Linux-Networking Digest #299, Volume #11 Wed, 26 May 99 23:13:44 EDT
Contents:
wu-ftpd Cracked? ("Rick Gocher")
Re: token ring driver - performance questions? (Owen Brotherwood)
Can WU imapd support Qmail's maildir? ([EMAIL PROTECTED])
Re: Is Linux Slower than Windows??(ppp) (Ray)
Re: LCP: timeout sending Config-Requests PPP FAILURE! LARGE MSG (Clifford Kite)
Re: IP Masquerading ("Curt")
[EMAIL PROTECTED] (Lorin Hochstein)
New knfsd trouble (was Re: NFS with Redhat 6 server and clients) ("G. Hugh SONG")
Re: IP Masquerade/Routing ("Curt")
Re: Looking for good reliable 100Mb Ethernet card for RH 6.0 (bryan)
Re: Can't see Samba Server from Win95 ("Fernando")
Re: Pinging for behind IP Masq?? (Stefan)
Looking to find ipmasqadm code ("John Antypas")
contest ("Ron Hill")
Re: Linux: ICMP Redirect, IP Source Routing unterdruecken (Helge Oldach)
Re: RH6 & PPPD 2.3.7 weirdness (Clifford Kite)
Using Linux as a Router ("DB")
Re: Recent problems with my cablemodem (using DHCP) ([EMAIL PROTECTED])
----------------------------------------------------------------------------
From: "Rick Gocher" <[EMAIL PROTECTED]>
Subject: wu-ftpd Cracked?
Date: Wed, 26 May 1999 21:02:26 GMT
I was told not to run the wu-ftpd that came with my RH box because there are
some exploits that will allow people to gain root access on my server. I
have found several older exploits however can't find anything very recent.
Does anyone know if this is a problem and if so, what ftp server should I
run on my RH5.2 box.
TIA,
Rick
------------------------------
From: Owen Brotherwood <[EMAIL PROTECTED]>
Subject: Re: token ring driver - performance questions?
Date: Thu, 27 May 1999 01:03:04 +0200
I too have been disapointed with my own setup: three identical Squid Proxy
machines (plus a smaller test machine), Mega whatsit's of HD/RAM/whatever
but with IBM ISA TR cards .
I also have had problems with IBM PCMCIA TR cards (I'm ashamed to say I
basically gave up ...)
- then Madge released their drivers on their web site.
I started with my PCMCIA Madge TR card: installed and working on my RH6.0
with none of the PCMCIA parameter corrections to high memory and I don't
have to worry about killing my system if my cable gets pulled out.
And also supports collection of all traffic, so my analyzer wish is
fulfilled
Then I bought a Madge PCI Mrk II TR card for a stationary machine RH6.0
because of analyzer possibilities and the prospect of PCI which could mean
better performance: now using ntop for statistics, yeah.
I've now received three more Madge cards: one of which I have installed in
_one_ of thee _identical_ squid proxy machines.
Having read a follow up message to this post, I did the following to see if
my last wish for performance could be granted:
>From a Madge PCI (RH6.0) to
IBM ISA (RH5.2) with RH 5.2 TR Driver from RH5.2 CD (Boxed set)
150 Opening BINARY mode data connection for Wine-990523.tar.gz.
226 Transfer complete.
3526153 bytes sent in 7.06 secs (4.9e+02 Kbytes/sec)
Madge PCI (RH5.2) with Madge Driver from http://www.madge.com
150 Opening BINARY mode data connection for Wine-990523.tar.gz.
226 Transfer complete.
3526153 bytes sent in 2.73 secs (1.3e+03 Kbytes/sec)
The above is a late night quick test (I ISDN'ed the three machines from
home) and not a proper benchmark (netperf against all the machine?)
Please also take time to visit http://www.linuxtr.com, they seem to have
more info and newer drivers to the IBM TR cards and are trying to fix and
improve the support.
As to if the above can help in removing your bottleneck, I don't know: but
at least it made me measure a difference between my machines, so I can go
to sleep knowing I didn't waste my money so who could wish for more?
[EMAIL PROTECTED] wrote:
> I'm using linux in a token-ring environment at work and am setting up a
> caching web proxy using Squid on the machine.
> It doesn't appear that the standard token ring driver is sharable, so I
> can't do any load balancing using more than one card which is a pity.
