Linux-Misc Digest #390, Volume #24 Sun, 7 May 00 12:13:03 EDT
Contents:
Re: SIMPLE HELP (urgent please) ("Peter T. Breuer")
Re: Like Defrag.exe in Win, but in Linux. (YamYam)
Re: "Core" file (YamYam)
How do I change screen resolution (James)
kernal 2.3 ("Michael Ahumibe")
Re: redhat 6.1 install woes (Christoph Kukulies)
Re: newbie question (Rob)
Redhat 6.1 and FDDI (DEFPA) (Christoph Kukulies)
Video Card? ([EMAIL PROTECTED])
Video Card? ([EMAIL PROTECTED])
ide-scsi CD-R Problem With Newest Kernels (.14 & .15) - write_g1?!?!?!? (Douglas E.
Mitton)
Re: Linux woes (Compaq for one) on the horizon ("Peter Hutchison")
Re: Like Defrag.exe in Win, but in Linux. (Stewart Honsberger)
Re: Need to find my IP address (Mario Klebsch)
Re: Need to find my IP address (Mario Klebsch)
Re: Installing lilo (Tony Steidler-Dennison)
Re: Enlightenment can't read Imlib (Lance)
SCSI lockup during formatting? (Peter Eddy)
Re: Redhat 6.1 and FDDI (DEFPA) (Robert Heller)
Re: Need to find my IP address (brian moore)
----------------------------------------------------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: SIMPLE HELP (urgent please)
Date: 7 May 2000 12:03:19 GMT
virgin <[EMAIL PROTECTED]> wrote:
: 1- Get the disk space used and then check it 2 hours later and find the
: difference and display it (such as for disk1 it was 550MB and now 580MB ...)
cron job. mv /etc/space /etc/space.ld; du > /etc/space; diff ...
: 2- Or find the differences in disk space used now and the last time this
: script is used.
ditto.
: 3- Also on the command line when I write ;
: df -h
: how can I filter (get rid of) the fifth and/or sixth column (fields).
man cut/man awk.
: 4- I also need a very simple script to send the result of a command to
: someone in the system or anyone outside the system.
done automatically in cron. Man crontab.
Have you actualy READ your class notes?
Peter
------------------------------
From: YamYam <[EMAIL PROTECTED]>
Subject: Re: Like Defrag.exe in Win, but in Linux.
Date: Sun, 07 May 2000 12:30:03 GMT
jcocasio wrote:
>
>
>
>
>
> In article <[EMAIL PROTECTED]>,
> [EMAIL PROTECTED] wrote:
> > On Sat, 06 May 2000 23:30:07 GMT, Federico Czerwinski wrote:
> > >Hey!, this questoin is a short one, just this, Is there any program
> for
> > >Linux that defrags the disk, just like Defrag.exe in windows? Where
> can i
> > >get it?, Thanx!
> >
> > This is a pretty FAQ, but nevertheless, any excuse to expound apon the
> > virtues of Linux. ;>
> >
> > The ext2 filesystem isn't as prone to fragmentation as the FAT
> filesystem
> > is. (FAT32 included). I've had a Linux system running for almost a
> year,
> > and I think the / partition was about 2.9% "non-contiguous" when I
> > installed a new version of SuSE.
> >
> > I'm sure there exists a defragmentation utility for Linux, but I
> highly
> > don't reccomend it. Besides that, it just isn't neccesary.
> >
> > Linux takes care of itself. Let it. :>
> >
> > --
> > Stewart Honsberger (AKA Blackdeath) @ http://sprk.com/blackdeath/
> > [EMAIL PROTECTED] (Remove 'thirteen' to reply privately)
> > Humming along under SuSE 6.4, Linux 2.2.14
> >
>
> That's about right. Linux just like its big brother UNIX looks for the
> next available block to write data to. Unlike winbloze that goes
> pretty much where it wants. Also if you set up your file system
> properly you should have no issues with fragmentation for a long time
> or even at all. I know there are some utilities out there but it's
^^^^^^^^^^^^^^^^^^^^^^^
And one of them is defrag as u may have it with ur linux
system.
-YamYam.
> just not needed.
