Linux-Setup Digest #415, Volume #20 Sat, 13 Jan 01 19:13:07 EST
Contents:
Re: Help ! Router firewall with only one NIC ("nag")
Re: Getting modem to be accepted by YAST2 config (drumvudu)
Re: Mouse stuck at the top of the screen after kde opens. (drumvudu)
Re: How to install linux using pcmcia cd-rom into notebook (Dave)
Re: User Permissons and Internet connection (David Efflandt)
Re: xfstt problem in SuSE linux (2.4.0 kernel) ("John D Lamb")
Re: IDE CD-RW vs ATAPI Zip drive (E J)
SBLive 1024 PCI (Nicolas Rinaudo)
Re: xfstt problem in SuSE linux (2.4.0 kernel) (Brian Gooyear)
Re: User Permissons and Internet connection (tbk)
Re: User Permissons and Internet connection (David)
During boot: can't open dependencies file ?!?!? (Edwin)
Re: Help ! Router firewall with only one NIC (James Knott)
Re: Help ! Router firewall with only one NIC (James Knott)
----------------------------------------------------------------------------
From: "nag" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,comp.os.linux.security
Subject: Re: Help ! Router firewall with only one NIC
Date: Sat, 13 Jan 2001 21:30:41 +0100
there is probably a broadcast taking place initially to find the modem but
after it should be point to point between the linux gateway box and the
modem.
anyone who knows for sure how it does work ??
so well, I can connect via pppoe to the internet via my isp [in this
configuration].
the problem is routing, masquerading & firewalling
what should be the routing table and the ipchains configuration in that case
?
if this is possible ...
thanks in advance,
nag
"Howard West" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> perhaps you know more about this than I but aren't you effectively
broadcasting
> all the pppoe traffic to your whole network?
>
> nag wrote:
>
> > why isn't it possible ?
> > I can probably receive on eth0 (pppoe) and send on eth0:0 (private
network)
> > and do the translation/routing between
> > those two interfaces. just some more collisions during intensive data
> > transfer (is it right ?) because the media is the same.
> >
> > as for the price of the nic cards, this is more of a curiosity problem
than
> > a cost problem.
> > but you I will probably switch to a simpler configuration if I have no
> > positive answer
> >
> > "default" <[EMAIL PROTECTED]> wrote in message
> > news:[EMAIL PROTECTED]...
> > > I don't this will work at all. For one thing, you would need two
> > interfaces
> > > on the gateway/firewall box, one to recieve and send on the "outside"
> > network
> > > and one to send and recieve on the "inside" network. With only one
card
> > you
> > > are either inside or outside, not both. The interface can't, to the
best
> > of
> > > my knowledge, perform a sort of context switch on each packet. You
would
> > > either be inside, in which case you are cut off from the external
network,
> > or
> > > outside, in which case you would be inaccessible to the internal
hosts.
> > >
> > > The other problem with this is that you are dumping raw pppoe traffic
> > right
> > > into your local network wire, which isn't really capable of handling
it.
> > You
> > > need to plug it into a host running pppoe so that it can translate the
> > packets
> > > into something that can be routed internally to your local hosts.
> > >
> > > How expensive could it be to get one more NIC?
> > >
> > > nag wrote:
> > >
> > > > hi,
> > > > I am trying to configure a linux box for masquerading/firewalling.
> > > > the configuration is not the classical one and I would need some
> > advises.
> > > > now I could just buy a 2nd nic card, but I am curious and I would
like
> > to
> > > > make it work that way.
> > > >
> > > > - adsl modem is connected to the hub via uplink
> > > > - the linux box (myoldbox) is connected to the hub via eth0
> > > > (debian stable 2.2.18pre21 running pppoe 2.6)
> > > > + the connection to my isp goes through the ppp interface ppp0
> > > > + eth0 is the only ethernet interface (1 nic)
> > > > - other machines are connected to the hub (1 nic per machine)
> > > >
> > > > q1? is this configuration as safe, efficient, reliable as having
the
> > > > gateway 'physically' routing to the private network (with 2 nics) ?
> > > >
> > > > and that is my current problem
> > > > q2? in this configuration, how do I configure the routes,
translation
> > and
> > > > ipchains ?
