Re: /etc/passwd migration between systems

[Bill Wohler]

Wayne Schlitt <[EMAIL PROTECTED]> writes:
> > As Jim Worthington said:
> > > 
> > > I do know that when I tried to copy my /etc/passwd file from a Sun
> > > system to Debian 1.1, the passwords would not decrypt correctly on
> > > login.
> 
> 
> Just another data point  I just copied my passwords from my SVR4
> Dell Unix system to Linux, and they worked fine.

  And another data point (for Crack).  My old binary of Crack works on
  Debian 1.1, HP-UX 9.*, and SunOS password files also.  I also
  recompiled it and it still works on the SunOS file.

  Also tried logging in with an HP-UX password.  Also worked fine in 1.1.

Bill Wohler <[EMAIL PROTECTED]>   ph: +1-415-854-1857  fax: +1-415-854-3195
Say it with MIME.  Maintainer of comp.mail.mh and news.software.nn FAQs.
If you're passed on the right, you're in the wrong lane.

more on procmail

[Rick Macdonald]

There have been a lot of posts about procmail, so I thought I'd jump in
with a question too.

I've installed procmail, but haven't done anything with it yet.

I use Netscape for a mail reader at home and office. It only uses file
locking on the system mail spool (the Inbox), not any of the folders
that it uses and maintains.

I figure (and Netscape sortof warns) that using a mail filter like
procmail or mailagent that intercepts mail and writes it to Netscape's
folders would be asking for trouble.

So, I figure I need a program to use as an external movemail program for
Netscape (I've been using its internal one) that has the smarts of a
mail filter.

Can procmail or mailagent be used as a batch "movemail" program like
this?
I figure that it would be safe, since it would only be executed by
Netscape when I click the GetMail button. It's not hard to remember not
to do anything else within the Netscape mail reader while it's
processing the button that I just clicked to get new mail.

Whether or not the above is possible, is there some other way to safely
combine a mail filter program and a mail reader that doesn't use locks
on it's mail folders?

-- 
...RickM...

Re: inn installation question

[Dirk . Eddelbuettel]

  Erik>  Can't really seem to figure out what is wrong. Indeed I do not have
  Erik> a /usr/bin/mail, could it be my use of sendmail instead of smail? If
  Erik> anyone has any hints on how to solve this one, I'll be very grateful!

Install the "mailx" package, it provides the simple /usr/bin/mail that is so
handy for scripts. If the inn package you use does not depend on mailx, you
should report this as a bug.

--
Dirk Eddelb"uttel http://qed.econ.queensu.ca/~edd

Screwing up on unsubscribe instructions (Was: Re: Date: 26-Aug-1996 10:47:04 -0400)

[Barid Bel Medar]

I'd just like to make sure my apology gets through for giving incorrect 
instructions on unsubscribing from this mailing list.  Use the WWW 
interface at  instead.

Again, sorry, I wasn't thinking...

On 26 Aug 1996, R Chris Ross wrote:

> 
> unsubscribe debian-user
> 
> 

--
Barid Bel Medar   [EMAIL PROTECTED]
Knights of the CosmosShayol Ghul Resort and Health Spa
--
"I  am  returning  this otherwise good typing paper to you because
someone has printed gibberish all over it and  put  your  name  at
the top." - English Professor, Ohio University
--

May I have your chat scripts, please?

[Mike Taylor]

I am assembling a collection of chat scripts in support of an effort by
our Debain developers to generate this information during setup.  We need
your help because the proceedure for logging in and starting ppp
frames varies from ISP to ISP.  With a reasonable database, most of the
variations can be anticipated.

If you would like to help, please send me the following (assuming you
use ppp and chat):

- Name and location of your ISP
- contents of your /etc/ppp/options file (if empty, so state)
- SANITIZED contents of your chat script
- command line that you use to fire up ppp
- If pppd is invoked from a shell script, I need that script
- Any other files/information/proceedures needed to bring up ppp on your
machine
- If you use a secrets file, say what kind.  I don't need the file.

Please sanitize this information before you send it to me.  I suggest
replacing your real username, password, and ISP phone number with
myusername, mypassword, and 555-1212.  Above all please do not send me
any information that could cause problems for you if it fell into
the wrong hands.  To be of use, I will have to put your information in
a place where others can get to it.  I will not intentionally deliver it
into the hands of crackers, but I make no guarentees.

If your scripts work, I want them.  Even if they are not pretty,
even if they fail sometimes.  If your scripts fail sometimes, tell me so.
If you do anything that you think might be out of the ordinary, please
tell me about that.

If you don't use chat, please describe your proceedure to bring up ppp
as precisely as possible and include the contents of supporting files
(example, a Seyon script).  Remember to sanitize.

If you use slip or some variant of slip supported by Debain, I'd like your
information also.  I don't use slip so I can't as specific as my request
for ppp - chat info, but I think you have the idea.

Please send your submissions to me by private e-mail.  Thanks in
advance for your help.

Mike Taylor

Irqtune: some stats (was Fix for your serial/PPP problems)

[Philippe Troin]

I've done some statistics regarding irqtune and kernels 2.0.14 and 
2.0.15.

The protocol was the following:
1) ftp to my ISP
2) put a file
3) get the same file 10 times with a 'mget file file '

I timed only the 'GET', not the 'PUT' as Linux buffers outgoing 
packets and tricks the figures.

Test I was done on a 2.0.14 kernel with a 100kb nul file (dd 
if=/dev/zero...)
Test II was done on a 2.0.14 kernel with irqtune and a 100kb nul file.
Test III was done on a 2.0.15 kernel with a 100kb nul file.
Test IV was done on a 2.0.15 kernel with irqtune and a 100kb nul file.
Test V was done on a 2.0.15 kernel with a 100kb random file.
Test VI was done on a 2.0.15 kernel with irqtune and a 100kb random 
file.

The same phone line was used for tests I and II, III and IV, and V 
and VI.
That is I didn't disconnect between I and II, III and IV and V and VI.
However I rebooted and disconnected between II and III, IV and V.

Here are the timings as reported by ftp, as well as the transfer 
rates as reported by ftp.
Then come the average figures for timings and transfer rates. The 
average for transfer rates were computed from the average time 
(100kb/).
Finnaly, I computed standard deviance for timings (not for transfer 
rate, they should be the same as timings if I remember correctly my 
stats (arghhh) lessons).

