Linux-Networking Digest #905, Volume #9 Sun, 17 Jan 99 01:13:44 EST
Contents:
Samba and "Sendmail" on a notebook ("Daniel Hogan")
Re: Commands ("Kyle Bowerman")
Re: This is Linux, not Windows, so why not superior flexibility AND idiot-friendly?
(Alexander Viro)
Re: [Q] : DNS problem ? ([EMAIL PROTECTED])
Connection freezes (SuprMath)
Re: This is Linux, not Windows, so why not superior flexibility AND idiot-friendly?
(Kevin Martin)
way to browse machines on a network??? (David Erie)
Re: linux installation via NFS ("Chip & Debby Piller")
firewall: http-gw question ("paul malabad")
Re: Linux Article - Please help (Lord Spurius)
Re: Gateway Simple Install.. (Philip Hirschhorn)
Re: Revenge of NT? (Paul)
Samba: request_oplock_break: no response (Mark Johnson)
Re: Pentium -> 486 via ftp done :) (Mikhail Bovineck)
Cyrus 1.5.19 compile error on RH 5.2 (Tim Gibson)
Re: External ISDN adapter - Does it need to use mlppp? ([EMAIL PROTECTED])
What Version to get? (Wee Moose)
Re: PPP: LCP protocol reject with USWEST ISP (Clifford Kite)
Re: Hacking Win95 for CHAP secret (Reinder)
Newbie / hosts file / domain name? ("jim")
Problems running dhcpcd for multiple interfaces (Jim Grusendorf)
----------------------------------------------------------------------------
From: "Daniel Hogan" <[EMAIL PROTECTED]>
Subject: Samba and "Sendmail" on a notebook
Date: 16 Jan 1999 03:44:47 GMT
I have my notebook computer configured to browse the internet using a
network. When I take this computer off the network and to boot it up, it
takes forever. It always stays on Sendmail for about 10 minutes and Samba
for about 30 minutes. Is there anyway to shorten the timeout wait?
------------------------------
From: "Kyle Bowerman" <[EMAIL PROTECTED]>
Subject: Re: Commands
Date: Fri, 15 Jan 1999 21:49:24 -0700
I use slackware and I do it a little diffrent
first you have to see what device is associated with you cd room: mine
is sonycd
mount /dev/sonycd /cdrom
Sechylmanos wrote in message ...
>In article <7775hd$39p$[EMAIL PROTECTED]>, "J.A Nothling"
><[EMAIL PROTECTED]> wrote:
>
>> Hi there,
>>
>> I am very new to Linux, can anybody tell me how to access disk drives
like
>> my CD-ROM and so forth in linux?
>>
>> It is a easy question, to which I am having a bit of trouble. I am
probably
>> looking in all the wrong places!
>>
>> Regards
>> J.A Nothling
>
>In article <[EMAIL PROTECTED]>, Arthur Chiu <[EMAIL PROTECTED]>
wrote:
>
>>> In Linux/Unix, storage device has to be mounted before it can be
accessed.
>>>
>>> Learn to use the command 'mount' first.
>
>You should definitely learn how to use the mount command, but what you
>need to do for accessing your CD-Rom or your Floppy disk is like this.
>(could be slightly different on your machine, I am using RedHat Linux...)
>
>to access the cd-rom:
>type "mount /mnt/cdrom" and then you will be able to go into the
>directories on the CD-ROM. The floppy drive is the same, only use "mount
>/mnt/fd0".
>
>Both of these assume that the devices have been set up in /etc/fstab.
>(which they should be automatically if you are using RedHat linux 5.X)
>
>Hope this helps...
>
>Sechylmanos
------------------------------
From: [EMAIL PROTECTED] (Alexander Viro)
Crossposted-To:
comp.os.linux.misc,comp.os.linux.portable,comp.os.linux.powerpc,comp.os.linux.setup
Subject: Re: This is Linux, not Windows, so why not superior flexibility AND
idiot-friendly?
Date: 15 Jan 1999 22:47:38 -0500
In article <36a10353.9119122@News>, George Marengo <[EMAIL PROTECTED]> wrote:
>> Is there anything outside of a ported unix tool
>> that does regular expressions in Windows?
>
>Is there some reason why it would matter if it was ported or not?
>If the tool is available, what difference does it make?
