Files
oldlinux-files/docs/mail-archive/linux-admin/Volume2/digest154
2024-02-19 00:23:35 -05:00

645 lines
24 KiB
Plaintext

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: Fri, 7 Oct 94 03:14:21 EDT
Subject: Linux-Admin Digest #154
Linux-Admin Digest #154, Volume #2 Fri, 7 Oct 94 03:14:21 EDT
Contents:
Re: [Q]: Linux on HP PC? (Greg Philmon)
Re: /sbin/hostname on 1.1.45- (Greck Cannon)
[ANSWER&?] Re: IDE/EIDE drive >528M?? (Kevin Burtch)
PCI-Herc video working??? (Tim Laren)
Help: nslookup works but others fail (G Cheng)
Re: PACKET SNIFFER FOR LINUX (Matthew Donadio)
dump for Linux? (Bill Braughton)
Re: New Adeptec SCSI not detected (Zeos Technical Support account for Internet)
Slip autoanswer (Peter Berger)
slack and named: Is this fixed in 2.1* (Peter Berger)
ftape + Highscreen/Vobis tape drive = ? (Marek Michalkiewicz)
Disk partitioning and moving filesystems. (Kristian Mowinckel)
Re: Mystery Chip...AMD (Rob Kean)
Re: [Q] 4M -> 8M RAM worth it? (Rob Kean)
Re: shutdown without root access (Randy Hootman)
----------------------------------------------------------------------------
From: philmon@netcom.com (Greg Philmon)
Subject: Re: [Q]: Linux on HP PC?
Date: Thu, 6 Oct 1994 13:32:52 GMT
> I stick to Linux. Will it run on a HP PC?? Together with X???
>
> Has anybody ever tried to install Linux on a HP PC?
>
I'm running Linux on an HP Vectra 486/66 XM. It works just fine.
Someone else said that they were never able to get the onboard ethernet
controller to work... I initially had this problem, but I grabbed a
patch that was posted to C.O.L.A that fixed it.
I don't recall the exact details, but it involved patching a single
file (lance.c). You can probably find it on sunsite.
The only other problem I encountered was getting the mouse to work.
It's a bus mouse and I had some weird conflicts with the mouse under
X, but only when the onboard ethernet was enabled.
But changing some CMOS settings cleared it up. It's been a while and
my memory fails me, but I believe that I ended up changing the ethernet
IRQ.
Finally, I had to change the IRQ used for the bus mouse and recompile
the kernel.
If you do get the HP, feel free to contact me via e-mail for the
specifics.
--
Greg Philmon | philmon@netcom.com | CIS: 71161,3445 | MCI: 588-5358
------------------------------
From: greck@scaredy.catt.ncsu.edu (Greck Cannon)
Subject: Re: /sbin/hostname on 1.1.45-
Date: 6 Oct 1994 15:43:45 GMT
Peter Tobias (tobias@server.et-inf.fho-emden.de) wrote in message <<370dlr$jd1@server.et-inf.fho-emden.de>>:
> Yes! The only thing you have to do is to change your /etc/hosts entries to
> something like this:
> #IP-Number Fully Qualified Domain Name Alias
> 152.1.43.22 scaredy.catt.ncsu.edu scaredy
> Functions like gethostbyname expect the FQDN at the second column (and the
> the alias names at the third column). You can now set the host name to
> either the FQDN (scaredy.catt.ncsu.edu) or one of the alias names (scaredy).
You are, of course, correct, and this is the way I have it set up now.
The problem is, w (and other commands) look much better if they return
just the first part of the name (so I'm aesthetic--sue me :). On Ultrix
you can evidently use this format:
ip short-form fqdn aliases
and it will use the short-form for stuff like w, but can still find the
machine's fqdn.