  I  IIIII IV V  VI
Count Time Rate  Time Rate  Time Rate  Time Rate  Time Rate  Time Rate
--
 156.4  1.8  27.2  3.7  59.2  1.7  27.3  3.7  67.2  1.5  26.9  3.7
 268.2  1.5  27.2  3.7  77.0  1.3  36.5  2.7  44.9  2.2  26.9  3.7
 340.7  2.5  41.0  2.4  58.4  1.7  27.1  3.7  80.0  1.2  26.9  3.7
 456.7  1.8  37.0  2.7  55.5  1.8  27.0  3.7  47.6  2.1  27.0  3.7
 563.0  1.6  36.6  2.7  44.8  2.2  41.1  2.4  68.1  1.5  27.0  3.7
 665.9  1.5  36.5  2.7  58.0  1.7  41.1  2.4  45.6  2.2  26.9  3.7
 768.9  1.5  41.2  2.4  46.2  2.2  41.0  2.4  62.1  1.6  26.9  3.7
 855.9  1.8  41.1  2.4  55.4  1.8  41.2  2.4  46.4  2.2  26.9  3.7
 973.2  1.4  42.2  2.4  70.0  1.4  36.5  2.7  60.4  1.7  27.0  3.7
1042.8  2.3  36.6  2.7  60.8  1.6  27.2  3.7  48.6  2.1  26.9  3.7
--
Avg.  59.17 1.7  36.66 2.7  58.53 1.7  34.60 2.9  57.09 1.7  26.93 3.7
Dev.  10.32   5.18   9.15   6.31  11.58   0.05

The figures are quite clear. Irqtune really improves my serial 
performances, even with kernel 2.0.15 (see previous messages, 2.0.15 
included a patch which might have helped serial performance).
What I don't understand is why a file consisting of zeros only do not 
get the same transfer rate as a random file. I suspect an obscure PPP
problem.

My machine is an AMD DX2-66 with 32MB RAM, 32MB swap, and I have fast 
serial controllers (16550A). The modem is an external 33.6 V.34/V.FC 
Cardinal connected to my ISP which has 33.6 too.

I'm not subscribed to the linux kernel mailing list, but you can 
forward this message over there.

I stick with irqtune right now. BTW, I saw no ill side effect when 
using it.

Phil.

Re: Squid 1.0.9 available at ftp.fuller.edu

[Miquel van Smoorenburg]

In article <[EMAIL PROTECTED]>,
Christoph Lameter  <[EMAIL PROTECTED]> wrote:
>?? Why double the work. This was an ad for the debianized version.

Oh I am sorry. I misunderstood. I've been keeping a debianized version of squid
since squid 1.0.3 at least, and I've been emailing with the original maintainer
(Craig Sanders) about it. It has been my intent for quite some time to make
the Interim release but I didn't hear anything anymore from Craig.

Oh well.

Mike.
-- 
+   Miquel van Smoorenburg   + Cistron Internet Services +  Living is a |
|  [EMAIL PROTECTED] (SP6)  | Independent Dutch ISP |   horizontal |
+ [EMAIL PROTECTED] + http://www.cistron.nl/+  fall+

Re: /etc/passwd migration between systems

[Bill Wohler]

Jim,

  My binary of Crack v4.1f that I compiled a couple of years ago (with
  ufc-crypt I believe) works fine under Debian 1.1.  I haven't tried
  to recompile it though.  I'd be happy to ship the binary to you in a
  pinch if you trust me (MH viamail or ftp preferred).

Bill Wohler <[EMAIL PROTECTED]>   ph: +1-415-854-1857  fax: +1-415-854-3195
Say it with MIME.  Maintainer of comp.mail.mh and news.software.nn FAQs.
If you're passed on the right, you're in the wrong lane.

Re: anybody seen this?

[Carlo U. Segre]

On Mon, 26 Aug 1996, Brian K Servis wrote:

> Just thought I would start here to see if I can find a solution.  I
> have a Micron PCI Pentium-90 PNP system with the EIDE controllers on
> the mother board.  On the primary controller I have two 1.2G Western
> Digital HD's.  The master is partitioned into 2 drives all for
> Win95. The slave is partitioned into 2 drives for Debian, swap and
> everything else(it is a home machine).  I have a EIDE CD-ROM on the
> secondary controller.  Here is the situation/problem.  If I am in
> Debian and then warm boot into Win95, Win95 forgets about my CD-ROM
> drive.  I have to cold boot the machine to get it back.  I have it set
> as G: just in case Win95 is trying to assign the linux partitions, it
> knows the second drive is there.  I am using  LILO to boot manage and
> have Win95 as the default(other non-linux family members use it).
> 

Your problem might be with the way Windows95 is set up.  I had a problem 
with Windows95 and a Hitachi CDROM on the second controller.  Windows95 
would not recognize it unless I changed the driver to the standard IDE 
controller as opposed to the bus-mastering controller.  The hard disks 
work fine with the bus-mastering controller on the primary interface.  
The fix is to go into the registry editor and work your way to:

MyComputer\HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\Class\hdc

This will contain 4 entries

Entry  is always set to the "Standard IDE/ESDI hard Disk Controller"

Entry 0001 describes the overall dual port IDE controller and is probably 
set to something like "Intel 82371FB Bus Master IDE Controllers"

Entry 0002 is the setting for the primary interface and is probably set to 
the following:
Devloader:  *IOS
DriverDesc: "Primary Bus Master IDE Controller"
InfPath:IDEATAPI.INF
InfSection: BMIDE_Child
PortDriver: ideatapi.mpd

Entry 0003 will probably be set up exactly as 0002 but I have changed it 
as follows to accomodate the CDROM problem:
DevLoader:  *IOS
DriverDesc: "Standard IDE/ESDI Hard Disk Controller"
InfPath:IDEATAPI.INF
InfSection: BMIDE_Child
PortDriver: ESDI_506.pdr

This fix has solved my problem and it might be that you are seeing the 
same kind of effect when you warm boot, since Linux handles the CDROM 
with no problems at all times.  The only difference is that on a cold 
boot to Windows95, the CDROM was never visible before the change.

Cheers,

Carlo



***
*Carlo U. Segre   *
*  Department of Biological, Chemical and Physical Sciences   *
*Illinois Institute of Technology, Chicago, IL 60616  *
*   Voice: (312) 567-3498  FAX: (312) 567-3494*
*  [EMAIL PROTECTED]  *
***

Making filesystems at installation time

[Bill Wohler]

Bruce Perens <[EMAIL PROTECTED]> writes:
>  In my 1.1 installation notes, I just have the users make one filesystem
> for simplicity's sake, and the resulting system does not seem to be any less
> robust, going by the number of complaints. Of course the user can get fancy
> once they know what they are doing.

  Funny how the mind works.  This made me think of the following
  suggestion for the installation which will help those who like to
  make lots of partitions.

  Currently, for me, one of the longest parts of the base installation
  is creating the filesystems.  One filesystem at a time you have to:
  a) choose to create a filesystem; b) choose a particular filesystem;
  c) elect to check for bad blocks and then after the check is done;
  d) choose a mount point.  I only have four additional Linux
  filesystems other than /, but this gets old real fast, especially
  when battling with NT and bleeding edge hardware and doing this over
  and over and over

  Ideally, it would be nice to have a form listing all the linux
  partitions.  Here, the mount point for each can be chosen.
  Prompting for the check for bad blocks would only happen once, and
  the filesystems would be created and mounted at one time so one
  could do other things which this is going on.

  If such a form isn't possible within the current constraints, a good
  compromise would be to allow the user to create all the filesystems
  *before popping back to the main menu*.  First the user should be
  asked whether to check for bad blocks for all filesystems created,
  then he should enter the mount points in turn, and then finally, all
  the filesystems should be created at once.