Dunno if the Borland's grep was ported or written from scratch (not a BFD)
but it was there around '87 (at least). One of the things that made the
damned parody on OS slightly more tolerable. But I *missed* awk. And sed.
And sh. And... aaaarrggh. Memories.
--
"You're one of those condescending Unix computer users!"
"Here's a nickel, kid. Get yourself a better computer" - Dilbert.
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: [Q] : DNS problem ?
Date: Sun, 17 Jan 1999 04:41:34 GMT
In article <77qbin$ilv$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] wrote:
>
>
> > Hey all !
> >
> > I'm having a problem which fix should be easy but as I am new to linux...
> > and I couldn't find any clear reference to it in the docs... So here's : I
> > did setup my PPP connection with the KDE dial-up utility or whatever it's
> > called. I get connected, I can tell this because a window "connected" with
> > connection time shows up. But now, whatever networking utility I load
> > (tried ftp, telnet, cftp, netscape, pine, kde specific stuff...) , when I
> > try to reach a distant computer I always get an error message telling me
> > that the host does not exist. So I guess it's a DNS config problem, right ?
> > I checked the DNS names, they are correct (my provider's), and I know these
> > DO work because I used to use them when I had Winwhatever_you_call_it.
> >
> > Any clue for helping me getting online would be GREATLy appreciated...
> >
> > Thanks for ya time
> > --
> > Djaak
> >
> > e-mail : [EMAIL PROTECTED]
> >
>
> this sounds similare to my problem, I ran ifconfig, and it showed I was
> connected to ppp with an ip address assigned and I can ping any IP address,
> but I cant ping a name like www.ibm.com when I could before. I used to be
> with Red Hat 5.1, then I totaly reloaded and with 5.2 (on a new HD) and the
> name resolve still wont work. I did find that I keep getting a core file
> caused by a script in /etc/sysconfig/network-scripts when one of the scripts
> calls a grep ^[g^] (or somthing like that).
>
> If any one can help please let us know.
>
> Thanks Sherm
>
Well I found my problem, I had a typo in the /etc/resolv.conf I had put a
207.17.177.10 instead of 204.17.177.10.
But I still would like to know why the linuxconf ppp scripts cause a segment
fault and create a core file. I just use manual scripts instead of the
linuxconf utility, and it works.
Sherm
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED] (SuprMath)
Subject: Connection freezes
Date: 17 Jan 1999 04:00:24 GMT
Whenever I try to transmit a large amount of data from my Windows computer to
my Linux box, the connection freezes. This amount of data seems to be about
15K. Both computers use 3c503 network cards and the Windows computer is
300Mhz, 64Mb of RAM and the Linux box is 50Mhz, 12Mb of RAM. All other network
operations work fine, and the Linux box can send as much data as it wants
without freezing. The connection locks when retrieving a file that is larger
than about 15K from Windows via HTTP or FTP. If anybody has any ideas I would
really appreciate the help.
=Brian
Oh, by the way, my FTP software (sending from Windows) gives this error
message:
Send error: Connection reset!
Transmitted 9216 bytes in 173.6 secs, (531.00 bps), transfer failed
Recieve error: Blocking call canceled!
------------------------------
Crossposted-To:
comp.os.linux.misc,comp.os.linux.portable,comp.os.linux.powerpc,comp.os.linux.setup
From: [EMAIL PROTECTED] (Kevin Martin)
Subject: Re: This is Linux, not Windows, so why not superior flexibility AND
idiot-friendly?
Date: Sat, 16 Jan 1999 04:19:59 GMT
>I guess my point is, what is the goal of users of the Linux platform,
>anyway?
Good question, but why the devil are you asking it in comp.os.linux.misc,
AND comp.os.linux.networking, AND comp.os.linux.portable, AND
comp.os.linux.powerpc, AND comp.os.linux.setup?
*I'm* not idiot-friendly. Neither is Usenet. If you don't pay attention
to where your followups are going, you can be tricked into all sorts of
unpleasant things. If you don't pay attention to where your followups go,
you're just a spam-sprayer waiting for someone to abuse you -- or use you to
abuse someone else.
------------------------------
From: David Erie <[EMAIL PROTECTED]>
Subject: way to browse machines on a network???
Date: Sun, 17 Jan 1999 04:14:40 +0000
==============DB57658A666F90007B1A9480
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
I was wondering if there was a way in Linux (Xwindows) to browse the
machines on a network. I'm used to the "network neighborhood" that
windows has...Does anyone know that this is possible?
thanks.