> > > > the private network mask is 192.168.0.0
> > > > loopback interface is up
> > > > eth0 on myoldbox is up with address 192.168.0.1
> > > > eth0:0 on myoldbox is up with address 192.168.0.2
> > > > ( I guess this alias interface is needed in that
case
> > for
> > > > routing purpose, not sure ? )
> > > >
> > > > thanks for your help, any pointer appreciated,
> > > > nag
> > >
>
------------------------------
Crossposted-To: alt.os.linux
From: drumvudu <[EMAIL PROTECTED]>
Subject: Re: Getting modem to be accepted by YAST2 config
Date: Sat, 13 Jan 2001 21:12:31 GMT
Patricia, go into yast1, system administration, network
configuration,configure a ppp network. Fill in the blanks for your service
and select auto detect modem, if it can be auto detected, that should do
the trick. I run a us robotics on suse 7,0 out of the box with no
problemo. julian
On Fri, 12 Jan 2001, Patricia Flickner wrote:
> I've been having quite a problem getting my US Robotics internal modem
> to be accepted and configured properly on Suse linux. I've followed
> the directions online and in the manual (and interestingly enough, the
> manual shows to use ttyS5, but in another part, it says the modem is
> assigned to ttyS1, but then the "help" I receive says it's ttyS2, none
> of which work). I get the commands to work just fine, but when I go
> back to YAST2 to finish the installation, it says to turn on my modem,
> which of course, I can't really do except that I did the command and
> hit the ok button. No matter what I've tried, it doesn't detect a
> modem, yet, when I installed the machine, it detected the hardware with
> no difficulties whatsoever. The video card was another matter until I
> bought a Voodoo 3x, then it went along just fine.
>
> Does anyone know what's going on? I've had it with Suse's lack-of-tech
> support and customer lack-of-service. They are the worst company I've
> ever dealt with next to Microsoft. And a 90-day support policy just
> means that they can take even longer to respond to your problems.
>
> Regards,
>
> Pat Flickner
>
>
>
>
> Sent via Deja.com
> http://www.deja.com/
>
------------------------------
From: drumvudu <[EMAIL PROTECTED]>
Subject: Re: Mouse stuck at the top of the screen after kde opens.
Date: Sat, 13 Jan 2001 21:23:30 GMT
yast1, system administration,integrate hardware into system,mouse
configuration, have fun
On Sat, 13 Jan 2001, stephencolin wrote:
> Hi I'm a novice at Linux but eager to get it up and running.
> However, I have had numerous problems trying to configure X and after
> trying numerous combinations I managed to get a good resolution and
> everything when KDE opened (after fiddling with it for days!). But the
> cursor remains 'stuck' at the top of the screen and jumps madly about
> when I move the mouse, but it won't move anywhere but along the top of
> the screen.
> I am using SuSE 7.0 with a GEForce2 MX card on an Athlon 850. With XFree
> version 4.0.2.
> Any suggestions gang??
> Please...................
>
> Stephen, Melbourne Australia.
>
>
------------------------------
From: Dave <[EMAIL PROTECTED]>
Subject: Re: How to install linux using pcmcia cd-rom into notebook
Date: Sat, 13 Jan 2001 21:30:07 -0000
Heesang wrote:
>
>
> I am trying to install linux using pcmcia cd-rom into my notebook in
> vain. I first made two disketts(one is boot.img and for pcmcia.img).
> However when I try to install , the boot.img doesn't detect the pcmica
> cd-rom. Of course the pcmica cd-rom works fine now(I installed linux with
> Backpack external cd-rom before). I don't know why I can't install with
> the pc card cd-rom . And to my knowledge, all of ATAPI cd-rom are
> supported by linux. How can I do this ? Any idea?
> My pcmcia cd-rom is Doshiba ATAPI 20x.
> Have a good one.
> Heesang
> .
>
>
Sounds like an issue I had once, installing Red Hat 7 on an old Dell
Latitude LM laptop, that had an interchangeable CD-Rom drive and floppy
drive. Going into BIOS setup with this laptop I could see that the only
choices for a boot device were the HDD and FDD; there was no BIOS support
for the CDROM. Sounds like that is the case with your laptop.
I got around this by creating a very small MS DOS partition (about 12 mgb),
installing a DOS-based CDROM driver (found it on the net off the CDROM
manufactures's site), and installing all the AUTOBOOT code and folders off
the Linux distro CDROM.
Next pulled the floppy drive, inserted the CDROM (with the Linux distro on
it), booted the laptop to MSDOS, and ran AUTOBOOT. This loads a LINUX image
from the MS/DOS autoboot folder then initiates the LINUX installation
program on the CDROM.