  The idea with all this installation stuff is to avoid having user
  actions interspersed with automatic actions that take a while.  This
  also goes for dpkg -GROEB --configure--all the packages that require
  user intervention for setup should be postponed until last.

  Thanks.

Bill Wohler <[EMAIL PROTECTED]>   ph: +1-415-854-1857  fax: +1-415-854-3195
Say it with MIME.  Maintainer of comp.mail.mh and news.software.nn FAQs.
If you're passed on the right, you're in the wrong lane.

Re: Question about dependencies ...

[Guy Maor]

On Mon, 26 Aug 1996, Richard G. Roberto wrote:

> package and I got a dependency error.  Mosaic needs libpng
> and zlib1.  I have libgr installed.  Although libgr does not
> conflict with these two other packages, they conflict with
> it.

libgr is an obsolete package.  Its functionality is replaced by
libtiff, zlib1, libpng, and a few others.


Guy

Q's about modifying a debian package install

[Al Youngwerth]

I'm still trying to get a grip on the whole theory of debian package 
installations and I have a few questions about how one should go about 
modifying an installation.

I just installed the apache package. The debian installation sets up apache to 
run out of the rc.d stuff but I want it to run under inetd (it's for "intranet" 
access and security is more important than speed). What's the "right" way to 
make this modification? To take apache out of the start-up process should I:

Change the apache script in /etc/init.d to just exit whenever its called?

Run "update-rc.d apache remove" (this generates the error: "update-rc.d: error: 
/etc/init.d/apache exists during rc.d purge)?

Just delete apache from /etc/init.d and its symlinks in all the /etc/rc.d 
directories?

I guess the right way to fix this problem would be to change the apache .deb 
package installation script stuff so it allowed for either type of installation 
and then future upgrades (or purges) would be done correctly?

What about installing non-debianized programs into my system. I've seen people 
refer to installing all non-debian stuff in /usr/local. Why? I'm writing a lot 
of little perl script utilities for my system. Should they all go in 
/usr/local? What about the configuration files for the perl script utilities I 
write. I've been putting them in /etc, should they go someplace?

Thanks in advance for answers to any of my questions.

Al Youngwerth
[EMAIL PROTECTED]

BTW is someone working on upgrading the debian apache package to 1.1.1?

serial connection

[Miro Torrielli]

Ooops, I meant a null-modem connection.

popclient

[Miro Torrielli]

Does anyone know how to use procmail with popclient?
When configured to use procmail, popclient exits saying that it
can't exec a few things. Any hints? :-)

Oh, and does anyone know how to establish a serial connection between
DOS and a debian box???

Thanx for the help...

Re: gcc can't find termcap library

[Bruce Perens]

Bruce:
> There is no varargs.h in libc5-dev_5.2.18-9 . Maybe older versions
> contain it. 

David:
> varargs.h is provided by gcc.

Oops. Never thought of that. Anyway, nothing should use it any longer.

Thanks

Bruce

Re: perl

[Andy Dougherty]

On 23 Aug 1996, Ed Donovan wrote:

> Andy Dougherty <[EMAIL PROTECTED]> writes:
> 
> > Perl uses 'csh' for filename globbing (that is, for <*.*> sort of
> > constructs.)  Yes, that's strange behavior, and the dependency probably
> > will eventually be removed, but not anytime soon.
> 
> But, but...I thought it was noted, here or in the bug reports, that the
> csh-usage was an optional include, and was being taken out.  Maybe this
> is a Debian stable/unstable issue?, because my current perl, and the
> current rex/binary-i386 Packages file say:
> 
> Package: perl
> Version: 5.003-2
> Pre-Depends: libc5, libgdbm1, libdb1  

It is possible to configure perl such that it doesn't require csh.
Running 

/usr/bin/perl -V:d_csh

shows that Darren has, in fact, done just that (for perl5.003-02),
so that for perl5.003-02, you don't need csh.  It may well be the case
that it was needed for previous versions such as the original user
was wondering about.

My remarks above (about removing the use of csh from Perl) were actually
referring to the upstream perl sources, though I see that's certainly not
obvious.   Sorry for the confusion.

Andy Dougherty  [EMAIL PROTECTED]
Dept. of Physics
Lafayette College, Easton PA 18042

Re: gcc can't find termcap library

[David Engel]

Bruce Perens writes:
> From: "Larry 'Daffy' Daffner" <[EMAIL PROTECTED]>
> > This is not necessarily always possible. The one case I know about is
> > when a program uses "varargs" instead of the newere, ANSI-standard
> > "stdarg".
> 
> There is no varargs.h in libc5-dev_5.2.18-9 . Maybe older versions
> contain it. 

varargs.h is provided by gcc.

David
-- 
David EngelOptical Data Systems, Inc.
[EMAIL PROTECTED]  1101 E. Arapaho Road
(214) 234-6400 Richardson, TX  75081

inn installation question

[Erik van der Meulen]

I seem to have managed to set up inn to process my uucp newsfeed (never been
able to get that far with Slackware). However, I keep receiving regular
(daily) mail messages:

 >Date: Mon, 26 Aug 1996 03:09:28 +0200
 >To: news
 >Subject: Cron <[EMAIL PROTECTED]> news.daily expireover
 >X-Cron-Env: 
 >X-Cron-Env: 
 >X-Cron-Env: 
 >X-Cron-Env: 
 >
 >/usr/lib/news/bin/news.daily: /usr/bin/mail: No such file or directory

Can't really seem to figure out what is wrong. Indeed I do not have a
/usr/bin/mail, could it be my use of sendmail instead of smail? If anyone
has any hints on how to solve this one, I'll be very grateful!

Regards,

Erik van der Meulen
[EMAIL PROTECTED]

Date: 26-Aug-1996 10:47:04 -0400

[R Chris Ross]

unsubscribe debian-user

Re: gpm_1.06-3.deb: libgpm.so.* are missing

[Shaya Potter]

On 26 Aug 1996, Martin Dinkloh wrote:

> In gpm_1.06-3.deb the files libgpm.so.* are missing.
> 
> I noticed this since mc wouldn't run anymore after updating to this package.
> I have not checked if there are any other files missing.
> 
> Martin Dinkloh
> 

You need to install the libgpm1 package to get those libraries.

Hope this helps,

Shaya
--
Shaya Potter
[EMAIL PROTECTED] 

Re: a4.sty and a4wide.sty

[Susan G. Kleinmann]

You can find a particular .sty file (or anything else in the CTAN archives)
by visiting:
  http://www.shsu.edu/cgi-bin/ctan-index

Then enter the name of the macro you want, in this case, a4.sty, and you'll
be returned a page with links to sites that provide it.

HTH,
Susan Kleinmann

Bug#4291: dselect and ftp

[Ian Jackson]

Package: base
Version: ?