--
David Erie derie@cc
Computer Engineering CS1502 Recitation TA
Georgia Institute of Technology
==============DB57658A666F90007B1A9480
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
I was wondering if there was a way in Linux (Xwindows) to browse the
machines on a network. I'm used to the "network neighborhood" that
windows has...Does anyone know that this is possible?
<br>thanks.
<pre>--
David
Erie
derie@cc
Computer
Engineering
CS1502 Recitation TA
Georgia Institute of Technology</pre>
</html>
==============DB57658A666F90007B1A9480==
------------------------------
From: "Chip & Debby Piller" <[EMAIL PROTECTED]>
Subject: Re: linux installation via NFS
Date: Sat, 16 Jan 1999 04:23:54 GMT
[EMAIL PROTECTED] wrote in message
<77nj1j$8ev$[EMAIL PROTECTED]>...
>Hi folks!
>Maybe someone can help me: I try to install Linux via NFS, but on the
client
>always the message "permission denied" appears when I confirm the directory
>on which the CDROM is mounted on the NFS-Server.
>I think I have to put in the IP-Number of the client in some special files.
>I did it in the etc/hosts and etc/exports, but no success.
>Is anybody able to describe me the correct way of installation?
>Many thanks
>
>Matthias
I am not sure which distribution you are trying to install, but I just went
through a NFS install of RedHat5.2 from a RedHat mirror where I work. So I
knew that I had a good IP address for me (client), and I was able to ping
the NFS server and knew it's name and IP address. I also knew the netmask,
gateway, and nameserver information.
The install program presents several screens that ask you for information,
Boot Protocol, Configure TCP/IP, Configure Network, and NFS Setup. I could
not get it to work until I figured out the right directory in NFS Setup.
For my site I thought the correct directory was
local/mirror/redhat/redhat-5.2/i386/RedHat which contains
RPMS/
base/
i386
instimage/
but this was too deep, I had to back out one level to the directory which
contains the RedHat directory (local/mirror/redhat/redhat-5.2/i386) and then
it worked fine.
Hope this helps.
------------------------------
From: "paul malabad" <[EMAIL PROTECTED]>
Subject: firewall: http-gw question
Date: Sun, 17 Jan 1999 00:11:54 -0800
When I visit some sites I get a user=unauth error in my messages file.
who is unauthorized, the client or the destination host? Please help.
The message file excerpt (sp?) follows:
Jan 17 00:00:23 ci http-gw[8691]: permit host=client4/192.168.0.4 use of
gateway
(V2.1)
Jan 17 00:00:23 ci http-gw[8691]: log host=client4/192.168.0.4 protocol=HTTP
cmd=
get dest=windowsupdate.microsoft.com path=/x86/W98/en/ie4//cucif.cab
Jan 17 00:00:26 ci http-gw[8691]: exit host=client4/192.168.0.4 cmds=1
in=6262 ou
t=0 user=unauth duration=4
Any help would be greatly appreciated.
Paul Malabad
------------------------------
From: Lord Spurius <[EMAIL PROTECTED]>
Subject: Re: Linux Article - Please help
Date: Fri, 15 Jan 1999 23:13:32 +0000
Ren wrote:
> Hi,
> I?m an IT bod who does a bit of freelance journalism on the side. I am
> writing an article on Linux for a UK Internet magazine and want to
> feature a few businesses or institutions that are using Linux. What I
> want to know is what is Linux like in a live environment, how well is it
> supported by the new groups of ?vendors? ? Also is anyone running heavy
> apps on it like Oracle.
>
> TIA -Ren
> [EMAIL PROTECTED]
Not sure if this is what you're lookin for, but Titanic was rendered
on a system of 160 433 mhz alphas (working as 1 system over 100mbs
ethernet). They oringinally tried NT, but it couldn't handle their
system. They then tried linux. The original install wasn't able
to handle it either, but thanks to the fact that it was open source
they were able to up the maximums to meet their needs.
------------------------------
From: [EMAIL PROTECTED] (Philip Hirschhorn)
Subject: Re: Gateway Simple Install..
Date: 17 Jan 1999 05:15:34 GMT
News Account ([EMAIL PROTECTED]) wrote:
: This is a multi-part message in MIME format.
