Linux-Networking Digest #796, Volume #10 Fri, 9 Apr 99 01:13:38 EDT
Contents:
Re: Firewall setup w/DSL and multi IPs (Ron Watkins)
IBM Turbo 16/4 Token Ring ISA problems (Snow Leopard)
Re: Downloads hanging - please help! (Ron Watkins)
PPP script for earthlink ISP server (jongmook)
Re: Duplicate IPs PLUS duplicate MAC addresses, has this been done ? (Ron Watkins)
Re: Dial Up NetWorkking (Newbie) (Jon-o Addleman)
Re: Downloads hanging - please help! ("William Evans")
servlet <-> netscape produces lots of RST packets (Steve Fink)
Re: Dialing into Linux box (Det 1 SOG)
Re: Thanks + ???? ("William Evans")
Re: SERIAL Network Printer (JLocke1122)
Re: pppd demand dialing (Bob)
I only want DHCP Client (Daniel Nissim)
Re: cant get into linux server - whats ipc$ ("ephigenie")
Re: Duplicate IPs PLUS duplicate MAC addresses, has this been done ? (Stuart R.
Fuller)
Re: Dialing into Linux box (Stuart R. Fuller)
Re: Why FTP/Telnet connection to Linux box is very slow ? (Jim Roberts)
Re: new to linux how do I read howto/FAQ under linux ("Snoopy :-))")
----------------------------------------------------------------------------
From: Ron Watkins <[EMAIL PROTECTED]>
Subject: Re: Firewall setup w/DSL and multi IPs
Date: Thu, 08 Apr 1999 18:43:18 -0700
I was told on the SINUS firewall list that this can be done by proxying using
multiple addresses, and in fact was probably the best way to firewall a
network that is too small to subnet further (as yours is.) I have not,
however, actually started doing this myself yet and I don't really know how.
I'll be experimenting with it over the weekend -- if I have any success I'll
let you know.
<<RON>>
jaydub wrote:
>
> I have recently gotten DSL service with an 8 IP setup for
> four home machines. I want to setup one machine as
> a firewall. I have seen alot of comments regarding
> using masquerading for people with single IP. Would
> I require anything like that when all machine already
> have real IPs?
>
> I am running the 2.2.5 kernel so any responses that
> provide ipchains info would be helpful.
>
> Jeff Warrington
> [EMAIL PROTECTED]
------------------------------
From: Snow Leopard <[EMAIL PROTECTED]>
Subject: IBM Turbo 16/4 Token Ring ISA problems
Date: Thu, 08 Apr 1999 17:46:44 -0400
Hi,
I am getting the following errors on the login screen every 30 min - 1
hr.
tr0: Line errors 01, internal errors 00, Burst errors 00
A/C errors 00, Abort delimiters 00, Lost frames 00
Receive congestion count 00, Frame copied errors 00
Frequency errors FF, Token errors 00
I have read the mini-howto on token ring adapters, and I have set this
card to auto mode. I am using RH 5.2. tr0 is detected during
initialization. It's connected to a live LAN.
Any ideas/suggestions?
James
--
======================
Do you dream of cats? http://www.idreamofcats.com
Wanna send me e-mail? [EMAIL PROTECTED]
------------------------------
From: Ron Watkins <[EMAIL PROTECTED]>
Subject: Re: Downloads hanging - please help!
Date: Thu, 08 Apr 1999 18:35:16 -0700
Right now the problem could be in a number of places, so you'll have to do
some detective work. Some questions I'm having in looking at this:
1. Does it fail with all clients? Try ftp, ncftp, netscape. Does it
always
hang in the
same place on all three clients?
2. Does it fail when you are running as root? How about as a user?
3. Does it fail with the older Linux kernels? Try 2.0.36.
4. Can you try it with a different modem?
5. If you run it without sound and X (console mode only, as many programs
shut down as
possible) does it still fail in the same place?
6. Can you try from another phone line, preferably across town?
(reaching a
bit here :)
7. Can you download the same file under Windows? (preferably on the same
machine if poss.)
I have never heard of a problem like this so I really haven't the faintest
clue where to look -- it's just detective work until you isolate it. You've
got an odd one for sure :-)
<<RON>>
"Scott E. Regener" wrote:
>
> I can't begin to express how incredibly frustrating this is for me. I'm a
> Linux newbie who has accomplished some very basic things so far: I've
> recompiled a kernel, gotten X and sound to work on an IBM Thinkpad 310ED and in
> general been impressed with the stability and performance of Linux (RH 5.2) but
> I've run into a problem that no one in the comp.os.linux.setup newsgroup could
> help with: some files will not download, no matter what method I try to
> download them. Other files, often from the same site, download without any
> problems at all. Activity ceases on the download, but other TCP/IP packets fly
> right through. Can anyone help me?