Kevin Buhr writes ("Re: dselect and ftp"):
...
> You did nothing wrong: when I installed Debian 1.1.5, I noticed that
> the "dpkg-ftp" had been *automatically* deselected, and I had to
> manually fix that.  I just forgot to file a bug report about it.

Ahh, the light dawns.  dpkg-ftp had been marked as installed in the
base system, but not as selected for installation.  dselect therefore
went for the default, which was to deinstall it because its priority
was too low.

This was three bugs:
 (a) The base system should have marked it as selected for
 installation.
 (b) When something is installed but with a wanted status of `unknown'
 dselect should not want to deinstall it.  I'll fix this in dpkg
 1.3.10.
 (c) dpkg-ftp should have been in the main distribution.  This is now
 fixed in unstable, I believe.

I'm reporting this to the bug system because of (a).

Ian.

Re: DPKG Remapping table

[Ian Jackson]

Randall Shutt writes ("DPKG Remapping table"):
>  Does anyone know where the dpkg remapping table is?  I need to remap
> pentium to i386 due to my recent installation of a pgcc compiler.  Trouble is,
> somehow, dpkg --print-arrcheticture returns pentium.  Tends to hose lots of
> install stuff and cause nice errors with dpkg -i.  Anyway, thanks alot.

The table is `archtable' in the source package, but it's compiled in
so you'd have to rebuild the package.

dpkg 1.3.10 will have `pentium' as an alias for `i386', and it will be
available shortly.

Ian.

Re: flop-py disk installation...........

[Susan G. Kleinmann]

Hi Mike and Candy --

You asked:
> Can anyone clue me in as to how to transfer package 
> information to a floppy in a form that linux can recognize?tried:
> direct copy from mswindows file manager,
Does this mean you had an MS-DOS formatted diskette in your floppy drive?
i.e., After you made the copy from Windows, could you successfully execute
a DOS command like DIR A: (or B:, whichever is appropriate), and see listed
the files you thought you'd copied?

When you booted Linux, did you then make the drive accessible by executing
mount -t msdos /dev/fd0 /floppy   OR  mount -t msdos /dev/fd1 /floppy
(whichever ...), before executing dselect?

> rawrite but no go also, 
Rawrite doesn't write a file _system_ onto a disk.  It just writes bytes
starting at a specified sector.  In order for a disk written by rawrite 
to be made accessible to other commands (e.g., dselect, or just plain 'ls') 
which are interfaced to the Linux file system, you must first copy the 
appropriate sectors of the diskette into a file.  
For example, to copy the whole diskette into a file in the /tmp 
directory, you'd use (for drive A:):
dd if=/dev/fd0 of=/tmp/my_file_name

Then you could tell dselect that your files were in a mounted file system,
in directory /tmp.

> booted from custom disk got message that dselectlockfile access denied,
> as well as the tragic news that the child died.

If you'd like to write and read files to a floppy that's got the 
Linux native file system on it, you must "format" the floppy using:

mke2fs /dev/fd0

Then mount the floppy using
mount -t ext2 /dev/fd0 /floppy

Of course, you could mount the floppy under some other directory if you want.
I'm just using the name "/floppy" because it's supplied with the base system,
and that's what it's there for.  The only advantage in using an ext2-formatted
floppy, as opposed to an MS-DOS formatted floppy is that you can use long
filenames with an ext2 formatted floppy.  The disadvantage is that you can't
read the floppies with DOS.

Be sure to unmount the floppy before you remove it from the drive.  
umount /dev/fd0


I'm not sure if this solves the particular problem(s) you were having.  If
not, send me private email, and I'll try to help you out.

Good luck,
Susan Kleinmann

Re: "man" crashes

[Ian Jackson]

Christoph Lameter writes ("Re: "man" crashes"):
...
> Here is the output from what you suggested:
> mandb: warning: /usr/man/man1/dpkg-source.1 is a dangling symlink
> mandb: can't open /usr/man/man1/dpkg-parsechangelog.1: No such file or
> directory
> mandb: warning: /usr/man/man1/dpkg-parsechangelog.1: bad symlink or ROFF
> `.so' request
...
> Seems that the man package cannot cope with some screwups of the dpkg
> updates.

Oops, I forgot to change the symlink target.  I'll fix this.

Ian.

Re: Bugs

[Ian Jackson]

Kevin M. Bealer writes ("Bugs"):
> I've submitted these two bugs to debian bugs, and got no
> response;  I submitted them again, and still got no response. 
> (Has the list address changed recently?)  I could try again, but
> then I would have to wait again, and so on, so...  (In
> otherwords, sorry for the offtopic.)

Kevin: please send them to [EMAIL PROTECTED], if you haven't
already.

Bruce: this is very bad.  Could you change the Pixar alias to point to
[EMAIL PROTECTED] ?  I've made this an alias for
[EMAIL PROTECTED]  We know that Pixar can deliver mail here.

It is definitely very bad if messages are just disappearing and not
even bouncing.  If all else fails the alias at Pixar should be
removed.

Thanks,
Ian.

Re: gpm_1.06-3.deb: libgpm.so.* are missing

[Martin Dinkloh]

Michael Meskes ([EMAIL PROTECTED]) wrote:
[...]
: It's not missing it just got deleted due to a bug in the postinst. You
: should be able to work around it by simply installing it once again.
[...]

I installed it several times, however /usr/lib/libgpm.so.* was not installed.

Another thing I noticed was that dpkg is not able to kill gpm itself, I must
do that by hand. Otherwise dpkg gets stuck during the installation.
That problem occured with gpm_1.06-1.deb already.

Martin

Re: a4.sty and a4wide.sty

[Dirk . Eddelbuettel]

  >> Hakan Ardo writes: > Sure it does, the only trouble with that one is
  >> that it only uses half of > the paper :) The rest is margins...
  >> 
  >> That's true. And I don't know how to solve it correctly. I just entered
  >> a latex preamble:

I always use "vmargin.sty" which allows to define the margins sizes
explicitly, and also allows me to switch from letter to a4 paper with one
command, eg
\usepackage{vmargin}
  \setpapersize{A4}  
  \setmargrb{3cm}{3cm}{3cm}{3cm}
selects a4 paper (default is letter), and then sets left, top, right and
bottom margins to 3cm. Vmargin is on CTAN hosts as 
tex-archive/macros/latex/contrib/other/misc/vmargin.sty

--
Dirk Eddelb"uttel http://qed.econ.queensu.ca/~edd

Re: Non-existent .deb's

[Ian Jackson]

Daniel Lynes writes ("Re: Non-existent .deb's"):
> On Thu, 22 Aug 96 00:28 BST, Ian Jackson wrote:
> >If you delete the `Packages' files, or fail to download them, dselect
> >will offer to scan the .deb files that are actually on your disk.
> 
> But, this method does not use the descriptions, or dependency lists,
> either.  I kind of like that information, as it allows me to see if I'm
> trying to install something I don't have the extra files for.

This method _does_ use the descriptions and dependency information
from the packages, just like using the Packages file does.

The difference is that some of the classification
(text/electronics/... &c and standard/important/... &c) will be wrong.