: ------=_NextPart_000_0008_01BE39E9.BFBAB4C0
: Content-Type: text/plain;
: charset="iso-8859-1"
: Content-Transfer-Encoding: quoted-printable
: Good day. I have 2 win98 machines i wish to access the internet =
: through my
: linux box. I have the ethernet card setup for my connection out to the =
: internet (DHCP it works).
: And the one to my regular internal network ( hard coded IP it works).
: =20
: Now I just need to know how to forward packets from my internet =
: connection
: to my windows machines.
: =20
: Please Help me im new to linux.=20
: =20
: JKinn.
: [EMAIL PROTECTED]
You need to set up IP Masqerade (and any other firewalling features
you want). There's an excellent description of the process at
http://rlz.ne.mediaone.net/linux/index3.shtml
--
======================================================================
Philip Hirschhorn [EMAIL PROTECTED]
[EMAIL PROTECTED]
======================================================================
------------------------------
From: Paul <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: Revenge of NT?
Date: Sun, 17 Jan 1999 05:15:55 +0000
Jason A Fletcher wrote:
>
> Got an odd kind of problem here, so I'll give a short version and a long
> version.
>
> Short: re-installing RedHat 5.2 (Linux kernel 2.0.36) on a Dell P2 400
> with 8 GB drive. The network card appears to be a 3Com 3c905B Cyclone.
[snip]
Ok, you have a 3Com 905B. THIS CARD IS PCI SO IGNORE THE PREVIOUS RESPONDANT'S
SUGGESTION TO USE ISAPNPTOOLS. My best friend has one of these. First thing you
need to do is run "ifconfig" at the command prompt, and you should see somewhere
in there:
....
eth0 Link encap:Ethernet HWaddr 00:80:C8:47:43:A4
....
I don't have it exactly in front of me, but it should be quite similar. Now, the
hex numbers "00:80:C8:47:43:A4" are the card's MAC, or hardware, address. If yours
says "ff:ff:ff:ff:ff:ff" then you need to turn off the computer and YOU NEED TO
UNPLUG THE COMPUTER FROM THE WALL for at least 10 seconds (this was advice that I
got from Donald Becker, the author of the driver for your card). This will
completely reset the card, and when you boot up the computer next time it should
show the correct MAC address, and the card should work. Windows puts the card in a
1/2 sleep mode that the Linux driver doesn't quite understand. The newest
development 3c59x drivers (yes, the driver is called "3c59x") fixes this.
Now, if Linux detects the correct MAC address of the card, but you still can't get
it to ping anything (use IP numbers since DNS could have problems of it's own,
Yahoo's IP is 204.71.200.72), take a look at the LEDs on the back of the card. If
you are connected to a 100Mbps network and the LEDs say that the card is set for
10Mbps, then you need to go to
http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html and get the latest
DEVELOPMENT version of the 3c59X.c driver, copy it to /usr/src/linux/drivers/net/
and then recompile your kernel (If you have never done this then read the
Kernel-Howto at http://metalab.unc.edu/LDP/HOWTO).
If you have gotten this far and it still doesn't work or something else went
wrong, I have a few more tricks up my sleve.
Paul Laufer
------------------------------
From: Mark Johnson <[EMAIL PROTECTED]>
Subject: Samba: request_oplock_break: no response
Crossposted-To: comp.protocols.smb
Date: Tue, 12 Jan 1999 08:49:54 GMT
I'm having an odd problem with Samba. I have it set up and working great
with user security. One logs in and has access that is expected. Copying
files from the server is great, very fast. However, when I copy files
from a Win98 client to the server, its very slow. I've noticed that at
times it will log this message:
1999/01/11 18:24:32 request_oplock_break: no response received to oplock
break request to pid 411 on port 1029 for dev = 309, inode = 28805
This does not appear to happen every time though, especially with small
files. But copying files to the server is VERY slow. I've looked through
the FAQs and web with no solutions. I've messed with OPLOCKS, and
locking, no good. As I said before, copying from the Samba server to the
Windows client is very fast.
Using Redhat 5.2 with samba-1.9.18p10-5. Very few changes to smb.conf.
Thanks for any suggestions!
------------------------------
From: [EMAIL PROTECTED] (Mikhail Bovineck)
Subject: Re: Pentium -> 486 via ftp done :)
Date: Sun, 17 Jan 1999 03:20:58 GMT
Reply-To: [EMAIL PROTECTED]
Well the problem is now fully resolved. It was really two problems.