>
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: YamYam <[EMAIL PROTECTED]>
Subject: Re: "Core" file
Date: Sun, 07 May 2000 12:30:05 GMT
Plathora wrote:
>
>
> This is a dump of the process image at the time of failure. It is used by
> programmers to debug the program "post mortum". You can safely erase the
> file and aas an added benifit take a look at the "ulimit" man page. There
If u're running bash shell, u can disable creating core files, by putting a line
in /etc/profile:
ulimit -c 0k
or if u're using tcsh shell, put in /etc/profile a line:
limit coredumpsize 0
Note!: This will be a global option -for all users-, so if u wanna make it a private
option for a specific user go to his HOME directory and put the appropriate above
line -depending on ur shell- in ur startup script. i.e., if u're using bash use the
file .bash_profile, or if u're using tcsh use the file .login
-YamYam.
> is/should be an option to disable "core" images from being created. There
> is on all the other Unix flavours but I can't remember the flag.
>
> Plathora.
>
> Federico Czerwinski <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
> > Hey! This site rules!, Here's the question: When i got an error (usualy
> > under X) a file is created, named "CORE", which is quite larger......can i
> > erase it? What is that file!? What's it for? Thanx a lot!
> >
> >
> > Federico
> >
> > --
> > Posted via CNET Help.com
> > http://www.help.com/
>
>
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: James <[EMAIL PROTECTED]>
Subject: How do I change screen resolution
Date: Sun, 07 May 2000 12:30:05 GMT
I recently installed linux Mandrake 7.0 on my computer, and when I
rebooted it after installation I found that the screen resolution was
640 x 480 not 800 x 600 as I had specified during the install.
I have tried using DrakConf and xsetup to change the settings, but
when I select a different resolution and colour depth and then reboot
everything still looks the same.
I also tried to make manual changes to the XF86Config file, but after
the reeboot they had all been overwritten.
I have used mandrake 7 on my computer before, and so I know that I
should be ablt to use a higher resolution.
Does anyone know a definite way to change screen resolution, and why my
manual changes are bieng erased?
Thankyou.
JL
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: "Michael Ahumibe" <[EMAIL PROTECTED]>
Subject: kernal 2.3
Date: Sun, 7 May 2000 13:16:55 +0100
hi
does anyone know where I can download kernal 2.3 from?
would I lose all my old kernal settings doing this?
thank you
mike
------------------------------
From: Christoph Kukulies <[EMAIL PROTECTED]>
Subject: Re: redhat 6.1 install woes
Date: 7 May 2000 13:23:01 GMT
Leonard Evens <[EMAIL PROTECTED]> wrote:
: Christoph Kukulies wrote:
:>
:> I tried to install RH 6.1 on a P90 ASUS PCI/P54SP4 board, 64MB,
:> 2 IDE (M-2.2 GB Maxtor, S-2.1 GB Quantum FB), ELSA Victory Erazor,
:> Intel Etherexpress NIC.
:>
:> Nothing fancy, as one can see.
:>
:> Installing runs fine, I install a minimum system, on root
:> partition of 2192 MB (or something like that), 350 MB swap on second IDE.
:>
:> But when I reboot first time I just can see the letters LI
:> (obviously from LILO) and then stop. Nothing anymore. I can only
:> CTRL-ALT-DEL.
:>
:> Tried this now three times, even with IDE Normal mode.
:>
:> --
:> Chris Christoph P. U. Kukulies [EMAIL PROTECTED]
: In principle, the installation should have created a boot
: floppy for you. (If it didn't, reinstall but use the
: upgraded Anaconda installation floppy images from the RedHat
: Errata or a mirror site. Actually, you could probably just
: choose the upgrade option if you have already installed.)
: Boot from that and use fdisk to
: show your partitioning. For IDE disks,
: fdisk -l /dev/hda
: fdisk -l /dev/hdb
: will exhibit the partitioning information. That way we can
: tell if you have encountered the 1024 cylinder limit.
: For completeness, also show us what
: more /etc/lilo.conf
: shows, but I expect that would not be the problem.
: If you aren't encountering the 1024 cylinder limit problem,
Well it looked like it was the 1024 cyls problem
in the first place. My BIOS had assumed totally bogus figures.