Maybe the best answer is to recompile w... <grin>
-greck
--
Greck S. Cannon \ [He's] only bitter on the outside--inside
sophomore CSC major \ he's got creamy nougat.
greck@ \ -Slappy Squirrel
scaredy.catt.ncsu.edu \
===========================
set your URL to http://www.catt.ncsu.edu
------------------------------
From: kburtch@pts.mot.com (Kevin Burtch)
Subject: [ANSWER&?] Re: IDE/EIDE drive >528M??
Reply-To: kburtch@pts.mot.com
Date: Wed, 5 Oct 1994 17:42:10 GMT
I have a setup (still installing) that works nicely with big drives...
/dev/hda has 1322 cylinders, 16 heads, and 63 sectors.
/dev/hdb has 2100 cylinders, 16 heads, and 63 sectors.
I booted with an older boot/root floppy set (kernel 1.0.0, slackware...)
using the lilo line (recommended for PS/1 users in the preboot message)
boot: ramdisk hd=1322,16,63
and it loaded up, recognizing my first hard drive (that already had one
240M MS-DOS partition).
Next I ran fdisk (linux) to create a second MS-DOS (*)primary partition
of 240M. * - MS-DOS FDISK.EXE will only create one primary partition, the
rest have to be inside an extended partition.
I then rebooted into MS-DOS to format the new drive. (to verify it worked)
So far, so good... everything fuctions perfectly. Reboot into linux, using
the same command as before...
I then used linux fdisk to create a ~24M swap partition and ~145M linux
native partition... Reboot again, mkswap & mke2fs etc. perform minor install
and reboot into MS-DOS to verify everything still works... no problems! :)
Next, using loadlin, boot linux (loadlin 1.5 is from heaven!!) with the
appropriate root=/dev/hda4 line... perfect! Run fdisk, and guess what?!
Linux (1.0) not only knew the *correct* size of hda, it also knew hdb!
(I thought I would have to have 1.1.50+ to automatically recognize them! :)
So far, with this setup, I have _no_ device drivers loaded, and 480M set up
for MS-DOS, and Linux filling the rest of the space on the disk with no
problems! I could have up to the MS-DOS limit (504M) using this method...
Here are the oddities...
After creation of the last partitions, if I load FDISK.EXE my partition
table lists the linux native (cyl 1025+) partition first, then the two
FAT partitions, then the swap. Apparently MS-DOS strips 1024 from the
number in the partition table, and then thinks the partition is at the
beginning of the disk. It seems to function properly, and since MS-DOS
cannot access anything over 1024, my linux partition is safe. Since linux
is intelligent enough (what did you expect? :) to recognize cylinders above
1024, my MS-DOS partitions are safe. (except from the virus MS-Windows ;)
When creating a partition in the _second_ drive with FDISK.EXE, the
second partition in the _first_ drive was no longer recognizable by
FDISK.EXE or MS-DOS. :( Well, I tried checking the partition table with
linux fdisk, it was OK. hmmm... Next step was to delete the MS-DOS
partition from the second drive: POOF! like magic my second partition in
the first drive is now an MS-DOS partition with a readable volume name again
(FDISK.EXE) and everything is working again. I tried making an extended
partition on hdb to host the MS-DOS partition. (thinking MS-DOS can only
have two total primaries) No luck, same problem. I deleted the MS-DOS
partition from the extended partition (left the empty extended partition)
and the problem still wouldn't go away. I deleted the extended partition
and POOF! (again) it worked. (linux worked in all instances! :)
Does anyone know enough about the partition tables to know why the creation
of a partition on hdb can effect the partitions on hda? Are they both stored
on hda? (I thought each drive had it's own partition table)
You're welcome for the install advice, and
thank you for any help any of you might offer me...
I plan to work on it some more tonight when I get home, so
I will try to keep you updated as to my 1.7G install. :)
(gotta love these drive prices!!)
Kevin Burtch
P.S. This is an IDE system, not SCSI. (if it makes a difference)
------------------------------
From: hacker@kaiwan.com (Tim Laren)
Subject: PCI-Herc video working???
