Linux-Setup Digest #203, Volume #19 Thu, 20 Jul 00 01:13:09 EDT
Contents:
POP-3 (Michael)
Re: suse manual (E J)
Re: Cannot install LILO (E J)
DHCP works for a while, then stops... (Alex Tchhailo)
Re: Ambient Technologies Inc. MODEM V.90 PCI MD5620 (E J)
Re: @Home Cable Modem Setup Problems (Bit Twister)
SC400 bootloader (Jeff Say)
Problem with X ("OK")
Re: smc-ultra ethernet card ("Mark VanBogart")
Re: Video Capture cards (Pete Zaitcev)
Re: IDE Tape Install Info (bullwinkle)
Re: DHCP works for a while, then stops... (Rob Kroll)
Re: Newbie modem / USB install problem ("John Gabriel")
Re: capture screen output during booting (MH)
Re: gtk-config glib-config and configure (Narbey Derbekyan)
Re: Simple question on Kernel Recompiling (Jason)
@Home Cable Modem Setup Problems (2) ("Dj")
----------------------------------------------------------------------------
From: Michael <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking
Reply-To: [EMAIL PROTECTED]
Subject: POP-3
Date: Thu, 20 Jul 2000 01:14:47 GMT
Hi all,
I have installed Redhat Linux 6.1 and I want to use it as a web server
(which is currently only connected to my local network and has some
access to the Internet).
Now, I would like to get the POP-3 service up and running.
What I did:
I activated the 3 lines in /etc/inetd.conf and HUPped the inetd.
When I use a telnet session to the server at port 110 it connects but
disconnects after 3 seconds (not showing anything; so I don't see
anything like POP-3 ready or something).
When I use the command prompt on the Redhat Linux 6.1 computer and type
'telnet localhost 110' I get the following:
Trying 127.0.0.1...
Connected to localhost.localdomain.
Escape character is '^]'.
and then something like 'Connection failed'
Has anyone got any idea of how I can get the POP-3 server up and
running?
Thanks in advance for your help.
Best Regards,
Michael
------------------------------
From: E J <[EMAIL PROTECTED]>
Subject: Re: suse manual
Date: Wed, 19 Jul 2000 18:16:57 -0700
I hope you mean KDE not Windoze. Use kpackage (GUI) or rpm to extract
suse_lxe.rpm
$ kpackage
H Briggs wrote:
> Hi, i recently tried to install suse linux 6 but encountered probs
> when trying to configure xfree and my vid card ( ati 3d rage pro agp
> x2) I downloaded and installed the xfree4 update but still didn't
> have any luck. before i try again i want to read the suse manual but
> its in a format iam unfamiliar with " suse_lxe.rpm" does anyone know
> what windows program will open this?
------------------------------
From: E J <[EMAIL PROTECTED]>
Subject: Re: Cannot install LILO
Date: Wed, 19 Jul 2000 18:19:50 -0700
Go to www.freshmeat.org
Get the latest lilo, it breaks the 1024 cylinder barrier.
Pascal & Bianca wrote:
> Help!
>
> I want too install Linux on a laptop Compaq Armada E500 450 Mhz. with a
> harddisk
> off about 12Gb. model ibm-dara-212000.
> I installed linux and wanted to install LILO, but it appears that i have
> more than 1024 cilinders in the first Linux boot partition.
>
> O.K., so i want to change that with fdisk.
>
> At first a message appears that the number of cilinders is set to 1559 and
> that is gives problems installing LILO
>
> I go to the extra functionality expert mode, type c and set the number of
> cil. too 1023.
>
> When i write to table and exit, and reboot, the cilinders are set back to
> 1559 head 240 and sectors 63.
>
> Do i have to change the sec and head settings too, and so, too what?
>
> Hope you can help, in the mean time ill be trying too figure it out myself.
>
> Pascal
------------------------------
From: Alex Tchhailo <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help
Subject: DHCP works for a while, then stops...
Date: Thu, 20 Jul 2000 01:21:53 GMT
Hi,
I am using Debian with 2.2.17 kernel. My eth0 (only NIC in my PC) is
RealTek RTL 8139 and is detected ok. To use my @home cable connection I
installed dhcpcd, which works for a while (up to 8 hours actually once),
but then just stops. I noticed however with ifconfig number of errors
for eth0 goes from 0 to 1 (if it means anything). I read in DHCP Howto
that Debian's version need not be configured, just installed. It also
says that if network connection stops after working for a while, it
might be gated interfering, but I'm not running gated.