It showed 4465/15/63 while the boot probe found
523/128/63.
I added the latter figures into the BIOS and it showed 43 MB
for the disk. That's probably modulo 2048 or something it seems.
Anyway, I re-installed a fourth or fifth time but still can't
boot. I *can* boot the system through the boot floppy though
which gets me a bit further but it's not tolerable in the long
run.
: there could be some geometry mismatch. You would have to give
: us more detail about your BIOS and whatever else you can tell
: us. And of course, there could just have been some glitch
: in the installation such that lilo was not run properly and
: doing again will resolve the issue.
Once when the system is booted, can I repair this? E.g. get a LILO and/or
BOOTINST or something written correctly to the disk?
: If you are encountering the 1024 cylinder limit problem,
: although there is a new version of lilo which overcomes the
Where is this new version of Lilo?
: problem, you probably can't use it because your BIOS probably
: doesn't support the extended BIOS call that is necessary.
: You can try adding the linear option and see what happens,
: but either you will have to repartition with /boot in
: a partition below cylinder 1024, use loadlin or something
: similar, or just boot from a floppy.
: --
: Leonard Evens [EMAIL PROTECTED] 847-491-5537
: Dept. of Mathematics, Northwestern Univ., Evanston, IL 60208
--
Chris Christoph P. U. Kukulies [EMAIL PROTECTED]
------------------------------
From: Rob <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.networking
Subject: Re: newbie question
Date: Sun, 07 May 2000 09:35:02 -0400
Gents, I am having the same problem, Ive tried 3 different nic cards and
get eth0 not found, or device not found, its kinda irritating, as i been
messing with this for a bout a month on and off. Did you get this problem
fixed????
adam wrote:
> josef,
>
> thanks for the response.
>
> yes, i think the rtl8139 driver is what i need. it is offered on the
> smc website for this card. so, i'm off to read the kernel howto right
> now. since i have a dual-boot system, i am hoping that i will be able
> to save a copy of the driver in windows and read the floppy in linux.
> or, is there a way to access my windows file system through linux?
>
> i'll let you know if and WHEN i need more help.
>
> regards,
> adam
>
> In article <[EMAIL PROTECTED]>,
> Josef Oswald <[EMAIL PROTECTED]> wrote:
>
> > I checked with the SuSE Hardware base ( since I am running SuSE and
> > there is only a 1211 (no _tx_) 100Mbit a driver is listed as
> rtl8139....
> >
> > Could it be that you need to compile your kernel and use the RealTeK
> > 8139 driver..
> >
> > I checked with make xconfig and the above Card is listed in
> > Western-Digital as 8139 support try that if it works :-)
> > sorry I wish I could help you more :-/
> >
> > if you need help in compiling the kernel just let me know :-)
> >
> > hth :-)
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
------------------------------
From: Christoph Kukulies <[EMAIL PROTECTED]>
Subject: Redhat 6.1 and FDDI (DEFPA)
Date: 7 May 2000 13:26:27 GMT
I want to configure a RH6.1 system as a router between
FDDI and 100MBit Fast ethernet.
I went into the graphical menu to add a module,
chose eth1 and can't find a module DEFPA.
Am I lost with Linux here? Should I go over to FreeBSD since I
know it's working there?
--
Chris Christoph P. U. Kukulies [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.questions
Subject: Video Card?
Date: Sun, 07 May 2000 13:23:42 GMT
Hi,
I planned to buy a new video card.
I checked some and I come up with two probable candidates:
a. ATI Rage Fury Maxx 64MB
b. Geforce 256 DDR 32 MB
My question is, are these two supported in XFree 3.3.6? I'm using
Mandrake7.0 GNU/Linux.
If somebody knows, I appreciate it if you can let me know by responding
to this post.
Thanks in advance.
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.questions
Subject: Video Card?
Date: Sun, 07 May 2000 13:21:53 GMT
Hi,
I planned to buy a new video card.
I checked some and I come up with two probable candidates:
a. ATI Rage Fury Maxx 64MB
b. Geforce 256 DDR 32 MB
My question is, are these two supported in XFree 3.3.6? I'm using
Mandrake7.0 GNU/Linux.