Firstly there seems to be something slightly fishy about APC's RedHat
ftp via plip process. In order to start the ftp installation via plip
you need to enter...
192.168.1.2 (IP of 2nd machine)
255.255.255.0 (netmask)
0.0.0.0 (gateway)
_ _ _ _ _ _ _ (primary nameserver)
The procedure will say "SIOCADDRT: Network is unreachable", and you
will be asked to either go back to the previous step, retry or abort.
I go back, and then re-enter...
192.168.1.2 (IP of 2nd machine)
255.255.255.255 (netmask) - note the change!
0.0.0.0 (gateway)
_ _ _ _ _ _ _ (primary nameserver)
And then the procedure starts.
The next problem is ftp-timeout. It is set at 15 mins by default (I
haven't figured out how to extend it yet!). If you spend too much
time over the selection of packages the connection will timeout. This
shouldn't be a problem and in fact the installation seems to restart
fine. BUT ALL IS NOT WELL and it will puke in a number of ways down
the track (in my case after about 20M of transfer).
I thought this was a CD problem and transferred all the files to HD
which worked but only because my package selection was becoming
streamlined and I was getting in under the 15 mark (usually by only a
min or so!). I can now install from the CD as long as I'm fast with
the package selectioin.
It seems you must select to format the root drive in order for the
installation to be happy so if you have a big drive good luck! I just
made it with 150M so I think if it is much bigger you could do a
double installation, formatting one time and then installing the next.
Or breaking up the partitions a bit.
Anyway that's my story - it took a week of frustration but the 486 is
now a standalone Linux box and all is well with the world :))
Cheers
On Mon, 11 Jan 1999 02:23:32 GMT, [EMAIL PROTECTED] (Mikhail
Bovineck) wrote:
>
>Hi - I have Linux installed on a hardrive on a pentium box which
>originated on a CD ( RedHat Distribution). I also have a 486
>with no CD which I would like to make a Linux box as well. In other
>words, I would like to access the pentium box's CDrom from the 486.
>
>Currently the 486 has muLinux installed on it, downloaded from the net
>and installed using 3 floppies.
>
>All I have to connect the two is a null modem cable. According to the
>RedHat docs I should be able to install from the CD via a network but
>do I need a network card for that? Is the /dev/cua1 port sufficient?
>
>If so, how do I configure the durn thing? All the Howto's seem to
>just be short of an answer that works!
>
>I have read all the docs from the LDP and the RedHat distribution but
>my head is spinning around! Could someone please please explain in
>English for me how I connect the two (i.e. what protocol, what scripts
>etc). I'm sure someone has had to do this before.
>
>I thank you in anticipation,
>
>Cheers :)
>
------------------------------
From: Tim Gibson <[EMAIL PROTECTED]>
Subject: Cyrus 1.5.19 compile error on RH 5.2
Date: Sat, 16 Jan 1999 19:11:28 -0800
While trying to 'make all' on my RedHat 5.2 system I get this:
/usr/src/cyrus-imapd-v1.5.19/cyradm/cyrmain.c:294: undefined reference
to `Tcl_Write'
/usr/src/cyrus-imapd-v1.5.19/cyradm/cyrmain.c:299: undefined reference
to `Tcl_Write'
/usr/src/cyrus-imapd-v1.5.19/cyradm/cyrmain.c:300: undefined reference
to `Tcl_Write'
/usr/src/cyrus-imapd-v1.5.19/cyradm/cyrmain.c:319: undefined reference
to `Tcl_Eval'
make[1]: *** [cyradm] Error 1
make[1]: Leaving directory `/usr/src/cyrus-imapd-v1.5.19/cyradm'
make: *** [all] Error 1
It is abridged slightly here. I just scolls with similar messages
before this.
I configured it using './configure --with-tcl=/usr/lib/tcl8.0' and 'make
depend' ran fine. What am I doing wrong? Thanks.
Please cc me via e-mail at [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.help
Subject: Re: External ISDN adapter - Does it need to use mlppp?
Date: Sun, 17 Jan 1999 05:37:22 GMT
Hello,
I am having the same problem you discussed in your posting to the linux
newsgroup. What program should I be using to connect? Using netcfg (part of
control-panel), I can connect on one B channel only. I tried entering the
commands Andrew Marcum listed in his posting:
ATS71=1
ATS80=1
AT&W
and now netcfg one register ppp0 after it connects on the one B channel.