>
> Has anyone used this driver in anger though (the standard token ring one
> that ships with all kernels; I'm using 2.0.36) as I'm concerned about
> the performance of the driver itself. It does seem robust enough, but
> Squid seems suprisingly slow even given the less-than-ideal hardware
> I'm running this on (P166 PC, 2 x 2GB IDE disks with web cache on its
> own disk, 64MB memory) and so I'm leaning towards the token ring driver
> being the bottleneck at the moment...
------------------------------
From: [EMAIL PROTECTED]
Subject: Can WU imapd support Qmail's maildir?
Date: Wed, 26 May 1999 20:13:37 GMT
Hey,
Does anyone know if there is a way to make Washington University's
imapd support the maildir in Qmail? I can't seem to find any info on
the subject matter.
- Steve
--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---
------------------------------
From: [EMAIL PROTECTED] (Ray)
Subject: Re: Is Linux Slower than Windows??(ppp)
Date: 27 May 1999 01:38:30 GMT
On 26 May 1999 10:19:12 -0400, Paul D. Smith <[EMAIL PROTECTED]> wrote:
>%% [EMAIL PROTECTED] (Jonathan Luckey) writes:
>
> jl> Linux can do this too using the irqtune program. Use it to rotate
> jl> IRQ priorities to put your serial port first usually IRQ 3 or IRQ
> jl> 4, and see if that tightens up your ping times.
>
>Interesting.
>
>Question 1: is the need for this very dependent on CPU speed? I have a
>P450 with 128M. I'm usually seeing OK modem performance (at least given
>that I only get about 26400 connections with my 56k modem :-/) but I
>wonder if raising the IRQ priority would give even better ones?
IRQ Tune isn't going to affect your connect speed at all. The 26400 you see
is determined by the two modems. About the only things you can do (from
software anyway) are:
1. Make sure your port speed is reasonable (57600 or 115200 for most
people)
2. Make sure you are not passing the modem any unnecessary init. strings
unless you know what they do. Virtually all modern modems come with
reasonable defaults so don't mess with them unless you have a good reason.
FWIW even a 486dx2/50 is fast enough to handle 50k+ connections in most
cases without resorting to IRQ Tune.
--
Ray
------------------------------
From: kite@NoSpam.%inetport.com (Clifford Kite)
Subject: Re: LCP: timeout sending Config-Requests PPP FAILURE! LARGE MSG
Date: 26 May 1999 20:41:25 -0500
Brandon Edens ([EMAIL PROTECTED]) wrote:
: Ok, I'm having problems with pppd 2.3.8 under RedHat 6.0
: Basically I'm getting the LCP: timeout sending Config-Request error.
: However, this is occuring on all 3 of my different ISPs. I will
: include the information for an attempt at connecting to MSN.
: Here is a copy of my /var/log/messages file:
: May 26 13:02:16 Zeos ifup-ppp: pppd started for ppp0 on /dev/ttyS0 at
: 115200
: May 26 13:02:17 Zeos pppd[2019]: pppd 2.3.8 started by root, uid 0
<snipped>
: May 26 13:02:52 Zeos chat[2021]: send (^M)
: May 26 13:02:53 Zeos pppd[2019]: Serial connection established.
: May 26 13:02:53 Zeos pppd[2019]: Using interface ppp0
: May 26 13:02:53 Zeos pppd[2019]: Connect: ppp0 <--> /dev/ttyS0
: May 26 13:02:54 Zeos pppd[2019]: sent [LCP ConfReq id=0x1 <asyncmap
: 0x0> <magic 0xdd60ef1> <pcomp> <accomp>]
: May 26 13:03:21 Zeos last message repeated 9 times
: May 26 13:03:24 Zeos pppd[2019]: LCP: timeout sending Config-Requests
: May 26 13:03:49 Zeos pppd[2019]: Modem hangup
: May 26 13:03:49 Zeos pppd[2019]: Connection terminated.
: May 26 13:03:50 Zeos pppd[2019]: Exit.
: The IP is expected which sends a return.
: The ~ shows that MSN has started pumping out PPP garbage
The first thing you might try is to use IP '\d\c' instead of IP ''
and ~ '\d\c' instead of ~ '' in the chat script to eliminate the
carriage returns which can sometimes confuse an ISP. The carriage return
sent by the CONNECT '' may be needed to force a login/password prompt,
but otherwise it too is a candidate for replacement by CONNECT '\d\c' .