I ended up with DOS (tiny; 12 mgb), Linux SWAP, and two LINUX native
partitions on the HDD when I was finished.
Kind of circuitous, but it worked, and I hope this helps you. Let me know.
DaveG.
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: [EMAIL PROTECTED] (David Efflandt)
Subject: Re: User Permissons and Internet connection
Date: Sat, 13 Jan 2001 22:04:33 +0000 (UTC)
Reply-To: [EMAIL PROTECTED]
On Sat, 13 Jan 2001 18:27:09 GMT, Trevor Tutt <[EMAIL PROTECTED]> wrote:
>I have finally found a real modem that will allow me to connect from
>Linux 6.1.
>The problem that I now face is that I can only connect as Root. If I
>try to connect from any other login, the connect options in Gnome are
>disabled. I am able to use the Dialup Configuration Tool and Debug the
>connection and it looks as though everything is in order, but it will
>not actually connect.
>
>I have checked the login in question to see if it has rx rights (at
>someone else's suggestion) and it does.
>I have a strong feeling that it does involve user rights, but I don't
>know where to look.
What I did in RH 6.1 was set up the connection with the Gnome Dialer
Configuration Tool as root, then edited the connection to tell it to let
pppd handle authentication (otherwise wvdial gets in the way trying to do
a shell login). Then I used linuxconf to set up ppp0 using PAP and
checked the box to allow any user to start/stop it.
Although, later on when I was removing and adding ISPs I think it became
confused about pap-secrets, so I had to make changes to that manually.
However, I boot to the console and use startx. I have never booted in GUI
mode (runlevel 5) so I do not know if that works differently.
In Mandrake I am using Kppp which I think by default allows any user to
set up a dialup connection. I also used linuxconf go set up connections
to start with /sbin/ifup. But since I upgraded my kernel, that does not
seem to detach from the terminal until the connection comes down (which it
does if I hit Ctrl-C).
--
David Efflandt [EMAIL PROTECTED] http://www.de-srv.com/
http://www.autox.chicago.il.us/ http://www.berniesfloral.net/
http://cgi-help.virtualave.net/ http://hammer.prohosting.com/~cgi-wiz/
------------------------------
From: "John D Lamb" <[EMAIL PROTECTED]>
Subject: Re: xfstt problem in SuSE linux (2.4.0 kernel)
Date: Sat, 13 Jan 2001 22:28:39 -0000
No. It's worse than that. xfstt is running without problems, but something
goes wrong when I try t restart the X server. Probably it's time for me to
upgrade XFree86 or switch to xfsft.
"John D Lamb" <[EMAIL PROTECTED]> wrote in message
news:93pl3f$4ss$[EMAIL PROTECTED]...
> I've compiled a 2.4.0 kernel for SuSE 6.4 on an i686- no problems there.
The
> machine boots to runlevel 3 (equivalent to RH runlevel 5) without any
> problems. And I get to log in and use KDE. Everything works (printers,
> disks, mouse, keyboard, graphics card, monitor, wacom tablet). The
problems
> start when I log out. Then the screen flashes a few times and leaves me
with
> a terminal login prompt.
>
> I think I've traced this one to xfstt. The /sbin/init.d/xfstt script start
> section currently contains
>
> echo "Starting xfstt"
> /usr/X11R6/bin xfstt --sync
> startproc /usr/X11R6/bin xfstt --daemon || return=$rc_failed
> echo -e "$return"
>
> If I try
> # init 2
> followed by
> # init 3
> I get the same problem as before. A quick check on the init scripts
reveals
> why:
> # init 2
> # /sbin/init.d/xfstt start
> returns "failed". But
> # init 2
> # /usr/X11R6/bin xfstt --sync
> # startproc /usr/X11R6/bin xfstt --daemon
> # /sbin/init.d/xdm start
> restarts kdm successfully
>
> Any suggestions for how to get /sbin/init.d/xfstt to work? I've tried
> putting a sleep command before the startproc in the hope that would give
> xfstt --sync time to finish, but that didn't work. Could I try to get the
> xfstt PID with checkproc and wait for it to complete before executing
> startproc? If so, what would the code look like? Or is there a simpler
way.
>
> JDL
>
>
------------------------------
From: E J <[EMAIL PROTECTED]>
Subject: Re: IDE CD-RW vs ATAPI Zip drive
Date: Sat, 13 Jan 2001 22:44:00 GMT
Give it up :) Put in the scsi emulation back in.