Date: 2 Oct 1994 23:59:06 -0700
I have an Intel chipset P5-90 motherboard with a Herc PCI video card with an
ET4000/32W chip. The docs say it works but sometimes I get a flicker when
going into X and then a black screen. I can escape out with ALT-CTL-BSPC. The
system works great in Dos Mode and running Windows, in all resolutions. Once
in X, if I can see it, it works great, never even a little flakey. It seems to
not work in batches, maybe a few times, then works great for a while. I have
tried two different motherboards and two different video cards. The version of
Linux is the Yggdrasil Plug-N-Play, Fall Release.
Any ideas???
I check this group but wouldn't mind an email if you got a fix.
Thanks
--
========================================================================
Hacker Electronics (818) 882-7980 Tue-Sat hacker@kaiwan.com
21010 Devonshire St (818) 772-9934 BBs/Fax hackerec@ix.netcom.com
Chatsworth, CA 91311 (818) 772-0435 ISDN FidoNet: 1:103/811
------------------------------
From: ubacr46@naga-1.uucp (G Cheng)
Subject: Help: nslookup works but others fail
Date: Thu, 6 Oct 1994 17:31:18 +0100 (BST)
As the document mentions the resolver supports up to 3 nameservers,
I set up three lines in resolv.conf. But if the first server stops
working, nslookup still works with alternative servers but other
domain-lookup all fail, such as ping, telnet.
My kernel is 1.1.45 and My /etc/host.conf:
order hosts nis bind
multi on
Thanks for your suggestion.
--
G Cheng ~{<=3G~} Department of Computer Science
G.Cheng@dcs.bbk.ac.uk Birkbeck College, University of London
------------------------------
From: donadio@mxd120.rh.psu.edu (Matthew Donadio)
Subject: Re: PACKET SNIFFER FOR LINUX
Date: 6 Oct 1994 16:08:50 GMT
CVL staff member Nate Sammons (nate@matisse.VIS.ColoState.Edu) wrote:
: So, where is one? I have heard that there is a "cool network monitor" for
: Linux, and I want to take a look at traffic on my network, and I am
: one of the "good guys"
Poke through the INDEX on sunsite. Either that or get a copy of
Stevens, _TCP/IP_Illustrated,_Volume_1_ and write your own. I wrote
one in about four hours.
--
Beaker aka Matt Donadio | Life is short, --- __ o __~o __ o
donadio@mxd120.rh.psu.edu | ride like ---- _`\<, _`\<, _`\<,
--- Penn State Cycling ---| the wind. --- ( )/( ) ( )/( ) ( )/( )
====================================URL: http://mxd120.rh.psu.edu/~donadio
------------------------------
From: wmb@sodapop.cc.latech.edu (Bill Braughton)
Subject: dump for Linux?
Date: 1 Oct 1994 01:32:48 GMT
Reply-To: wmb@sodapop.cc.latech.edu
Does anyone know by chance if there's a dump for Linux floating around? I am
in need of the dump command that usually resides with /etc on most other
Unices. I am attemping to setup the Amanda backup system so I can back my
Linux station up along with the rest of my network which already uses it to
back up. Please don't suggest dumpe2fs as it isn't complete enough to supply
Amanda with what it needs. And I am using the ext2 filesystem. Any suggestions,
ideas or comments can be posted or e-mailed to me at wmb@sodapop.cc.latech.edu.
Thanks. :)
Bill
------------------------------
From: support@zeos.com (Zeos Technical Support account for Internet)
Subject: Re: New Adeptec SCSI not detected
Date: 6 Oct 1994 09:51:47 -0500
The MadMan (madman@aloha.com) wrote:
: Hi. I'm trying to install Linux Slackware 2.0.0 on my machine,
: and am having a bit or a problem. Here's my stats:
: Zeos 486-66
: 16MB RAM
: Western Digital 220MB IDE Hard disk
: Adeptec AIC-6360 SCSI Host Adapter
: Quantum Empire 1GB SCSI HD
: Microcomm Desporte FAST ES 28.8 Modem
: I've installed the SCSI drivers, and everything works peachy thru
: DOS. I FDISked my 1GB and left about 200MB on it for the Linux Native
: partition. When I boot Linux, I use the SCSINET boot disk and the
: COLOR144 root/install disk. It goes thru all it's stuff, and when it
: gets to the SCSI Hosts part, it says "0 SCSI Hosts detected" and "out of
: that 0 are SCSI Hard Disk devices".