The \d inserts a delay of one second which may not be needed but which
shouldn't hurt.
If that doesn't at least change the problem to another one then you can
try eliminating the chat login/password as well as the IP and ~ expects,
use CONNECT '\d\c' instead of CONNECT '' to suppress the carriage
return there, and configure for PAP or CHAP or both. Unless you know
absolutely that PAP or CHAP authentication is not necessary. You may
well to *have* to use one of these authentications.
There are other things that could be wrong in the chat script and cause
the problem. A look at an exact copy of the chat script would be helpful.
Just to be sure check the type of UART configured for /dev/ttyS0 and
make sure it's correct for the type the modem actually uses. Moreover,
a 16450 UART is likely to be too slow for most modems today.
: Here is a copy of the route -n output:
: Kernel IP routing table
: Destination Gateway Genmask Flags Metric Ref
: Use Iface
: 10.49.99.152 0.0.0.0 255.255.255.255 UH 0 0
: 0 eth0
: 10.49.99.0 0.0.0.0 255.255.255.0 U 0 0
: 0 eth0
: 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0
: 0 lo
This will show a default routing through a PPP interface when things
work right.
: Here is a copy of ifconfig output:
: eth0 Link encap:Ethernet HWaddr 00:40:05:3C:24:61
: inet addr:10.49.99.152 Bcast:10.49.99.255
: Mask:255.255.255.0
: UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
: RX packets:291 errors:0 dropped:0 overruns:0 frame:0
: TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
: collisions:0 txqueuelen:100
: Interrupt:10 Base address:0x300
: lo Link encap:Local Loopback
: inet addr:127.0.0.1 Mask:255.0.0.0
: UP LOOPBACK RUNNING MTU:3924 Metric:1
: RX packets:48 errors:0 dropped:0 overruns:0 frame:0
: TX packets:48 errors:0 dropped:0 overruns:0 carrier:0
: collisions:0 txqueuelen:0
This will show a PPP interface configured when things work right.
: Here is a copy of the /etc/ppp/options file:
: noipdefault
: /dev/ttyS0
: lock
: crtscts
: debug
: passive
: defaultroute
These look good. I'm not used to seeing /dev/ttyS0 used for the modem
but since it dialed and connected it's OK.
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
/* The wealth of a nation is created by the productive labor of its
* citizens. */
------------------------------
Reply-To: "Curt" <[EMAIL PROTECTED]>
From: "Curt" <[EMAIL PROTECTED]>
Subject: Re: IP Masquerading
Date: Wed, 26 May 1999 20:46:16 -0500
You do not need a static IP.
Thunderbolt19 <[EMAIL PROTECTED]> wrote in message
news:7if7g5$cba$[EMAIL PROTECTED]...
> Do I have to have a static IP to do IP Masquerading if Im dialed into my
ISP
> via a modem? Thanks...
>
> --
> Eric Waters
> Total Distribution, Inc.
> [EMAIL PROTECTED]
>
>
------------------------------
From: Lorin Hochstein <[EMAIL PROTECTED]>
Subject: [EMAIL PROTECTED]
Date: Wed, 26 May 1999 21:57:00 -0400
subscribe linux-net
------------------------------
From: "G. Hugh SONG" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc,linux.redhat.misc
Subject: New knfsd trouble (was Re: NFS with Redhat 6 server and clients)
Date: Thu, 27 May 1999 10:06:26 +0900
Adnan Vora wrote:
>
> Another thing... I am using RedHat 5.2...
>
> On Wed, 26 May 1999, Adnan Vora wrote:
>
> > Date: Wed, 26 MAY 1999 13:39:03 -0500
> > From: Adnan Vora <[EMAIL PROTECTED]>
> > Newgroups: comp.os.linux.networking, comp.os.linux.misc, linux.redhat.misc
> > Subject: Re: NFS with Redhat 6 server and clients
> >
> > On Wed, 26 May 1999, Rich Piotrowski wrote:
> >
> > Thats exactly what my exports file says :
> >
> > /filesystem1 abc1.xyz.edu(rw) abc2.xyz.edu (rw) ....
> > /filesystem2 abc1.xyz.edu(rw) abc2.xyz.edu (rw) ....
> >
> >
> > The problem is only one client machine (out of 8) can access the remote
> > filesystem... besides even the one that can access the filesystem cannot
> > access the _other_ filesystem which is also exported..
> > These machines are (to the best of my knowledge) identically setup...