When I type ifconfig, all I see is leo0. Any ideas?
-Roddy
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: [EMAIL PROTECTED] (Wee Moose)
Subject: What Version to get?
Date: 17 Jan 1999 05:51:04 GMT
Ok, I am thinking about installing a trial network in my lab with the goal of
replacing the token ring/netware network.
I want to know what is the best for a mostly trouble free install
Red Hat?
Slackware?
FreeBSD?
Denebian?
ETC?
Carl Bogardus, Technology Instructor/Coordinator
Teachers' Page at http://members.aol.com/index.html
Chaparral Middle School
http://www.homeworkheaven.com/schools/chaparral
Remove junkbloc from address to e-mail me.
------------------------------
From: [EMAIL PROTECTED] (Clifford Kite)
Subject: Re: PPP: LCP protocol reject with USWEST ISP
Date: 16 Jan 1999 21:10:39 -0600
Russ Gritzo ([EMAIL PROTECTED]) wrote:
: folks:
: I started using uswest as an ISP in october of last year. Worked
: great. In late december I stopped being able to dial in to USWest, I
: can dial into other ISP's fine, and dial into USwest from Windoze ok,
: but not from my linux (slackware 2.0.30) box. Was using pppd 2.2,
: upgraded (along with necessary kernel patches) to pppd 2.3.5, but no
: luck. Problem seems to be manifested as an LCP protocol reject message
: while trying to negotiate a dynamic IP address, but the protocol being
: rejected is the IP network configuration request. Not a chap or pap
: problem, but seems to be more fundamental. USwest is no help, of
: course, since they don't do linux, but now this has become personal. :}
: They claim to have changed from 56flex to v.90 in late december, but
: this problem is irrespective of modem and modem protocol. Have tried
: with / without compression, and most every other ppp-option to no
: avail. Any ideas? edited debug logs attached.
..
: /var/log/debug:
: Jan 16 09:18:42 david pppd[871]: sent [LCP ConfReq id=0x1 <magic
: 0xa2f03242> <pcomp> <accomp>]
: Jan 16 09:18:42 david pppd[871]: rcvd [LCP ConfAck id=0x1 <magic
: 0xa2f03242> <pcomp> <accomp>]
: Jan 16 09:18:42 david pppd[871]: rcvd [LCP ConfReq id=0xe2 <asyncmap
: 0xa0000> <magic 0x924906a2> <pcomp> <accomp>]
: Jan 16 09:18:42 david pppd[871]: sent [LCP ConfAck id=0xe2 <asyncmap
: 0xa0000> <magic 0x924906a2> <pcomp> <accomp>]
: Jan 16 09:18:42 david pppd[871]: sent [IPCP ConfReq id=0x1 <addr
: 0.0.0.0> <compress VJ 0f 01>]
: Jan 16 09:18:42 david kernel: ppp: write frame, count = 20
: Jan 16 09:18:42 david kernel: FF 03 80 21 01 01 00 10 ...!....
: Jan 16 09:18:42 david kernel: 03 06 00 00 00 00 02 06 ........
: Jan 16 09:18:42 david kernel: 00 2D 0F 01 .-..
: Jan 16 09:18:42 david kernel: ppp_dev_xmit_lower: fcs is 1427
: Jan 16 09:18:42 david kernel: ppp_dev_xmit: writing 22 chars
: Jan 16 09:18:42 david kernel: ppp: receive frame, count = 26
: Jan 16 09:18:42 david kernel: FF 03 C0 21 08 E3 00 16 ...!....
: Jan 16 09:18:42 david kernel: 80 21 01 01 00 10 03 06 .!......
: Jan 16 09:18:42 david kernel: 00 00 00 00 02 06 00 2D .......-
: Jan 16 09:18:42 david kernel: 0F 01 ..
: Jan 16 09:18:42 david pppd[871]: rcvd [LCP ProtRej id=0xe3 80 21 01 01
: 00 10 03 06 00 00 00 00 02 06 00 2d 0f 01]
Here's some suggestions but they are admittedly shots in the dark. The
ISP has rejected the first IPCP protocol request as you observed and
doesn't even seem to recognize it. The kernel debug shows what appears
to me to be a perfectly valid IPCP request that matches that returned
in the ProtRej message. This is an unusual problem.