: My Question: Is there a new driver for Slackware that supports
: the AIC-6360 so that I can FDISK a Linux Native Partition and install the
: system?
You need to have a boot disk that gives you the LILO boot: prompt. At that
prompt, type:
ramdisk aha152x=0x340,11,7,1
Then, when you get Linux installed, you will have to install LILO and
add the following line to the /etc/lilo.conf file:
APPEND='aha152x=0x340,11,7,1'
That should ge you up and running.
Kevin
--
Zeos Technical Support | | ZEOS International, Ltd.
support@zeos.com INET | Support account for | Technical Support Dept.
uunet!support!support UUCP | Zeos Technical Support | 1301 Industrial Blvd N.E.
612-362-1207 FAX | | Minneapolis, MN 55413
------------------------------
Date: Wed, 05 Oct 1994 13:18:01 +0100
From: pit@p2.lxs.baboon.ch (Peter Berger)
Subject: Slip autoanswer
bob4@slb.com wrote:
> [...]
> When I posted this the other day, I got a followup from someone saying
> that was no good, because he didn't want the modem to answer the phone
> when Linux was crashed. I don't understand. I have a dialup system
> here in the office which I have set up as a Slip server. The modem
> is set to Auto Answer, and I don't have any problems.
It's a matter of taste. One can decide to have the computer (controlled by you
and your wishes) to control the modem. Or one can decide to let a dump machine
(the modem) react in a sometimes annoying way for the caller (e.g. if there's
no getty - or such - running..).
In short: It's a decision if one uses the (imho better) possibilities that
exist or not.
> Perhaps the person who replied before would care to expand his
> comments...
It could have been me, so I took the plunge..
> BTW, you could try reading the modem manual ! (grin ..)
Always a good advice ;-)
(hope your "DIP 5 switch"-tip will not be used by too many fellows - there's
different modems).
bye,
Peter
E-Mail: pit@lxs.baboon.ch
------------------------------
Date: Wed, 05 Oct 1994 16:58:02 +0100
From: pit@p2.lxs.baboon.ch (Peter Berger)
Subject: slack and named: Is this fixed in 2.1*
Hi All,
well, a little late "bug" report which may be already obsolet, but still may
help some fellows out there.. What I'm talking about:
Having the N disks (Slackware v2.0) installed one finds a file /etc/named.pid
on the hard disk, indicating that this will show the pid of named (if started).
So far, so good. BUT!
/usr/sbin/named does use/create its named.pid in /usr/tmp, not /etc !
The scripts /usr/sbin/named.restart|reload (they have softlinks in /etc) have
to be adjusted to use the right named.pid (and also to the path for named
inthere is wrong) or one may get some unexpected "surprise"..
.. just my 0.02$ worth
bye,
Peter
e-mail: pit@lxs.baboon.ch
------------------------------
From: ind43@sun1000.ci.pwr.wroc.pl (Marek Michalkiewicz)
Subject: ftape + Highscreen/Vobis tape drive = ?
Date: 6 Oct 1994 12:09:09 GMT
Hi,
I'm going to buy a tape drive for backups. Does anyone have any
experiences with the Highscreen/Vobis tape drive under Linux?
It is a cheap "250MB" (really 120MB) tape drive, which connects
to the floppy controller. Will it work with ftape?
Yes, I have RTFM (Ftape-HOWTO) but this tape drive is not listed
there. They at the Vobis shop don't know about Linux.