>
> About my setup: Kernel 2.2.4 & 2.2.5 tried, PPP connection (in kernel), RedHat
> 5.2 for almost everything.
>
> One file that consistantly fails, regardless of the site I try to pull it from
> is kde-base-1.1-whatever.rpm. Hangs just after the 100K point. NCFTP reports
> 'Error: Connection timed out. Transfer aborted.' Netscape Communicator just
> reports "stalled" forever. KFM gives no error, but activity stops.
>
> If anyone has ideas, I'd love to hear them. I've tried looking for an updated
> PPPD but haven't the foggiest where to look beyond mirrors of redhat
> distributions (which have the exact same rpm that I have installed already.)
> Like I said, I'm a newbie. Please help!
>
> --
> Scott E. Regener
> DSS Since 7/95
> DVD Since 6/98
------------------------------
From: jongmook <[EMAIL PROTECTED]>
Subject: PPP script for earthlink ISP server
Date: Thu, 08 Apr 1999 20:36:16 -0500
Reply-To: 221-10, Arnold, Dr., W., Lafayette, IN47906
Does anybody have a working chat script for Earthlink ISP server ?.
I am struggling to get it working.
Thanks
Jongmook
------------------------------
From: Ron Watkins <[EMAIL PROTECTED]>
Subject: Re: Duplicate IPs PLUS duplicate MAC addresses, has this been done ?
Date: Thu, 08 Apr 1999 18:41:17 -0700
I'd be on the phone to your vendor bigtime. You CANNOT have identical MAC
addresses on the same network without a great heap of jiggery and foolery.
They screwed up badly; you'll need to get them to bail you out of the problem.
Chances are, if they all have identical MAC addresses, there's a method of
changing them.
<<RON>>
------------------------------
From: [EMAIL PROTECTED] (Jon-o Addleman)
Subject: Re: Dial Up NetWorkking (Newbie)
Date: Fri, 09 Apr 1999 02:21:32 GMT
[posted and mailed]
Once upon a Tue, 6 Apr 1999 11:06:43 -0500, "KOC TIN Y. GO"
<[EMAIL PROTECTED]> wrote:
>Howdy,
> I have AT&T WorldNet as an ISP and I having trouble connecting through Red
>Hat Linux 5.2. I figured that the chat file is not setup up properly, but
>I'm not so sure what to change. When I called through Minicom, it doesn't
>say Logon but instead Please Sign On(I think). After I entered my username,
>there is a line of garbage and I can't do anything. Any ideas? BTW, can
>you e-mail me instead at [EMAIL PROTECTED] since I don't visit these
>newsgroup that much(though I should). Thanks
The PPP how-to should help you out. It sounds like you have a fairly
normal ISP - i.e. no PAP or CHAP (though you can's be certain). It
explains how to set up the chat script. Feel free to mail me if you
want some more help.
--
Jon-o Addleman
------------------------------
From: "William Evans" <[EMAIL PROTECTED]>
Subject: Re: Downloads hanging - please help!
Date: 08 Apr 1999 23:35:23 -0400
>>>>> "Edward" == Edward Lee <[EMAIL PROTECTED]> writes:
Edward> I am also having problem downloading a 60 M file.
Edward> I got 27M, 53M and 58M (really close) before ftp quit.
Edward> Ftp was designed decades ago when people only download small files.
Edward> What we need is an incremental ftp protocol where both the server and
client
Edward> do checksum on the files and only send the mismatched blocks. The problem
Edward> is getting everybody to support this protocol.
ncftp -C ftp://some.where.com/path/to/file
"-C" forces a continuation, aka a "reget". It just continues from
where the last attempt left off.
I believe it's based somewhere off of www.ncftp.com ... call it a hunch.