Ian.

anybody seen this?

[Brian K Servis]

Just thought I would start here to see if I can find a solution.  I
have a Micron PCI Pentium-90 PNP system with the EIDE controllers on
the mother board.  On the primary controller I have two 1.2G Western
Digital HD's.  The master is partitioned into 2 drives all for
Win95. The slave is partitioned into 2 drives for Debian, swap and
everything else(it is a home machine).  I have a EIDE CD-ROM on the
secondary controller.  Here is the situation/problem.  If I am in
Debian and then warm boot into Win95, Win95 forgets about my CD-ROM
drive.  I have to cold boot the machine to get it back.  I have it set
as G: just in case Win95 is trying to assign the linux partitions, it
knows the second drive is there.  I am using  LILO to boot manage and
have Win95 as the default(other non-linux family members use it).

I think this is kind of a catch 22 situation.  The Linux folks will
tell me to look at the Win95 side of it and the Win95 folks will tell
me to look at the Linux side of it.

Any pointers?

Thanks

Brian Servis
Mechanical Engineering
Purdue University
[EMAIL PROTECTED]
http://www.ecn.purdue.edu/~servis

Question about dependencies ...

[Richard G. Roberto]

Hi!

I tried dselect's ftp method last night for the first time.
It's pretty nice.  However, I tried to install the mosaic
package and I got a dependency error.  Mosaic needs libpng
and zlib1.  I have libgr installed.  Although libgr does not
conflict with these two other packages, they conflict with
it.  As libgr provides the functionality of zlib and libpng,
I guess these should conflict, but shouldn't this be mutual?
I mean shouldn't libgr conflict with zlib1 and ligpng?
Also, couldn't there be a couple of virtual packages that
these could provide (mutually exlisive) that things like
mosaic could depend on instead of depending on the actuall
packages?  I'm not sure that libgr is a fully functional
replacement for zlib1 and libpng, I'm just guessing.  But
I'd sure like to know what to do to get mosaic installed
without having to force the depencies.

Thanks in advance.

Richard G. Roberto
[EMAIL PROTECTED]
201-739-2886 - whippany, nj


--
***
Bear Stearns is not responsible for any recommendation, solicitation, offer or
agreement or any information about any transaction, customer account or account
activity contained in this communication.
***

Re: flop-py disk installation...........

[Dale Scheetz]

On Mon, 26 Aug 1996, Mike& Candy List wrote:

> I am soo confused.Can anyone clue me in as to how to transfer package 
> information to a floppy in a form that linux can recognize?tried:
> direct copy from mswindows file manager,rawrite but no go also, booted
> from custom disk got message that dselectlockfile access denied,as well
> as the tragic news that the child died.I'm so green that I expect to 
> have trouble,but if anyone can set me straight I'd be grateful.
> Mike List
> 
Well, first, get out of windows. Make sure rawrite is in the same
directory with the disk images you wish to copy (boot1440.bin, root.bin,
base14-1.bin, base14-2.bin, and base14-3.bin). Copy one image to each of
five formated disketts. To do this type rawrite at the dos prompt. You
will be prompted for an input file and an output file. The input file
should be one of the above .bin files. The output file/device should be
a:, unless your 3 1/2" drive is b:.
Once you have made these five disks, boot your computer with the boot disk
and follow the directions. I would also suggest that you print out the
installation directions (find them in the same place as the above disk
images) as they will be very helpful.
I you still have problems, don't hesitate to let us know. Some information
about the target machine (processor, memory, special interfaces...) will
give us a better idea about where your problems may lie.

Luck,
 
Dwarf

  --

aka   Dale Scheetz   Phone:   1 (904) 877-0257
  Flexible Software  Fax: NONE 
  Black Creek Critters   e-mail:  [EMAIL PROTECTED]

 If you don't see what you want, just ask --

Re: using procmail with emacs RMAIL

[Jean Orloff]

> On Sat, 24 Aug 96 17:42 EDT, Stephen Masterman <[EMAIL PROTECTED]> said:
 Stephen> Can someone point me to info for the quickest, easiest way to use
 Stephen> procmail with Emacs RMAIL to separate, say, the debian list into a
 Stephen> separate folder?  I normally have no problem putting hours into
 Stephen> something like this, but lately I'm incredibly busy and there seems
 Stephen> to have been loads of traffic on debian recently, so that I'd really
 Stephen> like to get it working as quickly as possible.

I switched to MH about a year ago, so this is just theory:

1) set up procmail
==
(with ~/.forward, etc) as explained by Dirk in a previous message

2) set up your ~/.procmailrc 

# Please check if all the paths in PATH are reachable, remove the ones that
# are not.
PATH=$HOME/bin:/usr/bin:/bin:/usr/local/bin:
MAILDIR=$HOME
LOCKFILE=$MAILDIR/.lockmail # default lock

# default action: keep in username mailbox
DEFAULT="/var/spool/mail/username"

#--
# Debian mailing lists go into ~/debian/inbox
:0:debian/.lock
* ^X-Loop:[EMAIL PROTECTED]
$MAILDIR/debian/inbox

3) from emacs,
==
do M-x rmail-input RET ~/debian/rmail RET 
to open/create the file ~/debian/rmail

4) in that buffer,
==
do M-x set-rmail-inbox-list RET inbox RET
to explain rmail where to get new messages from.

This all should work, now you only got to debug! ;-)

Amities,

Jean Orloff
+   +   +   +   +   +   +   +   ++
+Tel:(33)50.09.16.75   Fax:(33)50.09.94.95   http://lapphp0.in2p3.fr/~orloff/+
+   +   +   +   +   +   +   +   ++
In a Bucharest hotel lobby:
   The lift is being fixed for the next day.  During that time we 
   regret that you will be unbearable.
+   +   +   +   +   +   +   +   +

Re: dvips/metafont and Stylus 800

[Nils Rennebarth]

-BEGIN PGP SIGNED MESSAGE-

On 21 Aug 1996, Dirk Bernhardt wrote:
>I just installed the whole tex stuff. I now am trying to configure it
>to make the fonts for my Epson Stylus 800 printer.
>
>I've changed the dvips-file config.ps:
>
>D 360
>M epstylus
O.K.

>
>and added to Metafont's modes.mf
>
>mode_def epstylus = % Epson Stylus (360dpi)
>  mode_param (pixels_per_inch, 360);
>  mode_param (blacker, .35);
>  mode_param (fillin, 0);
>  mode_param (o_correction, .8);
>  mode_common_setup_;
>enddef;
Did you remake metafonts base file?