> > with the required differences ofcourse...
> > but still ... "permission denied"
> >
> > Any clues?
> >
> > Thanx
> > Adnan Vora
> >
> > PS the one that _can_ access a remote filesystem is not even the 1st
> > on the list...
> >
> >
> > > Date: Wed, 26 MAY 1999 12:15:36 GMT
> > > From: Rich Piotrowski <[EMAIL PROTECTED]>
> > > Newgroups: comp.os.linux.networking, comp.os.linux.misc, linux.redhat.misc
> > > Subject: Re: NFS with Redhat 6 server and clients
> > >
> > > On Wed, 26 May 1999 14:56:22 +0900, "G. Hugh SONG" <[EMAIL PROTECTED]>
> > > wrote:
> > >
> > > >Jon Paterson wrote:
> > > >>
> > > >> I am having a frustrating problem with a Redhat 6 server that was
> > > >> previously a redhat 5.2 server rebuilt.
> > > >>
> > > >> I have a Linux client (redhat 6.0 also) that is trying to connect to the
> > > >> server and is always getting the "permission denied" message.
> > > >>
> > > >> There is nothing wrong with the exports file, I have even deleted it and
> > > >> done the configuration through Linuxconf, and the same error exists.
> > > >>
> > > >> I think that it may have something to do with Knfsd, but am not sure
> > > >> where to look.
> > > >>
> > > >> can anyone help or point me in the right direction?
> > > >>
> > > >> regards,
> > > >>
> > > >> Jon Paterson
> > > >
> > > >I am having the same problem on both systems of Intel/Linux and
> > > >Alpha/Linux. The error message reads as
> > > >
> > > >fh_verify: ///permission failure, acc=3 error=13
> > > >
> > > >and something similar but related to nfs-something.
> > > >
> > > >I guess that it is not related to RedHat6.0. In my case,
> > > >it appears that it is related to the recent kernels, 2.2.8 and
> > > >2.2.9. It appears that it does no harm on the system. But,
> > > >certainly, it makes me quite nervous.
> > > >
> > > >Unfortunately, I don't know more than that.
> > > >
> > > >Regards,
> > > >
> > > >--
> > > >G. Hugh Song
> > >
> > > Simple!
> > >
> > > Redhat 6.0 NFS now defaults to Read-only. Check your setup in
> > > linuxconf again! Or check man exports. My exports file now loks like
> > > this.
> > >
> > > / machine.name(rw)
> > >
> > > Note the addition of the "(rw)" switch to allow read-write.
> > >
> > > Rich Piotrowski
> > >
> > >
> > >
> >
> > **********************************************************************
> > Adnan Vora Email : [EMAIL PROTECTED]
> Adnan Vora Email : [EMAIL PROTECTED]
My error messages on the server console are
fh_verify: ///permission denied, ...
nfs_stat_to_errno: bad nfs status return number ...
My guess is that we are all using the knfsd package and Kernel-2.2.?.
In /usr/src/linux-2.2.?/Documentation/Changes file, it is
explicitly stated that "NFS is currently under heavy revision" to
work as a kernel-based NFS.
I wish that the stable kernels work a little more "stably" than they
do now. I do not subscribe to the kernel development mailing list.
If someone do, please post this to the mailing list. Thank you.
Regards,
--
G. Hugh Song
------------------------------
Reply-To: "Curt" <[EMAIL PROTECTED]>
From: "Curt" <[EMAIL PROTECTED]>
Subject: Re: IP Masquerade/Routing
Date: Wed, 26 May 1999 20:52:25 -0500
I'm afraid you've lost me.
While I have not tried it, IP CHAINs in kernel 2.2 is advertised as capable
of doing a 'many to many' NAT.
IP MASQ is a 'many to one' NAT. I don't know if this helps or not.
Athol Marshall <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Curt,
>
> What I want to do is this :
>
> Set-up my client's hosts as
> 192.168.200.1
> 192.168.200.2 etc
>
> These same hosts have addresses
> 128.100.200.1
> 128.100.200.2 etc. within my client's private network.
>
> Now I set-up a route to 192.168.200.0 which will cause a PPP connection to
be
> opened to my client, where a dynamic IP address is negotiated.
>
> Anything sent to the real 128.100.0.0 would by-pass this route and get
sent-out
> via the default route to the internet. See - I want to make sure there can
> NEVER be a conflict.