If somebody knows, I appreciate it if you can let me know by responding
to this post.
TIA,
aollosa
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED] (Douglas E. Mitton)
Crossposted-To: alt.os.linux,comp.os.linux.hardware
Subject: ide-scsi CD-R Problem With Newest Kernels (.14 & .15) - write_g1?!?!?!?
Reply-To: [EMAIL PROTECTED]
Date: Sun, 07 May 2000 15:25:29 GMT
I have done several searches for this issue, there seem to be a lot of
people experiencing it BUT I have not been able to find a solution
yet.
In kernels V2.2.14 and .15 I get random cdrecord failures such as:
(Sorry, it wraps a little.)
Starting new track at sector: 0
Track 01: 175 of 311 MB written (fifo 100%).cdrecord: Input/output
error. write_g1: scsi sendcmd: retryable error
CDB: 2A 00 00 01 5E C0 00 00 10 00
status: 0x2 (CHECK CONDITION)
Sense Bytes: F1 00 05 00 01 5E C0 0C 00 00 00 00 10 02 00 00
Sense Key: 0x5 Illegal Request, deferred error, Segment 0
Sense Code: 0x10 Qual 0x02 (id crc or ecc error) [No matching
qualifier] Fru 0x0
Sense flags: Blk 89792 (valid)
cmd finished after 3.168s timeout 40s
write track data: error after 183894016 bytes
Sense Bytes: 70 00 00 00 00 00 00 0C 00 00 00 00 00 00 00 00 00 00
Writing time: 310.268s
Fixating...
Fixating time: 76.473s
cdrecord: fifo had 5740 puts and 5613 gets.
cdrecord: fifo was 0 times empty and 5206 times full, min fill was
95%.
This all works perfectly in kernel V2.2.13. I always use the same
config for each kernel upgrade/compile and use the "make oldconfig"
to incorporate it.
The other things I've tried are:
- Searching deja news
- reading the links off of the cdrecord home page.
- recompiling cdrecord on my system (V1.8a29)
- Installing the newest cdrecord (v1.8.1)
Does any one have any other experience to throw at this problem? My
next tact is to start comparing the SCSI source between 2.2.13 and
2.2.14/15. I don't hold out a lot of hope on this. The modules I've
read so far have revision dates well before this problem showed up.
Thanks in advance for any insight.
------------------------------------------------
Doug Mitton - Brockville, Ontario, Canada
'City of the Thousand Islands'
EMail: [EMAIL PROTECTED]
http://www.cybertap.com/dmitton
Other: mitton.dyndns.org
SPAM Reduction: Remove "x." from my domain.
------------------------------------------------
------------------------------
From: "Peter Hutchison" <[EMAIL PROTECTED]>
Crossposted-To: alt.linux,comp.os.linux.hardware,comp.os.linux.setup
Subject: Re: Linux woes (Compaq for one) on the horizon
Date: Sat, 6 May 2000 19:13:22 +0100
> If you run defrag in Win98, and show the details (I'm at work watching one
> go right now, whee!), it has this gigantic block of unmovable system files
> right up at the beginning. I don't know what these are, I've never worked
> on one of these until a couple of months ago when I started working here.
> Now I'm curious, so I guess I'll have to do some poking around for some
more
> info on these "freaks of the PC world".
That file is the swap file used for virtual memory (instead of a swap
partition like
on Linux). What you can do is turn off virtual memory (see System Control
Panel)
and do a defrag again....
Peter
------------------------------
From: [EMAIL PROTECTED] (Stewart Honsberger)
Subject: Re: Like Defrag.exe in Win, but in Linux.
Reply-To: [EMAIL PROTECTED]
Date: Sun, 07 May 2000 14:32:36 GMT
On Sun, 07 May 2000 08:28:26 GMT, [EMAIL PROTECTED] wrote:
>> Linux takes care of itself. Let it. :>
>That's about right. Linux just like its big brother UNIX looks for the
>next available block to write data to. Unlike winbloze that goes
>pretty much where it wants. Also if you set up your file system
>properly you should have no issues with fragmentation for a long time
>or even at all. I know there are some utilities out there but it's
>just not needed.