Please help, Alex.
my email is [EMAIL PROTECTED]
thank you
------------------------------
From: E J <[EMAIL PROTECTED]>
Subject: Re: Ambient Technologies Inc. MODEM V.90 PCI MD5620
Date: Wed, 19 Jul 2000 18:25:26 -0700
If it is a real modem, you have the driver already in RedHat.
If it is a lucent or PCTel go to www.linmodems.org
Other winmodem chipset, go buy a real modem.
Nemanja Todorovic wrote:
> If anybody knows where I could find drivers for my modem
> that would work with Red Hat 6.0-6.2 distribution please
> let me know
>
> Thanks !
>
> Greetings from
> linux fresh meet
>
> Nemanja Todorovic
------------------------------
From: [EMAIL PROTECTED] (Bit Twister)
Subject: Re: @Home Cable Modem Setup Problems
Reply-To: This_news_group.invalid
Date: Thu, 20 Jul 2000 01:37:04 GMT
Sounds like /etc/resolv.conf is not set correctly. Fake example:
search localdomain grlnd1.tx.home.com
nameserver 24.14.82.433
nameserver 24.14.28.434
On Thu, 20 Jul 2000 00:32:36 GMT, Dj <[EMAIL PROTECTED]> wrote:
>Hi there:
>
>Here's what I have: redhat 6.2 and 3COM 509(b) card. IP assigned by DHCP.
>I followed the instructions (used netcfg) to configure eth0, hostnames, DNS,
>etc.
>After I run dhcpcd and ifconfig I can see eth0 interface with assigned IP
>address. I can even see the number of RX/TX packets. I also can see traffic
>going back/forth on gnome network manager.
>The weird thing is that I cannot access anything outside -- I can't even
>resolve the names.
>
--
The warranty and liability expired as you read the message.
If the above breaks your system, it's yours and you keep both pieces.
Practice safe computing. Backup the file before you change it.
Do a, man every_command_here, before doing anything or running a script.
------------------------------
From: Jeff Say <[EMAIL PROTECTED]>
Subject: SC400 bootloader
Date: Wed, 19 Jul 2000 22:24:59 -0400
I have sc400 (486) based PC AT board,4M Flash and 16M DRAM SBC. I do not
want to use any BIOS to load Linux. Board supports headless operation
(no keyboard, no display, no floppy/hard disk) - only serial port can be
used as a console). I need a bootloader to boot Linux. I am trying to
find a bootloader which graps binary linux image and copies to DRAM,
then boots from there.
does anyone knows how to that? if you interested in this project, I will
pay your consultation fee!!
serious guys only!!
TIA,
Jeff Say
------------------------------
Date: Thu, 20 Jul 2000 10:28:10 +0800
From: "OK" <[EMAIL PROTECTED]>
Subject: Problem with X
I installed Linux Mandrake 6.5 and everything is working fine except the X
server. I suspect that it is a problem with my monitor or graphics card. I
am using a Trident 975 AGP card and a monitor brand called Likom. Everytime
I try higher resolution the monitor will be turned off by itself. And it can
only work in a very miserable low resolution and low colour mode. I have
tried changing various combinations of settings but to no avail. Sometimes
the entire system just hangs giving me a blank grey screen.
Hope to get some help here. Thanx
------------------------------
From: "Mark VanBogart" <[EMAIL PROTECTED]>
Subject: Re: smc-ultra ethernet card
Date: Wed, 19 Jul 2000 21:53:05 -0500
This is a multi-part message in MIME format.
=======_NextPart_000_009D_01BFF1CB.B7D8D840
Content-Type: text/plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
IF you are loading it as a mopdules, try compiling it into the kernel. =
I've had several carss that wouln't work as modules but would if built =
into the kernel.
"manhattanian" <[EMAIL PROTECTED]> wrote in message =
news:[EMAIL PROTECTED]...