Using ncftp, I came up with a small script:
=================================
PREFIX=""
KILLFILE="persistent-ftp.kill"
unset QUIETLY
while getopts ":hk:p:q" THISOPT ; do
case "$THISOPT" in
"h") echo "usage: `basename $0` [ -p prefix ] [ -hq ] [ -k killfile ] files" ;
exit ;;
"k") KILLFILE="$OPTARG" ;;
"p") PREFIX="$OPTARG" ;;
"q") QUIETLY=1 ;;
esac
done
shift $(($OPTIND - 1))
[ -f "$KILLFILE" ] && exit 1
for i in $* ; do
[ -z "$QUIETLY" ] && echo "$PREFIX$i"
if [ "${i#ftp://}" = "${i}" ]; then # doesn't start with ftp://,
# so use the PREFIX
name="$PREFIX$i"
else # starts with ftp://, so
# don't use the PREFIX
name="$i"
fi
while ! ncftp -r -d60 -C $name ; do
[ -f "$KILLFILE" ] && exit 1
sleep 5
done
done
=================================
You can name it whatever you want and change the defaults ... I just
like using it when I know that a site will most likely drop me
somewhere in the middle. To create a "shopping list", I just put the
filenames (paths as appropriate) in a file and do:
persistent-ftp `cat shopping-list-file-name`
*shrug* It works for me. Of course, it opens and closes a single
connection for each file, even though most of them are most likely on
the same server. *eh* I usually leave this unattended, anyway.
(BTW: the '-r -d60' in the script means "if the connection fails, wait
60 seconds and retry" ad infinitum. You can change this to whatever
you like, really, but for the sake of the sitemasters, don't flood
them with connection attempts just because they're already full of
anonymous users.)
If you prefer a graphical front-end, I just recently saw an
announcement for GTM, or GTransferManager:
GTransferManager allows the user to retrieve multiple files
from the web. These files can be retrieved in multiple parts
and each part retrieved on a separate session that the user is
connected to the Internet. This is most useful to users with
dialup connections. The program performs this tasks using wget
as it's back-end.
It requires, among other things, wget and GNOME, so if you aren't
running them, it's a whole lot to download and install just for the
pretty front-end. (Okay, wget isn't that much, but GNOME is a hefty
download ... and getting it may be another problem that you're trying
to fix with this! ;-)
Anyway, you can find it at http://camoes.rnl.ist.utl.pt/~bapm/gtm/
HTH
-bill
--
William Evans < william . evans @ computer . org >
------------------------------
From: Steve Fink <[EMAIL PROTECTED]>
Subject: servlet <-> netscape produces lots of RST packets
Date: Fri, 09 Apr 1999 02:56:06 GMT
I am running the servletrunner that comes with Sun's JSDK2.0, talking to a
Netscape 4.51 client. I keep getting error windows popping up: "Connection
reset by peer" in netscape. It doesn't always happen, and doesn't always
happen at the same point in the communication. I hate sporadic failures.
This is happening with kernel versions 2.2.3, 2.2.4, 2.2.5, and 2.2.5ac4. With
2.2.2ac7, which I've been told solved some fairly common networking problem or
other, I am still seeing it but much less commonly. With 2.2.0pre7, this never
happened.
The connection is to localhost:8080, so it's just going over the loopback
interface. When I do a tcpdump, it looks to my uneducated eye like *many* of
the connection closes are screwed up, and the fact that I'm only seeing the
error window once in a while is some bizarreness in Netscape.
Each connection is established (SYNs sent and ACKed both ways), the servlet
and netscape talk to each other happily, then netscape sends a FIN, the
servlet ACKs, the servlet pushes a little more data across, and netscape
barfs back a RST. Or the two roles are reversed. In fact, the roles reverse
every time this happens, but there are successful communications interspersed
(both sides close correctly by sending FINs back and forth and ACKing each
other.) Now, in the naive model of TCP networking that I keep in my head
whenever the cockroaches aren't making too much noise up there, this should
only happen when the recipient of the FIN packet doesn't bother to read out
all the data before closing its side of the connection. So yes, both my
servlet's and netscape's code could be buggy, but (1) why didn't I see this
with 2.2.0pre7; (2) why does everything work smoothly when I use lynx instead
of netscape; and (3) why does it sometimes work smoothly with my usual kernel
v2.2.2ac7?