Another belief Windy users seem to carry over, and this being into the world
of OS/2, is that "CHKDSK is outdated. Use ScanDisk instead."
The problem, it would seem, is that most people have been spoon-fed Win'**
all their lives, so they don't know what a real operating system can do. Even
a friend of mine, a huge OS/2 proponant, still believed in 8.3 filenames. My
question to him was always "Why?". OS/2's HPFS, much like Linux's ext2, won't
turn a filename into filena~1.ext.
Out of sheer curiosity, I just ran e2fsck -n on my / partition, and found it
to be a whopping 1.8% non-contiguous. This system was installed about a month
ago, and currently has an uptime of ~8.5 days. Were this a Win'** system
running FAT(32), I'd have had to defrag two or three times by now.
Especially considering I've done such things as download and untar the source
code of XFree86 (not a small undertaking of disk space, y'know?), and shortly
thereafter removed it. Shortly after installing my system, I decided to split
off my /opt directory to another partition. That meant moving 100M of
installed software off of the / partition and onto another (physically
located on another HDD). I've also moved my FidoNet echomail areas (about
200 or so files, some as small as 256 bytes, and others as large as 10M)
to another partition - and yet the drive remained at 1.8%.
I'd love to see any FAT system do that. :>
--
Stewart Honsberger (AKA Blackdeath) @ http://sprk.com/blackdeath/
[EMAIL PROTECTED] (Remove 'thirteen' to reply privately)
Humming along under SuSE 6.4, Linux 2.2.14
------------------------------
From: [EMAIL PROTECTED] (Mario Klebsch)
Crossposted-To:
comp.os.linux.development.apps,comp.os.linux.development.system,comp.os.linux
Subject: Re: Need to find my IP address
Date: Sun, 7 May 2000 14:48:15 +0200
[EMAIL PROTECTED] (Chris) writes:
>On Sat, 6 May 2000 17:51:38 +0200, [EMAIL PROTECTED] (Tobias Anderberg)
>wrote in comp.os.linux.development.apps:
>> int fd;
>> struct ifreq i;
>> fd = socket(AF_INET, SOCK_DGRAM, 0);
>> strncpy(i.ifr_name, "eth0", 5);
>> ioctl(fd, SIOCGIFADDR, (int) &i);
>> close(fd);
>> return (char *)inet_ntoa(((struct sockaddr_in *)
>> &i.ifr_addr)->sin_addr);
>Herein lies one of my biggest complaints about the Linux development
>environment: there should be no reason why an application programmer must
>rely on undocumented "catch-all" calls to accomplish simple tasks.
The SIOCGIFADDR seems to be prety standard anyway. You are right, it
is one of those not so well known features, that separate gurus from
wannabegurus. But in this code, there is a different IMHO major design
problem, that is commonly ignored: Where the hell does the
applikations get the NAME of the primary netword interface? In this
case, it is hardcoded, whoch is hardly portable.
73, Mario
--
Mario Klebsch [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Mario Klebsch)
Crossposted-To:
comp.os.linux.development.apps,comp.os.linux.development.system,comp.os.linux
Subject: Re: Need to find my IP address
Date: Sun, 7 May 2000 14:53:59 +0200
[EMAIL PROTECTED] (brian moore) writes:
>Yep, though for some applications, you need to know it (such as ftp and
>the god-awful PORT command it uses -- gads, that protocol needs
>replacing bad... all in all it's just plain evil).
But that should not be that bad. I would use getsockname() on this
ocasion.
73, Mario
--
Mario Klebsch [EMAIL PROTECTED]
------------------------------
Crossposted-To: comp.os.linux.setup
From: Tony Steidler-Dennison <[EMAIL PROTECTED]>
Subject: Re: Installing lilo
Date: Sun, 07 May 2000 15:19:59 GMT
I'm not sure about Debian. I'm using RH 6.1 on hdb. Win '98 is on hda.
Lilo works fine using the following /etc/lilo.conf:
boot=/dev/hda
map=/boot/map
install=/boot/boot.b
prompt
timeout=50
default=win
image=/boot/vmlinuz-2.2.12-20
label=linux
initrd=/boot/initrd-2.2.12-20.img
read-only
root=/dev/hdb1
other=/dev/hda1
label=win
This tells lilo that the MBR is on hda ('boot=/dev/hda'). It also makes
Win '98 the default boot partition (default=win), unless I key in 'linux'
in the 5 seconds after lilo appears.