Put the ignore statement to ignore both your zip disk (hdb) and cd-rw drive
(hd?).
You should be able to access your Zip DOS disk with the following command.
# mount /dev/sda4 /mnt/zip -t vfat # just guess that your zip drive is at
sda
John Foster wrote:
> After much help from this group and from the HOW-TOs and other
> documentation, I have been able to get my IDE CD-RW drive working under
> SCSI emulation - but one problem remains. I also have an ATAPI Zip-100
> drive, and I cannot find a setup that will get both working at once,
> except by intervention as root after login. I am running Redhat 6.2.
>
> The problem is that if I follow the HOW-TO instructions, the CD-RW works
> fine but the Zip drive doesn't. The instructions include adding
> /sbin/modprobe ide-scsi
> to /etc/rc.d/rc.local; afterwards cdrecord -scanbus then reports both
> the CD-RW and the Zip drive, and the Zip is not accessible. If I remove
> that line, the CD-RW doesn't work but the Zip does.
>
> What _does_ work is to remove the modprobe line, then log in as root and
> mount a Zip disc, and *then* type /sbin/modprobe ide-scsi.
>
> Support for the Zip drive is built in to the kernel, and I assume that
> what's happening is that the kernel doesn't grab ownership of the Zip
> drive until the first mount, leaving it free for ide-scsi to grab if
> that's run first. The Zip is on hdb, so I tried adding 'options ide-scsi
> ignore=hdb' in /etc/conf.modules, but it didn't like that :-(
>
> Does anyone know how I can either tell ide-scsi not to grab the Zip
> drive, or tell the kernel that it should grab it during the boot
> process?
>
> With thanks in advance
>
> John
------------------------------
From: Nicolas Rinaudo <[EMAIL PROTECTED]>
Subject: SBLive 1024 PCI
Date: Sun, 14 Jan 2001 00:13:21 +0100
Does any of you folks know how to set up an SBLive 1024 PCI under
slackware? It was pretty easy to do under redhat where emu10k1 was part
of the kernel, but I don't even know where to start looking with
slackware.
Thanx a lot,
Nicolas Rinaudo
------------------------------
From: Brian Gooyear <[EMAIL PROTECTED]>
Subject: Re: xfstt problem in SuSE linux (2.4.0 kernel)
Date: Sat, 13 Jan 2001 18:24:32 -0500
John D Lamb wrote:
> No. It's worse than that. xfstt is running without problems, but something
> goes wrong when I try t restart the X server. Probably it's time for me to
> upgrade XFree86 or switch to xfsft.
>
>
> "John D Lamb" <[EMAIL PROTECTED]> wrote in message
> news:93pl3f$4ss$[EMAIL PROTECTED]...
> > I've compiled a 2.4.0 kernel for SuSE 6.4 on an i686- no problems there.
> The
> > machine boots to runlevel 3 (equivalent to RH runlevel 5) without any
> > problems. And I get to log in and use KDE. Everything works (printers,
> > disks, mouse, keyboard, graphics card, monitor, wacom tablet). The
> problems
> > start when I log out. Then the screen flashes a few times and leaves me
> with
> > a terminal login prompt.
> >
> > I think I've traced this one to xfstt. The /sbin/init.d/xfstt script
> > start section currently contains
> >
> > echo "Starting xfstt"
> > /usr/X11R6/bin xfstt --sync
> > startproc /usr/X11R6/bin xfstt --daemon || return=$rc_failed
> > echo -e "$return"
> >
> > If I try
> > # init 2
> > followed by
> > # init 3
> > I get the same problem as before. A quick check on the init scripts
> reveals
> > why:
> > # init 2
> > # /sbin/init.d/xfstt start
> > returns "failed". But
> > # init 2
> > # /usr/X11R6/bin xfstt --sync
> > # startproc /usr/X11R6/bin xfstt --daemon
> > # /sbin/init.d/xdm start
> > restarts kdm successfully
> >
> > Any suggestions for how to get /sbin/init.d/xfstt to work? I've tried
> > putting a sleep command before the startproc in the hope that would give
> > xfstt --sync time to finish, but that didn't work. Could I try to get
> > the xfstt PID with checkproc and wait for it to complete before
> > executing startproc? If so, what would the code look like? Or is there a
> > simpler
> way.