Hi,=20
I have an old smc-ultra network card installed on Packard Bell 8950 =
computer. The computer can detect its existence (using ifconfig, at =
io=3D0x380 irq=3D10). I can also see it from conf.modules (alias eth0 =
smc-ultra). I also checked the host, ip address, rounter. Everything are =
fine. However, when I tried ping, it does not work.=20
Actually, at one time (several month ago, it worked) It does not =
working since last week I restall the RedHat 6.1. Anybody can remind me =
possible solution.=20
Also, the options smc-ultr io=3D0x380 irq=3D10 does NOT work at all.=20
Thanks.=20
--=20
_________________________
David
Tel: 212-802-6276
Fax: 212-802-6253
=20
=======_NextPart_000_009D_01BFF1CB.B7D8D840
Content-Type: text/html;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content=3D"text/html; charset=3Diso-8859-1" =
http-equiv=3DContent-Type>
<META content=3D"MSHTML 5.00.2919.6307" name=3DGENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><FONT face=3DArial size=3D2>IF you are loading it as a mopdules, =
try compiling=20
it into the kernel. I've had several carss that wouln't work as =
modules=20
but would if built into the kernel.</FONT></DIV>
<BLOCKQUOTE=20
style=3D"BORDER-LEFT: #000000 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: =
0px; PADDING-LEFT: 5px; PADDING-RIGHT: 0px">
<DIV>"manhattanian" <<A=20
href=3D"mailto:[EMAIL PROTECTED]">[EMAIL PROTECTED]</A>> =
wrote in=20
message <A=20
=
href=3D"news:[EMAIL PROTECTED]">news:39762467.445CA825@ba=
ruch.cuny.edu</A>...</DIV>Hi,=20
<BR>I have an old smc-ultra network card installed on Packard Bell =
8950=20
computer. The computer can detect its existence (using ifconfig, at =
io=3D0x380=20
irq=3D10). I can also see it from conf.modules (alias eth0 smc-ultra). =
I also=20
checked the host, ip address, rounter. Everything are fine. However, =
when I=20
tried ping, it does not work.=20
<P>Actually, at one time (several month ago, it worked) It does not =
working=20
since last week I restall the RedHat 6.1. Anybody can remind me =
possible=20
solution.=20
<P>Also, the options smc-ultr io=3D0x380 irq=3D10 does NOT work =
at all.=20
<P>Thanks. <PRE>--
_________________________
David
Tel: 212-802-6276
Fax: 212-802-6253</PRE> </BLOCKQUOTE></BODY></HTML>
=======_NextPart_000_009D_01BFF1CB.B7D8D840==
------------------------------
From: [EMAIL PROTECTED] (Pete Zaitcev)
Crossposted-To: comp.os.linux.misc,comp.os.linux.hardware
Subject: Re: Video Capture cards
Date: Thu, 20 Jul 2000 03:30:57 GMT
> Any good recommendations supported in the latest (stable) kernel? I'm
> looking to capture images from off a video feed (traffic cam) and then
> post the images to a INternet page. We're using something right now but
> it's not handling data well. Works but is kludged together.
> Willing to spend whatever on getting a good video line to a video
> capture card.
> Any recommendations would be appreciated.
> KJ
> --
Use something based on Brooktree Bt848 chipset.
There is a list of known boards inside the bttv.c driver.
--Pete
------------------------------
From: bullwinkle <[EMAIL PROTECTED]>
Subject: Re: IDE Tape Install Info
Date: Thu, 20 Jul 2000 03:30:04 GMT
Doc Shipley wrote:
>
> Wayne Vagtborg wrote:
> >
> > Anyone know where I can get some information or HOWTO/FAQ on
> > installing and using a HP Colorado 8Gb IDE tape drive with
> > RH 6.2 - just want to be able to tar/cpio/dump to the tape.
> > OS recognizes that the tape drive is present - hdd - but I
> > can't get it configured. Sure could use some good advice.
>
> You'll need the ide-tape module loaded. Depending on your distro, you
> can probably just "modprobe ide-tape". The drive may show up as
> /dev/hdX, but to use tar, mt, dump, etc, it's referred to as /dev/ht0.
> You can check operation, rewind, erase, or whatever with "mt", as in
> "mt -f /dev/ht0 status".