For one of the problematic runs, here's the beginning of the output of
tcpdump -i lo 2>&1 >dump.tcp.bug.3
grep 8080 dump.tcp.bug.3 | perl -ne 's/localhost\.8080/servlet/g;
s/localhost\.(\d+)/$1/; if (defined $p && $p != $1) { $p = $1; print "\n"; }
elsif (!defined $p) { $p = $1; }; print if ($x ^= 1)'
given that servletrunner uses port 8080 (the above perl hack just renames
localhost.8080 => servlet and localhost.nnnn => nnnn, puts a blank line
between tcpdump output for different ports, and suppresses every other line
of output because I didn't want to bother to look up how to tell tcpdump to
only give me incoming or outgoing packets):
20:43:51.274085 2155 > servlet: S 2966969:2966969(0) win 31072 <mss
3884,sackOK,timestamp 8002830 0,nop,wscale 0> (DF) 20:43:51.274152 servlet >
2155: S 4293852607:4293852607(0) ack 2966970 win 31072 <mss
3884,sackOK,timestamp 8002830 8002830,nop,wscale 0> (DF) 20:43:51.274184 2155
> servlet: . ack 1 win 31072 <nop,nop,timestamp 8002830 8002830> (DF)
20:43:51.275583 2155 > servlet: P 1:358(357) ack 1 win 31072
<nop,nop,timestamp 8002830 8002830> (DF) 20:43:51.275620 servlet > 2155: .
ack 358 win 30715 <nop,nop,timestamp 8002830 8002830> (DF) 20:43:51.662513
servlet > 2155: P 1:109(108) ack 358 win 31072 <nop,nop,timestamp 8002869
8002830> (DF) 20:43:51.662581 2155 > servlet: . ack 109 win 30964
<nop,nop,timestamp 8002869 8002869> (DF) 20:43:51.668378 servlet > 2155: P
109:751(642) ack 358 win 31072 <nop,nop,timestamp 8002869 8002869> (DF)
20:43:51.670583 2155 > servlet: . ack 751 win 31072 <nop,nop,timestamp
8002870 8002869> (DF) 20:43:51.737098 servlet > 2155: F 751:751(0) ack 358
win 31072 <nop,nop,timestamp 8002876 8002870> (DF) 20:43:51.737162 2155 >
servlet: . ack 752 win 31072 <nop,nop,timestamp 8002876 8002876> (DF)
20:43:51.737534 2155 > servlet: F 358:358(0) ack 752 win 31072
<nop,nop,timestamp 8002876 8002876> (DF) 20:43:51.737567 servlet > 2155: .
ack 359 win 31072 <nop,nop,timestamp 8002876 8002876> (DF)
20:43:52.722974 2156 > servlet: S 5205215:5205215(0) win 31072 <mss
3884,sackOK,timestamp 8002975 0,nop,wscale 0> (DF) 20:43:52.723040 servlet >
2156: S 8384749:8384749(0) ack 5205216 win 31072 <mss 3884,sackOK,timestamp
8002975 8002975,nop,wscale 0> (DF) 20:43:52.723071 2156 > servlet: . ack 1
win 31072 <nop,nop,timestamp 8002975 8002975> (DF) 20:43:52.723853 2156 >
servlet: P 1:428(427) ack 1 win 31072 <nop,nop,timestamp 8002975 8002975>
(DF) 20:43:52.723887 servlet > 2156: . ack 428 win 30645 <nop,nop,timestamp
8002975 8002975> (DF) 20:43:52.725768 2156 > servlet: P 428:497(69) ack 1 win
31072 <nop,nop,timestamp 8002975 8002975> (DF) 20:43:52.730580 servlet >
2156: . ack 497 win 31072 <nop,nop,timestamp 8002976 8002975> (DF)
20:43:52.730645 2156 > servlet: P 497:525(28) ack 1 win 31072
<nop,nop,timestamp 8002976 8002976> (DF) 20:43:52.740592 servlet > 2156: .