>
> Packets sent to 192.168.200.x should then be changed on the way out (after
the
> route has been determined) so the destination address in the packet is
> 128.100.200.x, and the router at the other end should then be able to
deliver
> them. This mapping would be handled by a lookup table maintained via a
program
> similar to ipfwadm.
>
> I know this isn't what's meant by IP masquerading. I think this is a
different
> problem, and one not covered by anything I have read. All the FAQs &
HOWTOs
> seem to assume you want to access a single private network or a single
private
> network and the Internet.
>
> It isn't just a matter of routing to an "illegal" address. What if 2
clients
> have both chosen to use the same private network number, say
192.168.100.0 -
> which is even the same one as I've chosen? I would then need to fool the
kernel
> into handling 3 different networks all using the same perfectly valid
network
> number.
>
> Do you know if it's worth pursuing this myself (ie learning more about the
> kernel and changing it), or is there some reason why this definitely
cannot
> work?
>
> thanks,
>
> Curt wrote:
>
> > If you're question is: Can I masquerade network
>
> > 128.100.0.0? The answer is
> > yes. The same way you'd masq 198.168.x.y. The only drawback being if
if
> > they try to access the REAL 128.100.0.0 network it will stay local.
Make
> > sure you reject access to the 128.100.0.0 network from outside too.
> >
>
>
>
------------------------------
From: bryan <[EMAIL PROTECTED]>
Subject: Re: Looking for good reliable 100Mb Ethernet card for RH 6.0
Date: Thu, 27 May 1999 01:25:51 GMT
the current wisdom seems to point to the intel etherexpress pro/100.
read the header comments (in the kernel source) to find out which
chipsets are preferred.
John Antypas <[EMAIL PROTECTED]> wrote:
: The subject says it all. I'm looking for the "safe" reliable 100Mb Ethernet
: card for RH 6.0. Years ago, it would have been the 3Com 3c509, then it was
: the tulip chipset. Now that many folks are dropping that, what is the solid
: card that (a) doesn't cost $75 per card and (b) works, automatically.
--
Bryan
------------------------------
From: "Fernando" <[EMAIL PROTECTED]>
Subject: Re: Can't see Samba Server from Win95
Date: Wed, 26 May 1999 18:46:47 -0300
Hi Mike,
It is not possible to route the protocol Netbios.
Like this, if I understood correctly, if Cisco router is between Win95 box
and Linux box (Samba Server), you should enable Netbios on TCP/IP in your
router.
I hope that helps
Fernando
Sao Paulo/SP - Brazil
Mike Dion <[EMAIL PROTECTED]> escreveu nas not�cias de
mensagem:7iefo2$8o5$[EMAIL PROTECTED]
> My Samba Server and WIN 95 computer are on separate network segments and I
> would like to share my Samba Server's filesystem with the Win95 computer.
>
> To date all of my attempts to configure Samba have failed and I have yet
to
> see any Samba services from my Win95 computer! The DOS command "net view
> \\computername" fails with an error 51 and message that the remote
> computer is not accepting or responding to requests.
>
> Background:
> +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> - Win95 computer on LAN with Windows NT Server "TIMNT" as primary
> Domain Controller and Master Browser; Domain is domain
> - PC can reach the Linux box by ping, traceroute, ftp, etc.
>
> - Samba Server (Linux) is on a separate network segment direct connected
> to the segment on which the Win95 PC resides
> - Samba server resides n the DMZ, between a Cisco router and
> firewall computer
> - Red Hat Linux V5.2 and have installed samba-1.9.18
> - testparm against smb.conf replies "Loaded services file OK"
> +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
> I'm stumped -- is the firewall the problem? It has never been a problem
in
> the past. The firewall has always permitted traffic to flow thru
providing
> that requests were initiated from the inside?
>
> Your help in this matter would be greatly appreciated!
>
> Sincerely,
>
> Mike D.
> Halifax, NS Canada
>
> ------------------ Posted via SearchLinux ------------------
> http://www.searchlinux.com
------------------------------
From: Stefan <[EMAIL PROTECTED]>
Subject: Re: Pinging for behind IP Masq??
Date: Thu, 27 May 1999 01:51:00 +0200
Matt Goebel wrote:
>
> Is there a way to ping from a client on a masqueraded connection?
Whats the Problem? The ICMP Packet is also Masqueraded so you would get
a reply.