(as root say
  mfmakemfbase
  mfmakecmmfbase
)

>Output written on cmr12.2602gf (128 characters, 254596 bytes).
Appears to create a font with standard definitions (which have a fairly
high resolution)

- --
Coming again: Best quotes of the net. Today:  | Nils Rennebarth
Kristian Köhntopp <[EMAIL PROTECTED]| Schillerstr. 61 
>>I'd also be interested in the comparison [of Linux] | 37083 Göttingen
>>with a cisco router. I assume a factor of about ten.| ++49-551-71626
>What? faster or slower?  | http://www.nus.
Cheaper!  | de/~nils

-BEGIN PGP SIGNATURE-
Version: 2.6.2i

iQB1AwUBMiGWMFptA0IhBm0NAQGtpAL/Q8Yniud7agtX+Zt5oaeP54rpE3kcBFrW
2Y+Mte4qwSPb0jHbCROZF56Wht3yCF08yOFrk3uA3Y5fXpnxsjaa6RmbkHucWlCu
VYxOv7rPP2w/0icS/W8eRM18NBgAksyV
=7/ay
-END PGP SIGNATURE-

Contents of a package? (was deb-view.el)

[Jean Orloff]

Good job with deb-view.el! This is by all means a useful tool. For a next
version: it would be great to have an entry in crypt-build-encoding-alist so
that opening any .deb from emacs goes through your processing. After all,
crypt++ is provided with w3-el. This might annoy people that download .deb
files through ange-ftp though, unless you manage to get the writing routines
right...

On a related subject, I have the following problem: I am eagerly waiting for
the next release of USSLite, which should fully support my MAD16 82C930
soundcard. I tried downloading the tar.gz from the author, but failed compiling
my custom kernel with it. No hurry, I'll wait...

However, the question is: how do I know which kernel-source.deb contains which
USSLite version... Is there a way around downloading a 6Mb package, just to
read one filename? Although less acute, I guess this type of problems must
occur with other packages too (eg: what is the version number of chat in a
given ppp.deb package?)

Amities,

Jean Orloff
+   +   +   +   +   +   +   +   ++
+Tel:(33)50.09.16.75   Fax:(33)50.09.94.95   http://lapphp0.in2p3.fr/~orloff/+
+   +   +   +   +   +   +   +   ++
"Outside of the killings, Washington has one of the lowest crime rates in
the country."
Mayor Marion Barry, Washington, D.C.
+   +   +   +   +   +   +   +   +

Re: a4.sty and a4wide.sty

[Hakan Ardo]

> Hakan Ardo writes:
> > Sure it does, the only trouble with that one is that it only uses half of 
> > the paper :) The rest is margins...
> 
> That's true. And I don't know how to solve it correctly. I just entered a
> latex preamble:
> 
> \begin_preamble
> \setlength\oddsidemargin {0mm}
> \setlength\evensidemargin  {\oddsidemargin}
> \setlength{\textheight}{23cm}
> \setlength{\textwidth}{17cm}
> \topmargin-20mm
> \end_preamble
> 
> That works. :-)

Yes, it does, very well. But there should be a way to get hold of the sty 
files, it should prevent a lot of trouble, not only for me.

Re: a4.sty and a4wide.sty

[Hakan Ardo]

> Hakan Ardo writes:
> > Then there is the trouble with lyx. I have just started to try it out, and 
> > it
>  
> > seems like it uses the sty approch, but it might be a misconfiguration of it
> > to...
> 
> I use a4paper with LyX all the time without a problem. However, a4wide
> doesn't work.

Sure it does, the only trouble with that one is that it only uses half of 
the paper :) The rest is margins...

Re: a4.sty and a4wide.sty

[Michael Meskes]

Hakan Ardo writes:
> Then there is the trouble with lyx. I have just started to try it out, and it 
> seems like it uses the sty approch, but it might be a misconfiguration of it
> to...

I use a4paper with LyX all the time without a problem. However, a4wide
doesn't work.

Michael

-- 
Michael Meskes   |_  __  
[EMAIL PROTECTED] |   / ___// / // / / __ \___  __
[EMAIL PROTECTED]  |   \__ \/ /_  / // /_/ /_/ / _ \/ ___/ ___/
[EMAIL PROTECTED]|  ___/ / __/ /__  __/\__, /  __/ /  (__  )
Use Debian Linux!| //_/  /_/  //\___/_/  //

Re: gpm_1.06-3.deb: libgpm.so.* are missing

[Michael Meskes]

Martin Dinkloh writes:
> 
> In gpm_1.06-3.deb the files libgpm.so.* are missing.
> 
> I noticed this since mc wouldn't run anymore after updating to this package.
> I have not checked if there are any other files missing.

It's not missing it just got deleted due to a bug in the postinst. You
should be able to work around it by simply installing it once again.

Michael

-- 
Michael Meskes   |_  __  
[EMAIL PROTECTED] |   / ___// / // / / __ \___  __
[EMAIL PROTECTED]  |   \__ \/ /_  / // /_/ /_/ / _ \/ ___/ ___/
[EMAIL PROTECTED]|  ___/ / __/ /__  __/\__, /  __/ /  (__  )
Use Debian Linux!| //_/  /_/  //\___/_/  //

flop-py disk installation...........

[Mike& Candy List]

I am soo confused.Can anyone clue me in as to how to transfer package 
information to a floppy in a form that linux can recognize?tried:
direct copy from mswindows file manager,rawrite but no go also, booted
from custom disk got message that dselectlockfile access denied,as well
as the tragic news that the child died.I'm so green that I expect to 
have trouble,but if anyone can set me straight I'd be grateful.
Mike List

Re: problem with the mailing list

[Steffen Mueller]

On Sun, 25 Aug 1996, Susan G. Kleinmann wrote:

Hi Susan,

> Does anyone else get repeated messages from 
> "Manager, AlisaMail" <[EMAIL PROTECTED]>
> 
> regarding  "Undeliverable Mail"?

Yupp, reiceived 6 of them.

Greetings,

Steffen

PGP Key is available under : http://www.xlink.net/~steffen/steffen.asc
-
Steffen R.Mueller   __  ___ _   _ email: [EMAIL PROTECTED]
NTG Netzwerk und Telematic GmbH \ \/ / (_)_ __ | | __ fax  : +49 721 9652 210
Geschaeftsbereich Xlink  \  /| | | '_ \| |/ / phone: +49 721 9652 211
Vincenz-Priessnitz-Str. 3/  \| | | | | |   <  RIPE : SM25-RIPE
D-76131 Karlsruhe, Germany  /_/\_\_|_|_| |_|_|\_\ WWW.Xlink.net/~steffen

Re: procps 1.01a-1 problems

[Michael Meskes]

Hakan Ardo writes:
> 
> Hi,
> I tested to install procps 1.01a-1, but after that all I get out of ps is:
> 
> ps: can't load library 'libproc.so.1.01a'

This is a bug in the postinst. I don't know if it has been reported yet.
Will check that.

You should be able to work around it simply by installing it once again.