ack 525 win 31072 <nop,nop,timestamp 8002977 8002976> (DF) 20:43:52.786284
servlet > 2156: P 1:247(246) ack 525 win 31072 <nop,nop,timestamp 8002981
8002976> (DF) 20:43:52.786333 2156 > servlet: . ack 247 win 30826
<nop,nop,timestamp 8002981 8002981> (DF) 20:43:52.786916 2156 > servlet: F
525:525(0) ack 247 win 31072 <nop,nop,timestamp 8002981 8002981> (DF)
20:43:52.786953 servlet > 2156: . ack 526 win 31072 <nop,nop,timestamp
8002981 8002981> (DF)
20:43:52.787618 2158 > servlet: S 1505501:1505501(0) win 31072 <mss
3884,sackOK,timestamp 8002981 0,nop,wscale 0> (DF) 20:43:52.787669 servlet >
2158: S 3505447:3505447(0) ack 1505502 win 31072 <mss 3884,sackOK,timestamp
8002981 8002981,nop,wscale 0> (DF) 20:43:52.787697 2158 > servlet: . ack 1
win 31072 <nop,nop,timestamp 8002981 8002981> (DF) 20:43:52.788343 2158 >
servlet: P 1:438(437) ack 1 win 31072 <nop,nop,timestamp 8002981 8002981>
(DF) 20:43:52.788373 servlet > 2158: . ack 438 win 30635 <nop,nop,timestamp
8002981 8002981> (DF)
20:43:52.791523 servlet > 2156: P 247:446(199) ack 526 win 31072
<nop,nop,timestamp 8002982 8002981> (DF)
20:43:52.791572 2156 > servlet: R 5205741:5205741(0) win 0
20:43:52.971157 servlet > 2158: P 1:109(108) ack 438 win 31072
<nop,nop,timestamp 8003000 8002981> (DF) 20:43:52.971220 2158 > servlet: .
ack 109 win 30964 <nop,nop,timestamp 8003000 8003000> (DF) 20:43:52.976764
servlet > 2158: P 109:1151(1042) ack 438 win 31072 <nop,nop,timestamp 8003000
8003000> (DF) 20:43:52.980598 2158 > servlet: . ack 1151 win 31072
<nop,nop,timestamp 8003001 8003000> (DF) 20:43:53.035171 servlet > 2158: F
1151:1151(0) ack 438 win 31072 <nop,nop,timestamp 8003006 8003001> (DF)
20:43:53.035234 2158 > servlet: . ack 1152 win 31072 <nop,nop,timestamp
8003006 8003006> (DF) 20:43:54.085509 2158 > servlet: P 438:883(445) ack 1152
win 31072 <nop,nop,timestamp 8003111 8003006> (DF) 20:43:54.085575 servlet >
2158: R 3506599:3506599(0) win 0
Please CC me on replies, unless your religion forbids it. I read the
newsgroup, but I'll get responses faster.
A hopeful thanks in advance,
Steve
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: Det 1 SOG <[EMAIL PROTECTED]>
Subject: Re: Dialing into Linux box
Date: Fri, 09 Apr 1999 02:56:16 GMT
Thanks Peter
I checked to make sure the default was still there.
Mgetty answers on first ring, it acts like it's doing something, then drops
the line before I get to the login prompt.
I just don't know where to look now.
I'm sure it's something real simple, that I have overlooked. I've read Mick
Dennis's piece on Auto PPP. the Mgetty Info page, the Mgetty man page,
Getty_ps documentation, login documentation, init documentation. Etc
Help!!!
Thanks
Bob M.
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: "William Evans" <[EMAIL PROTECTED]>
Subject: Re: Thanks + ????
Date: 08 Apr 1999 23:41:40 -0400
>>>>> "Mike" == Mike <[EMAIL PROTECTED]> writes:
Mike> When i entera url into the location area all i get returned
Mike> is an error message that home.netscape.co cannnot be locted,
Mike> it does not have a DNS entered.
Aha, you've hit on it there. You apparently don't have DNS
configured. This is important.
Two solutions:
{Longer} Setup and run a caching-only name server that forwards
request to a known name server or to the root nameservers; or
{Short} Add "nameserver w.x.y.z" to /etc/resolv.conf, where w.x.y.z
is the IP address (not name) of your ISP's name server. This will
effect changes immediately (that is, no need to reboot or bring
down/up the link). (You could also change the "search" parameter, but
that isn't *crucial* to getting it working.)
Actually, setting up a caching-only name server isn't that hard, but
it does take a little maintenance, and I realize that you just want to
get this thing working first of all.
HTH
-bill
--
William Evans < william . evans @ computer . org >
------------------------------
From: [EMAIL PROTECTED] (JLocke1122)
Subject: Re: SERIAL Network Printer
Date: 9 Apr 1999 04:33:25 GMT
I have tried the setserial, but still no go..The printer works fine.. That I
know...I even changed the printer settings to the default of setserial.. Still
nothing...
Any Ideas anyone?
------------------------------
From: [EMAIL PROTECTED] (Bob)
Subject: Re: pppd demand dialing
Date: Thu, 8 Apr 1999 20:16:14 -0600
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
> Would appreciate any help...