Thanks in advance.
Marek Michalkiewicz
ind43@ci3ux.ci.pwr.wroc.pl
------------------------------
From: nfykw@alfred.uib.no (Kristian Mowinckel)
Crossposted-To: comp.os.linux.help,comp.os.linux.misc
Subject: Disk partitioning and moving filesystems.
Date: 7 Oct 1994 00:01:03 GMT
Disk partitioning and moving filesystems.
=========================================
Firstly I must mention that I have read
Wirzenius's newsarticle of 2 jul 1994 18:21 about disk partitioning.
However I have some questions.
I recently bought a 420Mb Conner ide drive.
And I have a 240Mb Samsung ide drive.
Currenly my 240Mb Samsung drive is my hda
and my 420Mb Conner drive is my hdb.
The partitions are:
Filesystem 1024-blocks Used Available Capacity Mounted on
/dev/hda2 216012 173057 31799 84% / % linux root
/dev/hdb1 385436 1 365529 0% /hdb1 % new linux part
/dev/hda1 5212 2636 2576 51% /dosc % dos part
On /dev/hdb2 I have a 18Mb swap partitition.
Earlier I had a 17Mb swap partitition on /dev/hda3,
but that I deleted.
Since the 420Mb Conner drive is faster I would
like to let this become hda(by changing jumpers on the drives).
Then would like to have one big linux partitition on this,
besides the swap partitition.
On 240Mb Samsung drive I would like to have a small
dos/mswindows partitition and a news spool partition
since I am planning to set up a news server on my box when I get
my slip connecting.
How can I accomplish this?
Especially installing lilo on the 420Mb Conner drive.
Greetings from,
Kristian
/\_/\
(o o)
+---------------------ooO-(o)-Ooo---------------------+
Name: Kristian August Mowinckel +
Postal Address: +
Hamnegata 6 +
7800 Namsos +
+
Country: Norway +
+
Phone: + 47 74 27 43 61 +
+
E-mail: nfykw@alf.uib.no +
+-----------------------------------------------------+
------------------------------
From: rkean@scsn.net (Rob Kean)
Crossposted-To: comp.sys.ibm.pc.hardware.systems,comp.os.linux.misc
Subject: Re: Mystery Chip...AMD
Date: Thu, 6 Oct 1994 01:52:06
In article <36vcaa$2fj@tequesta.gate.net> gterry@gate.net (G. E. Terry) writes:
>RYAN Colin Patrick (ryan@ecf.toronto.edu) wrote:
>: While responding to an add inteh local paper for a $99 486 upgrade it came
>: to light that this upgrade was a quote "486/66 Mhz" which was a "faster chip and less expensive than the i486DX2-66". This propted my query on what the hell
>: this chip was and the response was AMD. I was not aware of this chip. I was
>: under the impression that all the 66's 75's 100's etc (non-Pentium) were
>: overclocked 33 Mhz chips. Does a 'real' 66 Mhz chip exist? If so (and I dont'
>: think so" do traditional mother boards ( ie that could handle a DX2) support
>: this chip. And Finally, if this is true is it compatble and reliable.
>This is just an clock doubled 33. I would say it might have been an error
>in the ad. I have one of these chips. The story I read was that Advanced
>Micro Devices bought the masks for the 80286 from Intel a few years back.
>They decide to try to test the copywrite & patent laws by cloning the
>386 & 486 chips. Well they were very successful and Intel sued. In the
>following trial, Intel LOST. I hear it was due to the naming system that
>they used. It made the chips generic in the eyes of the court. That is why
>there is no 80586, and we have the Pentium.
>As far as reliability, the AMD 486DX2/66 is an exact duplicate of Intel's
>chip, or so I read in PC Magazine. And at an average of 100 to 150 dollars
>less than Intel, I am a buyer!
Actually, in the ever running catch up game between AMD and Intel the price
difference has fallen to $20. BUT, this is about to change--BIG SHOCK!! 8^)..