Anyway I would try matching the asyncmap of a0000 that the ISP requests
for itself - failure at the IPCP stage often occurs because an ISP ppp
implementation fails to continue using the default asyncmap (ffffffff)
when none is requested by the client. This usually takes longer though,
with several exchanges before one side or the other finally quits.
I also would drop any "escape FF" option should one be present, it can
cause trouble on occasion.
As a matter of course I would make sure that the boot-up setserial command
included the spd_vhi option and the appropriate /dev/ttySx was set for
38.4 kbps .
If these fail then I would try configuring for PAP since most ISPs do
that - and sometimes the ISP personnel haven't a clue as to what changes
occur in software they install. And I'm not sure what happens if the
ISP expects an AuthReq for PAP and gets an IPCP request instead.
It doesn't seem likely that the change to v90 caused your problem since
the ISP would need to maintain compatibility for those customers without
v90 capability. But I don't know enough to rule this out entirely.
--
Clifford Kite <[EMAIL PROTECTED]> Not a guru. (tm)
/* Better is the enemy of good enough. */
------------------------------
From: Reinder <[EMAIL PROTECTED]>
Subject: Re: Hacking Win95 for CHAP secret
Date: Sun, 17 Jan 1999 05:34:33 +0000
Leon Wood wrote:
> My ISP has flatly told me they will not help me configure Linux.
> Clearly they've aligned themselves with the forces of darkness in
> Redmond. If I knew what the CHAP secret was I think I could do it
> despite their obstructions. Win95 DUN works so the secret is in there
> somewhere, can anyone tell me where and how to find it?
>
> TIA
>
> Leon Wood
To connect your linux machine to your ISP using CHAP do this :
first edit the file /etc/ppp/chap-secrets.
it reads something like this:
# client server secret ........................
here you fill in your username and password, thus
username * password
then edit the file /etc/resolv.conf.
put the IP addresses of your nameservers in here, like
search whatever.something
nameserver what.ever.it.is
nameserver add.another.one.here
then make a simple script to dial in.
make sure you've got a symbolic link from /dev/modem to /dev/cuaX
<--where your modem is
then make the script :
#!/bin/sh
pppd connect 'chat -v "" ATZ OK ATDTtelephonenumber CONNECT "" ' \user
username modem defaultroute \ crtscts /dev/modem 38400 asyncmap 0
put it in /etc/ppp or something.
make is executable using chmod +x /etc/ppp/scriptname
------------------------------
From: "jim" <[EMAIL PROTECTED]>
Subject: Newbie / hosts file / domain name?
Date: Fri, 15 Jan 1999 21:32:16 -0800
I'm setting up TCP/IP on a home network with 2 Win95 machine and a
Linux box. I won't be connected to any other network, except when one of
the Win95's uses dial-up networking.
OK, I've figured out what addresses I can use for this type of private
network (I'm gonna use 192.168.1.1/2/3). Everything works OK, pings OK all
around.
Now I'm writing a hosts file. I know the names of the hosts, but the
docs suggest that I need a domain name. What sort of generic domain name
should I use? "localdomain"?
Thanks
-Jim McQueen
------------------------------
From: Jim Grusendorf <[EMAIL PROTECTED]>
Subject: Problems running dhcpcd for multiple interfaces
Date: Sat, 16 Jan 1999 05:35:42 GMT
I'm on my way to installing an IP masquerade firewall for the first
time, using RedHat 5.2, 2.0.36 kernel. Right now, I'm just trying to
get both ethernet interfaces working. Both must use DHCP-assigned
addresses.
Modules are being successfully loaded for both interfaces (eth0=ne,
eth1=tulip). During boot, DHCP works for eth0, but fails for eth1. It
doesn't matter which network is connected to which interface, and gives
the same result if I make the tulip eth0 and the ne eth1. If, after
logging in, I run "dhcpcd -d eth1", I get the error
"bind(OpenSendSocket) Address already in use".
I'm assuming dhcpcd should handle multiple interfaces -- am I right?
Let me know if you need more information. Thanks!
Jim Grusendorf
[Sending unsolicited advertising to any address in the northmount.net
domain constitutes agreement to pay the recipient a $50 US handling fee
for each such message, due immediately upon receipt of such message.]
------------------------------
** 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
******************************