------------------------------
From: "John Antypas" <[EMAIL PROTECTED]>
Subject: Looking to find ipmasqadm code
Date: Thu, 27 May 1999 02:46:04 GMT
Since the list has been so helpful, I figure I'd ask away. It's so nice to
see Usenet with people back as opposed to ads.. Thanks people -- hope I can
lend a hand to someone soon.
I just brought up a RH 6.0 server (2.2.5 kernel). Ipchains was easy enough
to get going. However, port forwarding INTO the net seems to require a tool
called ipmasqadm. This tool claims to be found on
file://juanjox.linuxhq.com. Of course, linuxhq isn't reachable, let alone
juanjox. Where can I find this tool or its author?
John.
------------------------------
From: "Ron Hill" <[EMAIL PROTECTED]>
Subject: contest
Date: 26 May 1999 16:23:41 GMT
Enter to win at http://www.gennow.com
Check out our deals !!
Complete Computer packages from 699.99 USD 999.99CDN
check it out and enter to win at http://www.gennow.com
------------------------------
Crossposted-To: de.comm.internet.routing
From: [EMAIL PROTECTED] (Helge Oldach)
Subject: Re: Linux: ICMP Redirect, IP Source Routing unterdruecken
Date: Wed, 26 May 1999 22:04:47 GMT
In <7ifh5i$ejq$[EMAIL PROTECTED]> Bernd Eckenfels <[EMAIL PROTECTED]> writes:
| In comp.os.linux.networking Detlef Bosau <[EMAIL PROTECTED]> wrote:
| > Wenn ein Design Stuss ist, und icmp redirects sind Stuss, zwar
| > aus der Historie erklaerbar, da hat man ja alles moegliche in den
| > DoD Salat gemengt, aber es ist und bleibt Stuss, kann man es nun
| > langsam ja mal verabschieden.
| Hmm... ich finde sie praktisch. Spart eine Menge Arbeit.
Klar. Aber kein Geld: Das verdreifacht gegen�ber einer seri�sen
Konfiguration die Anzahl der Pakete und verlangt ergo nach Fast
Ethernet. Ohne Not, wohlgemerkt.
------------------------------
From: kite@NoSpam.%inetport.com (Clifford Kite)
Subject: Re: RH6 & PPPD 2.3.7 weirdness
Date: 26 May 1999 14:24:06 -0500
Peter Schwenk ([EMAIL PROTECTED]) wrote:
: May 26 10:23:15 localhost kernel: registered device ppp0
: May 26 10:23:15 localhost pppd[606]: pppd 2.3.7 started by root, uid 0
: May 26 10:23:15 localhost pppd[606]: Using interface ppp0
: May 26 10:23:15 localhost pppd[606]: ioctl(SIOCSIFDSTADDR): Cannot
: assign requested address(99)
: May 26 10:23:16 localhost pppd[606]: Exit.
: I have no idea what the "pppd[606]: ioctl(SIOCSIFDSTADDR): Cannot assign
: requested address(99)" part of the messages means. Does anyone have a
: clue about this? I use the "demand" option of pppd at home with a
: desktop PC just fine, so this is really odd.
Try using
pppd 192.168.0.1:192.168.0.2 ipcp-accept-local ipcp-accept-remote demand ...
This worked for me. The options above are only those needed for demand
dialing, use them in addition to others that you use. The addresses
don't have to be the ones specified above, most likely any two on the same
subnet will do. Oh, and you likely need to setup for dynamic IPs with
"echo -n 1 > /proc/sys/net/ipv4/ip_dynaddr"
--
Clifford Kite <kite@inet%port.com> Not a guru. (tm)
/* Those who can't write, write manuals. */
------------------------------
From: "DB" <[EMAIL PROTECTED]>
Subject: Using Linux as a Router
Date: Thu, 27 May 1999 02:55:10 GMT
I want to use Linux to act as a router between a 10BaseT network and a DSL
connection to the Internet.
Can anyone provide information about how to configure a recommended version
of Linux as a router or point me to resources where I can research this?
Thank You!
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Recent problems with my cablemodem (using DHCP)
Date: Wed, 26 May 1999 21:13:07 GMT
I should add that following boot, the networkings works perfectly, but
after an hour or so, I can't ping anymore.
--== Sent via Deja.com http://www.deja.com/ ==--
---Share what you know. Learn what you don't.---
------------------------------
** 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.networking) 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-Networking Digest
******************************