>
> --
> Doc Shipley
> Network Stuff
> Austin, Earth
Doc gave you some good information. I have a 20g HP Colorado EIDE
tape drive on a RH6.2 system. You should have everything you need if you
had any ide devices detected during the install. You probably want to use
/dev/nht0 (non-rewinding) when you do any back-ups.
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
Crossposted-To: comp.os.linux.help
Subject: Re: DHCP works for a while, then stops...
From: [EMAIL PROTECTED] (Rob Kroll)
Date: Thu, 20 Jul 2000 03:50:55 GMT
[EMAIL PROTECTED] (Alex Tchhailo) wrote in <[EMAIL PROTECTED]>:
>Hi,
>I am using Debian with 2.2.17 kernel. My eth0 (only NIC in my PC) is
>RealTek RTL 8139 and is detected ok. To use my @home cable connection I
>installed dhcpcd, which works for a while (up to 8 hours actually once),
>but then just stops. I noticed however with ifconfig number of errors
>for eth0 goes from 0 to 1 (if it means anything). I read in DHCP Howto
>that Debian's version need not be configured, just installed. It also
>says that if network connection stops after working for a while, it
>might be gated interfering, but I'm not running gated.
>Please help, Alex.
>my email is [EMAIL PROTECTED]
>thank you
>
>
Your problem could be related to the lease time that @Home allows.
(Although I think that the default lease is 1 day, with a maximum lease of
30 days.)
In any case, why are you running with DHCP over Rogers@Home anyway? (I
think it is Rogers in Alta. Could be Shaw.) Use static. The only times I
know of where somebody's IP address has changed is when they move or switch
from a Lancity to a Terayon modem.
I, for example, have had my IP address since I started with @Home. They've
threatened to cycle the IP address, but they haven't. And they likely won't
until I switch over (permanently) to 3-meg ADSL.
HTH
-RK
------------------------------
From: "John Gabriel" <[EMAIL PROTECTED]>
Subject: Re: Newbie modem / USB install problem
Date: Wed, 19 Jul 2000 22:55:37 -0500
Reply-To: "John Gabriel" <[EMAIL PROTECTED]>
Brian Mc Kinney <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> E J wrote:
>
> > Kinky! Put on a helmet when your girlfriend gives you the boot :).
> > Your modem is probably a brainless (controllerless) wonder called a
> > winmodem..
> > External modem->almost always real modem
> > ISA modem->usually a real modem
> > PCI modem->usually a winmodem
> > Check the website of you modem manufacturer. If it contains lucent or
> > PCTel chips, you can download drivers to make those brainless wonders
work
> > under linux. The website is www.linmodems.org.
> > You might have to buy a new REAL modem.
> >
> > George Dowd wrote:
> >
> > > Hi, I am hoping someone can help me out. I have been trying to fix
this
> > > problem for the last 2 days and have checked the posts and
documentation
> > > as well, no luck. Any longer and my girlfriend is going to give me
the
> > > shoe.
> > >
> > > I installed Mandrake 7.1 on a new Gateway laptop. Everything is fine
> > > but there is no modem and I can't seem to find or configure it. On
boot
> > > up, I get the following message:
> > >
> > > insmod usb interface [failed]
> > >
> > > This is a dual boot system with win2k and the modem works fine there.
I
> > > am not sure if this usb problem is related to the modem?
> > >
> > > specs are:
> > > Intel 82371AB/EB PCI to USB Universal Host Controller
> > > PCI bus 0, device 7, function 2
> > > i/o range : FF80-FF9F
> > > Interrupt Req: 05
> > >
> > > Actiontec 56K v.90 modem
> > > PCI bus 0, device 11, function 0
> > > com3 port
> > >
> > > please please please if anyone has ideas I am desperate. thanks,
> > >
> > > George Dowd
> > > [EMAIL PROTECTED]
> > > 212 861-5598
>
> I notice that you are installing on a laptop.
> Most (? all) PCMCIA modems are "real" modems and should work with linux
> You may need to buy one of these if your laptops internal modem is a
winmodem
>
> Brian Mc Kinney
>
Dear Brian
I am just now finding my way around the world
of DSL and USB, so my words should be taken with a grain of salt. BUT..
It sounds as if the modem interface from your laptop is a
USB (Universal Serial Bus) port. This is a new interface
which is intended to replace the serial and parallel ports
we know and love. It is a "bus" that can have many devices
connected to it such as modems, scanners, keyboards,.....
and can carry data perhaps about 100 times faster than
a serial port.