> >
> > JDL
> >
> >
>
>
Don't know if this will help but my xfstt stuff also points to a directory:
/usr/X11R6/bin/xfstt --dir /usr/X11R6/lib/X11/fonts/TrueType/ --sync
/usr/X11R6/bin/xfstt --daemon --dir /usr/X11R6/lib/X11/fonts/TrueType/
Maybe it's looking for something to sync and is freaking out cause it
doesn't know where to look.
Brian
------------------------------
From: tbk <[EMAIL PROTECTED]>
Subject: Re: User Permissons and Internet connection
Date: Sat, 13 Jan 2001 23:26:47 GMT
Thanks for the help!
I can now read news and browse the net with relative "safety"
------------------------------
From: David <[EMAIL PROTECTED]>
Subject: Re: User Permissons and Internet connection
Date: Sat, 13 Jan 2001 23:37:39 GMT
David Efflandt wrote:
>
> On Sat, 13 Jan 2001 18:27:09 GMT, Trevor Tutt <[EMAIL PROTECTED]> wrote:
> >I have finally found a real modem that will allow me to connect from
> >Linux 6.1.
> >The problem that I now face is that I can only connect as Root. If I
> >try to connect from any other login, the connect options in Gnome are
> >disabled. I am able to use the Dialup Configuration Tool and Debug the
> >connection and it looks as though everything is in order, but it will
> >not actually connect.
> >
> >I have checked the login in question to see if it has rx rights (at
> >someone else's suggestion) and it does.
> >I have a strong feeling that it does involve user rights, but I don't
> >know where to look.
>
> Then I used linuxconf
linuxconf? What's that?
Oh That's right, I remember now, never mind, I exterminated that bug
after installation.
--
Confucius say: He who play in root, eventually kill tree.
Registered with the Linux Counter. http://counter.li.org
ID # 123538
Completed more W/U's than 99.002% of seti users. +/- 0.01%
------------------------------
Crossposted-To: comp.os.linux,comp.os.linux.misc,nl.comp.os.linux.installatie
Subject: During boot: can't open dependencies file ?!?!?
From: [EMAIL PROTECTED] (Edwin)
Date: 13 Jan 2001 23:58:33 GMT
Hi,
I've downloaded and compiled kernel 2.2.12 on my RedHat 7 installation. Went
through all the moves:
make dep,
make clean,
make bzImage,
make modules,
delete old modules directory,
make modules_install,
copied System.map to /boot/System.map-2.2.12
copied bzImage to /boot/vmlinuz-2.2.12
(LILO was already configured for this kernel name)
Then, when I boot up the new kernel, I get this:
...
Jan 14 00:35:50 localhost modprobe: modprobe: Can't open dependencies file
/lib/modules/2.2.12/modules.dep (No such file or directory)
Jan 14 00:35:50 localhost rc.sysinit: Initializing USB controller (usb-uhci):
failed
Jan 14 00:35:50 localhost fsck: /: clean, 106161/384768 files, 373048/769104
blocks
Jan 14 00:35:50 localhost rc.sysinit: Checking root filesystem succeeded
Jan 14 00:35:50 localhost rc.sysinit: Remounting root filesystem in read-write
mode: succeeded
Jan 14 00:35:51 localhost depmod: depmod:
Jan 14 00:35:51 localhost depmod: *** Unresolved symbols in
/lib/modules/2.2.12/misc/eicon.o
Jan 14 00:35:51 localhost rc.sysinit: Finding module dependencies: succeeded
Jan 14 00:35:51 localhost modprobe: modprobe:
Jan 14 00:35:51 localhost modprobe: Can't locate module es1371
Jan 14 00:35:51 localhost rc.sysinit: Loading sound module (es1371): failed
Jan 14 00:35:51 localhost fsck: /boot: clean, 29/6024 files, 4012/24066 blocks
Jan 14 00:35:51 localhost fsck: /home: clean, 704/384768 files, 13483/769104
blocks
Jan 14 00:35:51 localhost rc.sysinit: Checking filesystems succeeded
...
However, modules.dep is there in /lib/modules/2.2.12:
total 44
drwxr-xr-x 8 root root 4096 Jan 14 00:35 .
drwxr-xr-x 5 root root 4096 Jan 14 00:12 ..
drwxr-xr-x 2 root root 4096 Jan 14 00:12 block
drwxr-xr-x 2 root root 4096 Jan 14 00:12 fs
drwxr-xr-x 2 root root 4096 Jan 14 00:12 ipv4
drwxr-xr-x 2 root root 4096 Jan 14 00:12 misc
-rw-r--r-- 1 root root 4724 Jan 14 00:35 modules.dep
-rw-r--r-- 1 root root 99 Jan 14 00:35 modules.pcimap
drwxr-xr-x 2 root root 4096 Jan 14 00:12 net
drwxr-xr-x 2 root root 4096 Jan 14 00:12 video
What's wrong here? Shouldn't this work? Have I forgotten something?