Michael

-- 
Michael Meskes   |_  __  
[EMAIL PROTECTED] |   / ___// / // / / __ \___  __
[EMAIL PROTECTED]  |   \__ \/ /_  / // /_/ /_/ / _ \/ ___/ ___/
[EMAIL PROTECTED]|  ___/ / __/ /__  __/\__, /  __/ /  (__  )
Use Debian Linux!| //_/  /_/  //\___/_/  //

Re: Debian on a Compaq Prosignia 4/66

[Lazaro . Salem]

Yes Dave, 
You can have Debian Linux on your compaq. This is from a previous posting I 
sent to this list:
>I find this LOADLIN scheme a very flexible one. I find the related DOS 
> applications rdev, pfdisk and ext2tools good companions to this DOS 
> booting scheme. A DOS readln or readkey utility (or BE ASK in NORTON 
>UTIL) 
> is also handy for creating extremely configurable DOS bat files to manage 
> the most complex scenarios you can imagine.
> Misc: This is the only method I found to boot linux on Compaq with PCI on 
> board as linux need to relocates the PCI services through a DOS driver 
> provided by COMPAQ. Syslinux or LILO cannot make linux see the PCI chips 
> (Linus Torvald said). 
> 
> Disclaimer: LILO is very good,... this is just an alternative. 
>
> Lazaro 
> <[EMAIL PROTECTED]>

This is what you have to try:

1) Get the Software package sp116.zip provided by Compaq (at www.compaq.org?). 
   Run sp116.exe contained in it to extract MOVEPCI.SYS and MOVEPCI.DOC. 
   Read that documentation. If it suits your machine (it works for my Deskpro
   590 and reported to work with Deskpro XL both compaq boxes with PCI on board)
   keep reading. Otherwise get more info from COMPAQ and come back later to
   point 3.
2) Now that the driver lets linux recognize the PCI on board, you need to
   replace the SYSLINUX method of for booting. A good alternative is the DOS 
   executable LOADLIN. Once you get it:  
   Make a DOS boot floppy with DEVICE=MOVEPCI.SYS as  the first line in
   CONFIG.SYS. Then you can load Linux with loadlin from DOS from config itself 
   with a SHELL=LOADLIN .. or from AUTOEXEC.BAT Or from the DOS command
   line. 
   If everything went smooth to this point, you shall see that Linux detects
   the PCI chips on board during bootup.

You may also want to make a customized Debian-type boot disk by: 
a)copying everything except syslinux in the Debian standard 1996-07-14 boot disk
(you don't need a special kernel) onto a _DOS_bootable_ disk; b)preppending 
DEVICE=MOVEPCI.SYS to the config.sys; and c) copying the LOADLIN stuff. I find 
adding some menuing DOS utilities and a small editor very handy to fix things on
the road in case I need it.

And that's all! just a pity XFree86 doesn't support the Compaq graphic cards. 
But you can try (10'minutes working demo) the X server from ftp.xinside.com to 
test X applications on the Debian linux box. Just keep track of the files  
installed as I had some problems with the scripts provided. 

If you get stucked or find this guide too brief to make ANY sense out of it :-) 
just write me privately for more details (I have written something about 
this where is it?) 

Also, have a look at  a related WEB page:
http://www-c724.uibk.ac.at/liedl.html";>Klaus Liedl,mailto:[EMAIL PROTECTED]"> [EMAIL PROTECTED]
Page last modified: Mon Oct  9 10:41:10 MET 1995

This kind soul used to mantain a page related to the deskpro XL. I find the info
very useful from fixing Linux on my Deskpro 590 with a PCI Net card on board.
cheers, 
lazaro <[EMAIL PROTECTED]>

__ Reply Separator _
Subject: Debian on a Compaq Prosignia 4/66
Author:  debian-user@lists.debian.org at cclink
Date:23.08.96 21:55


We were just given a Compaq Prosignia 4/66 with 40 MB RAM and a 1.05 GB
SCSI drive to user as a DB Server.  Our preference is to run it under
Linux.  However, it has a built-in SCSI controller.  So far we have
tried
Debian 1.1, 1.1.6, and all of the Debial 2.0.5 alternate boot disks with
no luck in detecting and working with the SCSI controller...

DOS does work on the system as did Novell 3.12.  The system arrived
without
documentation.  So far the email to Compaq asking for the SCSI
controller
equivalent has not been answered.

Any suggestions?  

TIA,

Dave

Re: "man" crashes

[Ed Donovan]

Hamish Moffatt <[EMAIL PROTECTED]> writes:

> When I try to view a manual page that's located in /usr/local/man,
> man says it's updating a local index, then gives a segmentation fault.
> Any ideas on how to fix this?

Christoph Lameter <[EMAIL PROTECTED]> writes:

> Any way to fix those problems?

This is going back a step: I waited, thinking I might be duplicating
Dale's answer--but I'd suggest one more thing.  I had this problem a
little while ago.  (I wasn't using an experimental version of dpkg,
though I didn't ever find out just what caused it.)  What worked for me
was running mandb with the -c option to completely rebuild the database:

-c, --create
  By default, mandb will try to update any previously
  created databases. If a database does not exist, it
  will create it. This option forces mandb to  delete
  previous databases and re-create them from scratch.
  This may be necessary if a database becomes corrupt
  or  if  a new database storage scheme is introduced
  in the future.   

I ran it from the command line as root, and without specifying any path,
letting it read /etc/manpath.config.  I don't recall what directory I
ran it from, but don't think it mattered.  Segfaults went away as silently as
they'd come.

I didn't get the impression you've tried this yet, and if you haven't,
do, because it sounds pretty much like my experience.  Hope it helps,
 
-- 

Ed Donovan  [EMAIL PROTECTED]

Re: "man" crashes

[Kevin M Bealer]

On Sat, 24 Aug 1996, Christoph Lameter wrote:

> On Sat, 24 Aug 1996, Dale Scheetz wrote:
> 
> dwarf>On Sat, 24 Aug 1996, Christoph Lameter wrote:
> dwarf>
> dwarf>> Seems that the man package cannot cope with some screwups of the dpkg
> dwarf>> updates.
> dwarf>> 
> dwarf>Sounds like you upgraded to an experimental version of dpkg. Which 
> version
> dwarf>do you have? I'm running 1.3.6 with none of these problems.
> 1.3.9 as of today.
> 
> Any way to fix those problems?
> 
> {}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}
> {}Snail Mail:   FTS Box 466, 135 N.Oakland Ave, Pasadena, CA 91182
> {}
> {}FISH Internet System Administrator at Fuller Theological Seminary   
> {}
> {}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}{}
> PGP Public Key  =  FB 9B 31 21 04 1E 3A 33  C7 62 2F C0 CD 81 CA B5 


I would suggest removing any of the files that it lists as
problems, then remove and reinstall the manpages package, and
possibly the actual "man" utility as well.


[EMAIL PROTECTED]/GNU__1.1___Linux__2.0.14___
"Don't worry about people stealing your ideas.  If your ideas
are any good, you'll have to ram them down people's throats."
-- Howard Aiken

gpm_1.06-3.deb: libgpm.so.* are missing

[Martin Dinkloh]

In gpm_1.06-3.deb the files libgpm.so.* are missing.

I noticed this since mc wouldn't run anymore after updating to this package.
I have not checked if there are any other files missing.