I myself need a LINUX driver for a DSL (fast telephone line-between
270Kbit/sec and 8 megabit/sec) and am looking for information about how to
write one. As I
discover information I will try to keep you informed, if
I fail (I am 69 years old and absentminded, having been in computing since
1956) please send me E-mail to remind me.
Meanwhile there are "ordinary" modems that plug into
a PCMIA slot on a laptop. They are a bit expensive to
buy at retail, but I have been buying combination modem/
ethernet PCMIA cards at auction on www.ebay.com for
about $50.00 each. They usually sell for about $50.00
and are advertised as 33.6 kit/sec, upgradable to 56 Kbit
using a firmware download from the US Robotics/3COM
website.
I have also seen plain old 56K PCMIA cards on Ebay.
Last of all, I think there is web site www.linuxmodems.com
which may help you. I have still to login to it though, having
only begun this quest about 2 hours ago.
I just looked in my notes about auctions. My vendor for the
PCMCIA modems was [EMAIL PROTECTED], and their phone is 1-800-235-1046. A
chap called Jason Morgan
looked after the sale thru Ebay. I think perhaps that if you
call him and offer $50.00 + $5.00 shipping, he would sell you a PCMCIA card
after haggling a little.
Yours Sincerely
John Gabriel (VP R&D for MHGAI)
E-mail to [EMAIL PROTECTED] or [EMAIL PROTECTED]
------------------------------
From: MH <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x,comp.os.linux.misc
Subject: Re: capture screen output during booting
Date: Wed, 19 Jul 2000 20:59:10 -0700
Reply-To: [EMAIL PROTECTED]
WesRight wrote:
>
> mikey wrote:
> >
> >
> > Hi,
> >
> > Is there a way to capture screen output from the kernel during
> > booting ??
> >
> > Regards,
> > Mike
>
> Greetings Mike, others have given you some good advice as far as the
> dmesg command goes, but here is something you may want to try.
>
> When the bootup is complete and you're sitting at the login prompt, hit
> the right <shift> and <pg-up> or <pg-down>. It must be the right shift
> key and then page-up or down will scroll all the way back to the beginning
> of the boot process.
>
Now that's cool. There have been many times when that little trick
would have come in handy.
------------------------------
From: Narbey Derbekyan <[EMAIL PROTECTED]>
Subject: Re: gtk-config glib-config and configure
Date: Thu, 20 Jul 2000 04:11:00 GMT
I forgot I had made this posting. Well the problem was finally
fixed and here's hoping the explanation will help someone else.
It looks like the problem was related to the complete
non-existance of /usr/include/linux (and /usr/include/asm) on my
system. The configure script actually compiles a small C test program
with the libraries that the software depends on. It seems during my
kernel upgrade those two links under /usr/include went away and hence
causing problems. Things were compiling just fine and dandy before
then. A symlinking to the appropriate directories under
/usr/src/linux/include fixed it.
config.log had the appropriate clues as mentioned in the error
message below (imagine that :)
narbey
Narbey Derbekyan <[EMAIL PROTECTED]> wrote:
: Greetings, I'm having some issues with gtk-config and
: glib-config when used by the configure scripts when compile various
: gtk software. I'm receiving the following error:
: *** Could not run GLIB test program, checking why...
: *** The test program failed to compile or link. See the file config.log for the
: *** exact error that occured. This usually means GLIB was incorrectly installed
: *** or that you have moved GLIB since it was installed. In the latter case, you
: *** may want to edit the glib-config script: /usr/bin/glib-config
: Exact same error for gtk-config. This was working just fine. The same
: software that compiled just fine are now breaking when I run
: ./configure
: I have the following system:
: RedHat 6.2 upgraded with Helix Gnome.
: I had run into some lilo problems the other week and the only
: thing I could do was reinstall. Ever since then gtk-config and
: glib-config have been giving problems when I run ./configure. I did
: not know they existed until they broke.
: The scripts are there and I can run them from the command line
: and receive the various output meant to be used with gcc. But I
: receive the above error when I run ./configure as the same user.