Regards,
Edwin
------------------------------
From: James Knott <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,comp.os.linux.security
Subject: Re: Help ! Router firewall with only one NIC
Date: Sun, 14 Jan 2001 00:06:15 GMT
You can assign an alias to an interface, as the user has
done. However, one address should be internal and the other
external.
default wrote:
>
> I don't this will work at all. For one thing, you would need two interfaces
> on the gateway/firewall box, one to recieve and send on the "outside" network
> and one to send and recieve on the "inside" network. With only one card you
> are either inside or outside, not both. The interface can't, to the best of
> my knowledge, perform a sort of context switch on each packet. You would
> either be inside, in which case you are cut off from the external network, or
> outside, in which case you would be inaccessible to the internal hosts.
>
> The other problem with this is that you are dumping raw pppoe traffic right
> into your local network wire, which isn't really capable of handling it. You
> need to plug it into a host running pppoe so that it can translate the packets
> into something that can be routed internally to your local hosts.
>
> How expensive could it be to get one more NIC?
>
> nag wrote:
>
> > hi,
> > I am trying to configure a linux box for masquerading/firewalling.
> > the configuration is not the classical one and I would need some advises.
> > now I could just buy a 2nd nic card, but I am curious and I would like to
> > make it work that way.
> >
> > - adsl modem is connected to the hub via uplink
> > - the linux box (myoldbox) is connected to the hub via eth0
> > (debian stable 2.2.18pre21 running pppoe 2.6)
> > + the connection to my isp goes through the ppp interface ppp0
> > + eth0 is the only ethernet interface (1 nic)
> > - other machines are connected to the hub (1 nic per machine)
> >
> > q1? is this configuration as safe, efficient, reliable as having the
> > gateway 'physically' routing to the private network (with 2 nics) ?
> >
> > and that is my current problem
> > q2? in this configuration, how do I configure the routes, translation and
> > ipchains ?
> > the private network mask is 192.168.0.0
> > loopback interface is up
> > eth0 on myoldbox is up with address 192.168.0.1
> > eth0:0 on myoldbox is up with address 192.168.0.2
> > ( I guess this alias interface is needed in that case for
> > routing purpose, not sure ? )
> >
> > thanks for your help, any pointer appreciated,
> > nag
--
Replies sent via e-mail to this address will be promptly
ignored.
To reply, replace everything to the left of "@" with
"james.knott".
------------------------------
From: James Knott <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,comp.os.linux.security
Subject: Re: Help ! Router firewall with only one NIC
Date: Sun, 14 Jan 2001 00:04:58 GMT
nag wrote:
>
> hi,
> I am trying to configure a linux box for masquerading/firewalling.
> the configuration is not the classical one and I would need some advises.
> now I could just buy a 2nd nic card, but I am curious and I would like to
> make it work that way.
>
> - adsl modem is connected to the hub via uplink
> - the linux box (myoldbox) is connected to the hub via eth0
> (debian stable 2.2.18pre21 running pppoe 2.6)
> + the connection to my isp goes through the ppp interface ppp0
> + eth0 is the only ethernet interface (1 nic)
> - other machines are connected to the hub (1 nic per machine)
>
> q1? is this configuration as safe, efficient, reliable as having the
> gateway 'physically' routing to the private network (with 2 nics) ?
>
> and that is my current problem
> q2? in this configuration, how do I configure the routes, translation and
> ipchains ?
> the private network mask is 192.168.0.0
> loopback interface is up
> eth0 on myoldbox is up with address 192.168.0.1
> eth0:0 on myoldbox is up with address 192.168.0.2
> ( I guess this alias interface is needed in that case for
> routing purpose, not sure ? )
You'll need to set one of those to the assigned IP address
or DHCP as required. While it's not as safe as two separate
NICs, it shouldn't be too bad, as 192.168 addresses are not
routable over the internet. I'm doing similar with my OS/2
system.
--
Replies sent via e-mail to this address will be promptly
ignored.
To reply, replace everything to the left of "@" with
"james.knott".
------------------------------
** 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 by posting to comp.os.linux.setup.
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
******************************