AMD is about to release, yes you guessed it, their 486DX2-80MHz!!!!!
From what I've heard through my venders, It will run about $20 more than an
Intel 66MHz.
rhk
rkean@scsn.net
------------------------------
From: rkean@scsn.net (Rob Kean)
Subject: Re: [Q] 4M -> 8M RAM worth it?
Date: Thu, 6 Oct 1994 02:06:58
In article <rcas1.781417441@giaeb> rcas1@giaeb.cc.monash.edu.au (Robert Casiano) writes:
>I`ve just installed linux at home. My pc is 486DX33 with 4M of RAM.
>It looks good but x runs very very slow. Thus, it swaps a lot to
>hard disk, and I fear that its life span is being shortened
>considerably. So I might buy more memory. With my tight budget, I
>could only afford another 4M. Question is, is it worth it to buy the
>4M now, or just wait for probably another year and get 16M?n If I
>buy the 4M now, wuold it improve things a lot, or not much
>improvement?
>Cheers :-)
>Robert
It doesn't matter if you're running X, MS Windows, OS/2, NT or any other GUI
that you can think up, upgrading from 4 to 8 Mb of RAM is worth magnitudes
more than the cost of the upgrade. I've have seen (on the high, exagerated
end of examples) as much as a 60% increase in performance.
And, next year, if you can afford 16Mb, buy it. You can always convert the
extra 4Mb into cache on a caching SCSI or IDE controller, or sell it to
someone.
Good luck...
rhk
rkean@scsn.net
------------------------------
From: rph@netcom.com (Randy Hootman)
Subject: Re: shutdown without root access
Date: Wed, 5 Oct 1994 23:29:24 GMT
Have you tried out the utility sudo? It allows you to give other certain
root priviledges.
Randy
: I am trying to set up my system so that others in my family can start
: and stop it without root privileges. The /etc/passwd I inherited from
: the Slackware system installation included:
: sync::5:0:sync:/sbin:/bin/sync
: shutdown:*:6:0:shutdown:/sbin:/sbin/shutdown
: halt:*:7:0:halt:/sbin:/sbin/halt
: I suppose I should use shutdown to bring the system down, since it calls
: halt. When I tried logging in as 'shutdown', I was asked for a password
: and of course nothing was accepted. I removed the '*' from the entry in
: /etc/passwd (so no password was needed), then got the message:
: shutdown: must be root.
: Finally I changed '6' to '0' (so shutdown had the same userid as root),
: and got a usage message:
: Usage: shutdown [-krhnf] time [warning message]
: -k: don't really shutdown, only warn.
: -r: reboot after shutdown.
: -h: halt after shutdown.
: -n: don't sync before reboot or halt.
: -f: do a 'fast' reboot.
: -c: cancel a running shutdown.
: -t secs: delay between warning and kill signal.
:
: I then tried using the entry for halt. When I removed the '*' from its
: entry in /etc/passwd and tried logging in as 'halt', I got the message:
: -halt: can't idle init
: Finally, I put this into /etc/inittab:
: # What to do at the "Three Finger Salute".
: ca::ctrlaltdel:/sbin/shutdown -t3 -h now
: This means I can't trigger a reboot from the keyboard - I have to
: actually press the reset button. No big deal.
: My instructions to shut down the system are:
: 1. logout
: 2. press cntl-alt-del.
: My questions for the net:
: Can you really run shutdown as a login shell?
: When would you log in as 'halt'?
: When would you log in as 'sync'?
: Is my method safe? (filesystem damage, security, etc.)
: Is there a better way?
: - Jim Van Zandt <jrv@vanzandt.mv.com>
:
--
///////////////////////////////////////////////////////////////////////
"In recognizing the humanity of our fellow beings,
we pay ourselves the highest tribute." - Thurgood Marshall
//////////////////////////////////////////////////////////////////////
Randy Hootman Randysoft Software (408) 229-0119
------------------------------
** 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
******************************