: I installed Helix Gnome on top of all this hoping that the gtk
: and glib rpms from the Helix cd would clear things up. Well Helix
: Gnome is working just fine but I'm still running into the same
: problem. I check and the new gtk and glib libraries from the helix cd
: were installed.
: I can't make heads or tales of what's broken. If someone could
: give some pointers, I'll be eternally greatfull.
: thanks
: narbey
------------------------------
From: Jason <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux.networking,comp.os.linux.redhat,comp.os.linux.help,linux.redhat.install
Subject: Re: Simple question on Kernel Recompiling
Date: Thu, 20 Jul 2000 04:17:06 GMT
I think it may have solved one of the anomolies I was getting, namely the "every
third compile seems to work better than the others" phenomenon I was experiencing.
So here's my compile process now...
rm -rf the /lib/modules/KernelToCompileDirectory
make mrproper
make menuconfig (cuz I'm a textual junkie)
make dep
make bzlilo
make modules
make modules_install
cp bzImage to /boot/vmlinuz
lilo
shutdown -r now
Note the following: I have my stable kernel hidden as vmlinuz.working (not to mention
it's a 2.3.XX), and LILO refers to it as "safety", so I don't have to worry about
being destructive with either my sources, modules, or vmlinuz. Like I said, it takes
care of MOST of my problems (but the modules.dep file doesn't exist until the second
reboot). However, my problem remains. Here's what it does, followed by what I've
found out (since I've just recompiled 7 times from scratch many different ways):
What it does:
Bringing up interface eth0 insmod: /lib/modules/2.4.0-test4/net/smc-ultra.o: insmod
eth0 failed
Delaying eth1 initialization.
What's in the modules.dep file (just the one relevant line):
/lib/modules/2.4.0-test4/net/smc-ultra.o: /lib/modules/2.4.0-test4/net/8390.o
What I've found out:
When you have 2 eth cards, the kernel will only load 1 card, the other will be
forgotten. Also, invariably, the card you want as eth1 will load as eth0 from the
kernel. The only choice is to load both as modules. On top of that, you MUST
specify the drivers for each, and the IRQ for at least one, or at least one card
(usually both) will balk on boot.
PNP doesn't matter... I have a smc-ultra.o card and a tulip.o card. When both are
compiled in the kernel, the smc card is recognized as PNP, but only the tulip card is
loaded. It is also loaded as eth0, and I want it to be eth1.
So there's my problem, and more info than you probably needed. If I compile both
drivers into the kernel, one loads. As modules, none load (though under all
circumstances it detects the SMC card as a PNP device). I can also, under all
circumstances, simply reboot to my "safety" in lilo, and the devices work fine... so
I don't believe it's a config problem... unless there's something in the changes file
I didn't understand (read: skipped over) about dualing ethernet cards. Any help from
you or the rest of the Linux community would bring tears of joy to my eyes. Thanks
all.
A very red-eyed,
Jason
Roger Atkinson wrote:
> Not knowing (you didn't post them) what commands you have entered to
> make your new kernel makes it hard to help. So, I'll just try to explain
> what I do to see if this helps.
>
> After installing RedHat and the Source in /usr/src/linux (which is
> usually a link to the kernel version you are building) I decide whether
> I want to replace the existing kernel or make a new test kernel. I
> usually do the later and after testing it satisfactorily I set it up to
> be the primary boot kernel later. If I am making a new test kernel I
> edit the .Makefile and add something like ".1" or "-1" which will make a
> kernel that will install next to the kernel I currently boot. I then
> execute the following as root from the /usr/src/linux dir:
>
> 1) make mrproper <---gets rid of everything not needed including the
> old config file.
> 2) cd into the configs dir and cp the base config file for my machine
> into /usr/src/linux/.config (cd back into /usr/src/linux when done).
> 3) make xconfig <---since the config file has all the default Red Hat
> config all I need to do is modify the configuration for my SCSI and
> ethernet which I usually put into the kernel instead of making them
> modules. NOTE - I do not have a dual eth machine so I don't know if you
> should build in two different eth drivers but i don't see why not.
>
> After you close and save off the new config file, you must:
>
> 4) make dep <---make the dependencies.
> 5) make clean <--- I'm not sure if this is required but it always works
> for me.
> 6) make zImage or bzImage <---usually a bzImage as the default Red Hat
> config makes a large kernel.
