Morrison                         The Liberal Party of Canada:
 University of British Columbia, Canada     proof that those who do not
 Electrical Engineering                     remember the past are condemned
 jmorriso@rflab.ee.ubc.ca        VE7JPM     to repeat it

 Day 12 of Canada held hostage: the socialist Liberal tax & spend tyranny

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

From: sweh.womble@spuddy.UUCP (Stephen Harris)
Subject: mv feature?
Date: Sat, 06 Nov 93 14:01:11 GMT

In article <RFRANKEL.93Nov4113826@obelix.obelix.us.oracle.com> rfrankel@us.oracle.com writes:

> Anyway, no, I NEVER use dos. from a Unix SYSV standard man page:
> 
>       Only mv allows file1 to be a directory. In which case, the
>       directory rename occurs only if the two directories have the
>       same parent; file1 is renamed target.
> 
> This is the standard behavior i have expected in ~15 years of using
> (mostly att based) unix systems.

Hee hee, good fun.

Actually there are two differing standards for the "mv" command.
The old SYSV style system (I have used a r2 and r3 system) didn't allow
mv to move directories.  Instead there was a seperate command called mvdir
(or similar).  One system I knew had a "mv" command that called "mvdir" if
you gave it directories.  Hmm.

BSD systems allowed the renaming of directories with rename(), so the "mv"
command was allowed to move directories - as long as they stayed on the
same filesystem.  Cross file-system moving was not allowed because the
rename() call does not work in this case, and it would require a recursive
copy.

Linux (GNU actually) mv follows the BSD semantics.

[ QUERY: did SYSV have a rename() call, or was mv(1) written in terms of
  link() unlink() ?  This would explain the behaviour.... ]

                            Stephen Harris
       sweh.womble@spuddy.uucp     ...!uknet!axion!spuddy!sweh.womble

*  Meow! Call Spuddy the Cat for Usenet access in the UK.  Call 0203 364436 *

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

From: metcalf@CATFISH.LCS.MIT.EDU (Chris Metcalf)
Crossposted-To: comp.os.linux
Subject: syslogd that handles /proc/kmsg ?
Date: 7 Nov 1993 04:45:32 GMT

I'm curious to know what the canonical solution for handling /proc/kmsg is.
Presumably, the right thing is to run a version of syslogd that reads from
kmsg, adds some arbitrary priority, and logs the result appropriately.
Are there any such out there which have already been hacked like this?

Currently I run a hacked-up xconsole-style program by Craig Fields
from MIT Athena, which monitors both /dev/console via TIOCCONS and
/proc/kmsg.  But this isn't a complete solution in that I can't use
syslogk or whatever to copy kmsg output to a log file at the same time,
since only one process gets to see kmsg output.

I'll summarize email, or people can followup in comp.os.linux.admin.

Thanks,
-- 
                        Chris Metcalf, MIT Laboratory for Computer Science
                        metcalf@cag.lcs.mit.edu   //   +1 (617) 253-7766

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

From: rbauer@ecst.csuchico.edu (Robert Bauer)
Subject: Re: Finger and Last problems
Date: 7 Nov 1993 07:31:19 GMT

In article <1993Nov6.105548.78@plisgyn.demon.co.uk>,
root <root@plisgyn.demon.co.uk> wrote:
> When using finger on its own it shows logged in users o/k but if you finger
> a user (who is not logged in) it says they have never logged in. Last gives
> times for reboot ( these are highly suspect ) but every user has crash in
> place of its second time. I believe that the problem is that the programs

I had this problem on my system until I installed the three utility
distributions (one each for /etc, /bin/ and /usr/bin) recently announced.

These packages seemed to fix earlier problems with 'last' and 'w',
however they created at least one more; specifically, programs 
which attempt to get the user's login name with 'getlogin' (or
some similar function) don't work properly.  Getlogin returns
'LOGIN' as the user name...I suspect this is due to a conflicting
format for utmp.  Can anyone confirm this?  Are there any fixes
yet?

Regards,

Robert
rbauer@ecst.csuchico.edu



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

From: rbauer@ecst.csuchico.edu (Robert Bauer)
Subject: Re: Finger and Last problems
Date: 7 Nov 1993 07:39:37 GMT

>In article <1993Nov6.105548.78@plisgyn.demon.co.uk>,
>root <root@plisgyn.demon.co.uk> wrote:
>> When using finger on its own it shows logged in users o/k but if you finger
>> a user (who is not logged in) it says they have never logged in. Last gives
>> times for reboot ( these are highly suspect ) but every user has crash in
>> place of its second time. I believe that the problem is that the programs
>

I might add that there is a newer version of finger floating around on
sunsite or tsx-11.  It's called finger-523* (something).  This is a 
rewritten version of the original non-working finger which was 
distributed with linux...the author has fixed the last login time
problem, and has fixed (or added) the support for mailbox status, 
and special passwd fields, such as office phone no, etc.

Note that this version works on my system; however I had installed
an upgraded utilities package which may be using different utmp
file format.  Your mileage may vary.

Regards, 

Robert
rbauer@ecst.csuchico.edu


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


** FOR YOUR REFERENCE **

The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:

    Internet: Linux-Admin-Request@NEWS-DIGESTS.MIT.EDU

You can send mail to the entire list (and comp.os.linux.admin) via:

    Internet: Linux-Admin@NEWS-DIGESTS.MIT.EDU

Linux may be obtained via one of these FTP sites:
    nic.funet.fi				pub/OS/Linux
    tsx-11.mit.edu				pub/linux
    sunsite.unc.edu				pub/Linux

End of Linux-Admin Digest
******************************