If you edit your /etc/lilo.conf file to something similar to this, you
need to be sure, afterwards, to use /sbin/lilo to commit lilo to the MBR.
Tony
**************************
Email is packaged by intellectual weight, not volume. Some settling of contents may
have occurred during transmission.
On Sun, 7 May 2000, wmuser wrote:
> Date: Sun, 07 May 2000 11:15:49 +0200
> From: wmuser <[EMAIL PROTECTED]>
> Newsgroups: comp.os.linux.misc, comp.os.linux.setup
> Subject: Installing lilo
>
> I've recently installed Debian on my pc, but I failed to get lilo
> working
> properly. The problem seems to be that my linux partition is on the
> second hd. From what I've read in the doc-files a solution would be
> to install lilo in the MBR, that is on /dev/hda if I got it right. The
> problem
> is it doesn't say how you actually install lilo there: I installed it on
> the
> second hd while in the Debian installation tool (which seemingly didn't
> give me the option to install on the first hda, which only contains a
> w95
> partition). How do I install lilo?
>
> --
> Alex Borghgraef
>
>
>
------------------------------
From: Lance <[EMAIL PROTECTED]>
Subject: Re: Enlightenment can't read Imlib
Date: Sun, 07 May 2000 15:30:10 GMT
Hello there,
I'm running RedHat 6.2 (with Gnome) on one of my older computers, a
Pentium 150, 48mb RAM with 850mb and 1.2gb hard drives. {All drives are
Linux natives}.
Recently, I upgraded from Enlightenment (0.15.5-48) to the latest
Enlightenment (0.16.4-1, RPM package) on my system with Imlib (1.9.8).
Along with it, I also upgraded to the latest libraries (glib, fnlib, et
cetera), as much as I could. The installation(s) went smoothly, with no
error warnings whatsoever.
However , upon rebooting my system, Enlightenment gave an error, saying
that the Imlib package could not be read. From what I have observed, it is
possible for Enlightenment to run with Imlib(1.9.8), isn't it?
Enlightenment doesn't seem to explicitly state that it requires Imlib2.
I sincerely hope someone can help me out with this. I'm quite new to
this, just a few days in fact.
Thanks for everything. Best regards...
Lance
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: Peter Eddy <[EMAIL PROTECTED]>
Subject: SCSI lockup during formatting?
Date: Sun, 07 May 2000 11:49:28 -0400
Hi, I've just installed a new SCSI disk and I'm experiencing lockups
during formatting of the disk. I'm able to partition it (I gave it a
128M swap partition and the rest to one big ext2 partition) but when I
format it using the -c (check for bad sectors) my machine freezes about
1/3 of the way into the process.
One thing that makes me nervous, since I'm unexperienced with them, is
that the Hitachi is an 80 pin SCA disk that required an adapter. The
adapter has optional termination, I've tried with and without the
termination with the same problem or typical no-termination problems.
Here's my info:
RH Linux 6.1, kernel 2.2.12-20, Intel
BusLogic BT-950, driver version 2.1.15, Aug 1998, Firmware 5.02
SCAM enabled, level 2
The disk in question is a Hitachi model DK319H-18WC, Rev: HF8G
and it's the SCSI ID is 2. It shares the bus with one other disk,
an IBM DGHS09U 9M disk with SCSI ID 1.
Thanks for any advice,
Peter
------------------------------
From: Robert Heller <[EMAIL PROTECTED]>
Subject: Re: Redhat 6.1 and FDDI (DEFPA)
Date: Sun, 07 May 2000 15:58:39 GMT
Christoph Kukulies <[EMAIL PROTECTED]>,
In a message on 7 May 2000 13:26:27 GMT, wrote :
CK> I want to configure a RH6.1 system as a router between
CK> FDDI and 100MBit Fast ethernet.
CK>
CK> I went into the graphical menu to add a module,
CK> chose eth1 and can't find a module DEFPA.