> 7) make modules <------ must be run since the default Red Hat config
> contains lots of modules which I use like the sound driver, zip ppa
> driver, etc.
> 8) make modules_install <----- here is where it gets interesting.
> Notice above I modified the .Makefile. This will cause a new directory
> and modules set to be put into /lib/modules/<kernel version>. If you
> don't change the Version number you are making (in the .Makefile) then
> you *must* move the old modules out of the way before you do the "make
> modules_install" or it will look as if it is doing what you want but it
> will, in fact, not install anything. This is the default behaviour and
> the probable cause of all your module problems. In other words your
> modules do not match your kernel and they must.
> 9) If everything builds OK IE: good exit code, I then edit the
> /etc/lilo.conf file to have an entry for the new kernel I am going to
> install. This is usually vmlinuz-<version> for me. YMMV. To keep it easy
> to remember I usually make sure that the .Makefile is one version more
> than what is currently being booted in /etc/lilo.conf. This makes sure
> I don't make typo mistakes.
> 10) make install <--- this will scan through the last build directories
> verifying that everything got made, install the new kernel into the
> /boot dir and run /sbin/lilo for you making sure that everything is
> ready for the next reboot.
> 11) reboot and at the lilo prompt, enter the tag name (what you labeled
> your new kernel in /etc/lilo.conf) of the new kernel. I usually use
> linux_new until I have tested that all is well and then later install it
> as the default Linux kernel to boot.
>
> I hope this helps you and others as I have seen a lot of posts about
> kernel building and I want to try to help if I can. BTW Every thing I
> have listed above is mentioned in either HowTos or the kernel readme
> that Linus keeps up to date and it should all be required reading before
> anyone starts to build there own new kernel.
>
> Ciao, Roger A.
>
> Jason wrote:
> >
> > Hello...
> >
> > I tried that, and unfortunately it did the same thing. I moved the modules.dep
> > to oldmodules.olddep so it couldn't find it, and "touch"ed a new modules.dep
> > file. I got the same number of the following error that I got even before I
> > wiped out the modules.dep file (this is guessing, because I don't know how to
> > copy the bootup errors to a file yet):
> >
> > depmod: *** Unresolved symbols in /path/goes/here/somemodulename.o
> >
> > ...and I get that for tulip.o, and a couple other net drivers. Any ideas?
> >
> > --Jason
> >
> --snip rest of article----
------------------------------
From: "Dj" <[EMAIL PROTECTED]>
Subject: @Home Cable Modem Setup Problems (2)
Date: Thu, 20 Jul 2000 04:18:30 GMT
Thanks for the help. Here's my /etc/resolv.conf:
search rchdsn1.tx.home.com
nameserver 24.4.48.33
nameserver 24.4.48.34
domain rchdsn1.tx.home.com
The problem is that I can't ping IP addresses (even without DNS resolution).
Eth0 is definitely working -- I can see traffic from my pings, etc. but
nothing gets to my screen. Any ideas guys ? Are you aware of any other
diagnostic tricks I can do to trouble shoot deeper ?
I would appreciate any help I can get....
"Bit Twister" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Sounds like /etc/resolv.conf is not set correctly. Fake example:
>
> search localdomain grlnd1.tx.home.com
> nameserver 24.14.82.433
> nameserver 24.14.28.434
>
>
> On Thu, 20 Jul 2000 00:32:36 GMT, Dj <[EMAIL PROTECTED]> wrote:
> >Hi there:
> >
> >Here's what I have: redhat 6.2 and 3COM 509(b) card. IP assigned by DHCP.
> >I followed the instructions (used netcfg) to configure eth0, hostnames,
DNS,
> >etc.
> >After I run dhcpcd and ifconfig I can see eth0 interface with assigned IP
> >address. I can even see the number of RX/TX packets. I also can see
traffic
> >going back/forth on gnome network manager.
> >The weird thing is that I cannot access anything outside -- I can't even
> >resolve the names.
> >
>
>
> --
> The warranty and liability expired as you read the message.
> If the above breaks your system, it's yours and you keep both pieces.
> Practice safe computing. Backup the file before you change it.
> Do a, man every_command_here, before doing anything or running a script.
------------------------------
** 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.setup) 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-Setup Digest
******************************