Martin Dinkloh

Re: problem with the mailing list

[Bruce Perens]

Susan:
> Does anyone else get repeated messages from 
>"Manager, AlisaMail" <[EMAIL PROTECTED]>

You get one for each posting. The version of the Microsoft Mail gateway they
are running at Lockheed simply assumes you can figure out the address that is
failing from the "To:" address in the message. I sent the manager there a
message asking who he or she thinks it could be for - at least half for
rhetorical value, as there isn't much way to tell and it makes the point
that their software is broken.

Probably the best way to tell is to get the distribution list from Anders.

Bruce

Re: Booting problems

[Bruce Perens]

From: "David C. Winters" <[EMAIL PROTECTED]>
>   I'm running into a problem with one, lone Linux box right now.
> After the BIOS bootstraps, the drive and CD-ROM and recognized, and the 3.5"
> floppy is seen, instead of saying "LILO" as a preface to LILO's booting into
> Linux, the following comes up:
> 
> xFA:

This is from Debian's master boot record. Press "2" to boot partition 2,
'F' to boot the floppy, etc. I suggest you re-install the LILO package and
say "no" to the first question (which asks if you want to use the existing
LILO configuration) and "yes" to all subsequent questions.

Either the partition is not bootable for some reason, or you have a shift key
down when the system boots. Otherwise you would not see this.

Bruce

Re: "Extended" memory?

[Mr Stuart Lamble]

In article <[EMAIL PROTECTED]> bhaskar wrote:
: I see that the Debian installation instructions contain the following.
: >Extended vs. Expanded Memory
: >If your system provides both extended and expanded memory, set it
: >so that there is as much extended and as little expanded memory
: >as possible. Linux requires extended memory and can not use
: >expanded memory.
: 
: I'm confused, since unices make no such distinctions between memory.
: This is under the BIOS settings instructions, so presumably there
: were systems where the BIOS somehow configured the memory to behave
: like extended memory?

Indeed. As an example, at home I have a 486, and two 286 class systems.
One of the 286 machines is based around the NEAT chipset, which allows
you to do all sorts of neat stuff (pun not intended, but I'll keep it
there anyway. :-) It has 4MB of RAM, 2MB of which is earmarked for EMS;
the rest is the base/extended. Changing the size of the EMS block is
trivial - just modify the CMOS settings, and reboot.

Linux switches into 386 enhanced mode, which treats memory as one big
continuous block. _However_, this only works if the memory is set up
to work as extended memory. EMS was designed to get around the 640KB
limit on 8086, 8088, and 80186-based systems. (yes, there _was_ an
80186 chip; it just wasn't widely used in the same way that the 8088,
80286, 80386, and 80486 were.) [yes, I know, it's the i486...sheesh.
I was on a roll there. :-) ]

Basically, the simple answer is that you can "see", and use, extended
memory in 386 enhanced mode, which Linux runs under. You cannot "see"
expanded memory, making it useless under Linux.

-- 
Windows is not the answer. Windows is the question. Linux is the answer.
http://sunsite.unc.edu/mdw/ for all your PC software requirements.

"Extended" memory?

[Bhaskar Manda]

I see that the Debian installation instructions contain the following.
>Extended vs. Expanded Memory
>If your system provides both extended and expanded memory, set it
>so that there is as much extended and as little expanded memory
>as possible. Linux requires extended memory and can not use
>expanded memory.

I'm confused, since unices make no such distinctions between memory.
This is under the BIOS settings instructions, so presumably there
were systems where the BIOS somehow configured the memory to behave
like extended memory?

Thanks for an explanation!
-bhaskar

Booting problems

[David C. Winters]

I'm running into a problem with one, lone Linux box right now.
After the BIOS bootstraps, the drive and CD-ROM and recognized, and the 3.5"
floppy is seen, instead of saying "LILO" as a preface to LILO's booting into
Linux, the following comes up:

xFA:

(where x == /dev/hdax according to whichever partition is enabled as the
active partition--I get "2FA:" if it tries to boot off of /dev/hda2 or
"3FA:" for /dev/hda3)

When this problem first came up, I found that I'm able to boot via
floppy into Linux--but the partition table was fine, as was LILO and the
/etc/lilo.conf file.  I did a full reinstall onto /dev/hda2 anyway, with the
same result.  The full, initial install onto /dev/hda3 was just another
attempt at it (that partition is earmarked for NT on the machines in my
cluster), but I got "3FA:" instead.  (Just booting into Linux and leaving it
at that isn't an option.  Because it's a cluster machine, I need it to boot
off of the HD so I can BIOS-lock the floppy out of the boot cycle.)
I did have this problem on a second machine, but a full reinstall
took care of it--but didn't teach me anything.  Neither machine had any more
on it than the five-floppy initial installation set puts on it.  I'm using
1.1.


David [EMAIL PROTECTED]
Office: 3503 WeH, x86720
MTFBWY

Re: procps 1.01a-1 problems

[Carlo U. Segre]

On Sun, 25 Aug 1996, Hakan Ardo wrote:

> Hi,
> I tested to install procps 1.01a-1, but after that all I get out of ps is:
> 
> ps: can't load library 'libproc.so.1.01a'
> 
> The libriry I have is libproc.so.0.99, which package contains the new version,
> and shouldn't the dependency infomation in procps have told me that?
> 

I had the same problem and it went away after reinstalling the procps 
package by hand.  I think that the library is in the package itself.

HTH

Carlo

***
*Carlo U. Segre   *
*  Department of Biological, Chemical and Physical Sciences   *
*Illinois Institute of Technology, Chicago, IL 60616  *
*   Voice: (312) 567-3498  FAX: (312) 567-3494*
*  [EMAIL PROTECTED]  *
***

Re: 1.1 Install problem

[Carlo U. Segre]

On Sun, 25 Aug 1996, Kevin Murphy wrote:

> 
> hda: read_intr: status = 0x59 {DriveReady SeekComplete DataRequest Error}
> hda: read_intr: error = 0x40 {UncorrectableError}, LBAsect = 2247884,
>  sector = 131084 
> end_request: I/O error, dev 03:04, sector 131084 Kernel
> panic: EXT2-fs panic (device 03:04): read_inode_bitmap: Cannot read inode
>bitmap_block_group = 8 inode_bitmap = 65542
> 
.
. 
.
> 
> The motherboard is very new and has a fancy AMI BIOS.  It has a built-in
> PCI IDE controller, and the BIOS setup allows you to tweak settings for the
> IDE, like LBA mode, block mode, 32-bit mode, and PIO mode.  During the
> installation, LBA mode was on, block mode was on at 8 sectors, 32-bit mode
> was on, PIO mode was auto, which meant the machine was coming up with PIO
> 4.  I don't know if this has anything to do with it.
> 

I have had this happen to me when the PIO level was set too high.  Try 
setting it manually to a lower value and work your way up the highest 
functioning mode.

Cheers

Carlo

***
*Carlo U. Segre   *
*  Department of Biological, Chemical and Physical Sciences   *
*Illinois Institute of Technology, Chicago, IL 60616  *
*   Voice: (312) 567-3498  FAX: (312) 567-3494*
*  [EMAIL PROTECTED]  *
***