CK>
CK> Am I lost with Linux here? Should I go over to FreeBSD since I
CK> know it's working there?
No, you need to re-build the kernel. The DEFPA driver cannot be added
as a module (I don't know why). Make sure your kernel sources are
installed, go to /usr/src/linux and do a make xconfig (turn on the
DEFPA driver), make clean, make depend, make bzImage. *Make sure you
have at least a boot floppy*. When the build completes there should be
a bzImage file in /usr/src/linux/arch/i386/boot/. Copy this file to
/boot/vmlinuz-2.2.12-20-defxx, copy /usr/src/linux/System.map to
/boot/System.map-2.2.12-20-defxx, and then edit /etc/lilo.conf to
include an additional image (DON'T DELETE THE ORIGIAL image= BLOCK):
image=/boot/vmlinuz-2.2.12-20-defxx
label=linux-defxx
initrd=/boot/initrd-2.2.12-20.img
read-only
root=/dev/sda1
run lilo (MAKE SURE YOU HAVE A BOOT FLOPPY, JUST IN CASE). Make sure
/etc/lilo includes
prompt
timeout=50
and reboot. Select 'linux-defxx single' at the boot prompt and watch
the machine boot up and note that the fddi card is seen. The device
will be named fddi0. If all is good, edit /etc/lilo.conf to have a
line like:
default=linux-defxx
run lilo again and exit from the SI bash prompt to get the rest of the
way up. You can now fire up X11 (locally or remotely) and use netconfig
to configure your fddi card. Note: you won't have to add a module. The
fddi device will show up as an available interface.
CK>
CK> --
CK> Chris Christoph P. U. Kukulies [EMAIL PROTECTED]
CK>
--
\/
Robert Heller ||InterNet: [EMAIL PROTECTED]
http://vis-www.cs.umass.edu/~heller || [EMAIL PROTECTED]
http://www.deepsoft.com /\FidoNet: 1:321/153
------------------------------
From: [EMAIL PROTECTED] (brian moore)
Crossposted-To:
comp.os.linux.development.apps,comp.os.linux.development.system,comp.os.linux
Subject: Re: Need to find my IP address
Date: 7 May 2000 16:02:54 GMT
On Sun, 7 May 2000 14:48:15 +0200,
Mario Klebsch <[EMAIL PROTECTED]> wrote:
> [EMAIL PROTECTED] (Chris) writes:
>
> >On Sat, 6 May 2000 17:51:38 +0200, [EMAIL PROTECTED] (Tobias Anderberg)
> >wrote in comp.os.linux.development.apps:
>
> >> int fd;
> >> struct ifreq i;
> >> fd = socket(AF_INET, SOCK_DGRAM, 0);
> >> strncpy(i.ifr_name, "eth0", 5);
> >> ioctl(fd, SIOCGIFADDR, (int) &i);
> >> close(fd);
> >> return (char *)inet_ntoa(((struct sockaddr_in *)
> >> &i.ifr_addr)->sin_addr);
>
> >Herein lies one of my biggest complaints about the Linux development
> >environment: there should be no reason why an application programmer must
> >rely on undocumented "catch-all" calls to accomplish simple tasks.
>
> The SIOCGIFADDR seems to be prety standard anyway. You are right, it
> is one of those not so well known features, that separate gurus from
> wannabegurus. But in this code, there is a different IMHO major design
> problem, that is commonly ignored: Where the hell does the
> applikations get the NAME of the primary netword interface? In this
> case, it is hardcoded, whoch is hardly portable.
That's what SIOCGIFCONF is for.
See section 16.5 of Unix Network Programming by W. Richard Stevens.
Again, though, you will still have problems determining -which- IP
number is relevant.
For example, a relatively common setup is to have a machine with 2 NICs,
one going to a private network shared with something like a Network
Appliance NFS server.... it would be wrong to assume the address used on
that NIC is public, since it's probably in RFC1918 space.
--
Brian Moore | Of course vi is God's editor.
Sysadmin, C/Perl Hacker | If He used Emacs, He'd still be waiting
Usenet Vandal | for it to load on the seventh day.
Netscum, Bane of Elves.
------------------------------
** 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: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.misc) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Misc Digest
******************************