>
> I have an RH5.2 (2.0.36) system running very nicely on a 586 running
> as a fax gateway ttyS02 email tx/rx and ppp dialin on ttyS3, I would
> like to
> setup demand dialing on ttyS01 using pppd - is this possible ? I have
> read the hands-on how-to
> and it doesn't seem to suggest how to specify which modem pppd is to use
> !
>
> Any help would be greatly appreciated
>
> TIA Frank.
>
>
>
Frank,
Dug though a bunch of old notes to find this great resource. This will do
it for you ( it did me)... http://www.brasscannon.com/Linux/
BTW > you'll need to download a source upgrade for pppd ( can't remember
where off hand - check the ftp sites) , patch the source, rebuild the
pppd stuff, recomiled the kernel. This will give you a kernel with pppd
that supports demand dialing.
HTH,
Bob
------------------------------
From: Daniel Nissim <[EMAIL PROTECTED]>
Subject: I only want DHCP Client
Date: Fri, 09 Apr 1999 02:10:11 GMT
I have a linux box that needs to be a DHCP client only. How do I remove
only the server part ? I am running Redhat 5.2.
Thanks
[EMAIL PROTECTED]
------------------------------
From: "ephigenie" <[EMAIL PROTECTED]>
Subject: Re: cant get into linux server - whats ipc$
Date: Thu, 8 Apr 1999 01:17:52 +0100
seems to me that your win98 box can't give your linux box the correct
passwd.
try the patch for win95/98 for getting access, if you don't wan't to
recompile samba. just put into your registry.
begin 666 win95_plainpassword.reg
M4D5'141)5#0-"@T*6TA+15E?3$]#04Q?34%#2$E.15Q3>7-T96U<0W5R<F5N
M=$-O;G1R;VQ3971<4V5R=FEC97-<5GA$7%9.151355!=#0HB16YA8FQE4&QA
@:6Y497AT4&%S<W=O<F0B/61W;W)D.C P,# P,# Q#0H`
`
end
------------------------------
From: [EMAIL PROTECTED] (Stuart R. Fuller)
Subject: Re: Duplicate IPs PLUS duplicate MAC addresses, has this been done ?
Reply-To: [EMAIL PROTECTED]
Date: Fri, 09 Apr 1999 04:00:02 GMT
Judge (postmaster@localhost) wrote:
: This is a multi-part message in MIME format.
:
: ------=_NextPart_000_017B_01BE8132.B41B79C0
: Content-Type: text/plain;
: charset="iso-8859-1"
: Content-Transfer-Encoding: quoted-printable
:
: Hi,
: I'm living an installation nightmare at the moment.
:
: We have a network of thin-client netstations with an RH5.1 =
: proxy/bootp/nfs/everything server at it's center.
:
: The netstations were all shipped out with identical MAC addresses, =
: biiigg boo-boo. We've tried manually assigning different IPs to the =
: stations, bypassing part of the requirement for a bootp server, but =
: we're still running into problems when the proxy (squid BTW) is handling =
: web traffic.
:
: My question is, what if I assign the *same* IP address to each =
: netstation ? They already have identical MAC addresses. I'm hoping the =
: stations will start rejecting packets further up the IP stack, oops, =
: wrong sequence number, bye-bye.
: Obviously this will flog the network and the CPUs of the stations, but =
: will it work ?
The bottom line is that 2 machines having the same MAC address violates the
way that Ethernet works. Until each machine has a unique MAC address, you can
pretty much forget about getting it to work. You might be able to get some
hack going (not very likely), but even if you did, it'll be a support
nightmare.
I'd like to know how it is that each machine has the same MAC address? I
thought that they were supposed to ALL be unique?
Stu
------------------------------
From: [EMAIL PROTECTED] (Stuart R. Fuller)
Subject: Re: Dialing into Linux box
Reply-To: [EMAIL PROTECTED]
Date: Fri, 09 Apr 1999 04:00:02 GMT
Det 1 SOG ([EMAIL PROTECTED]) wrote:
: In article <01be77dd$d8cf61e0$227bcdd0@PANTHER>,
: "Russell S. DiPesa" <[EMAIL PROTECTED]> wrote:
: > To all,
: > I have RH5.2 and am trying to dial into the modem that I have connected
: to
: > it. I have added the necessary line to "inittab" to respawn mgetty for
: > each new call. I am able to dial the modem and get a handshake from it.
: > But, after attempting to negotiate the username and password, it
: > disconnects. I believe I have the proper permissions set for the username
: > that I am dialing in with. Any suggestions?
:
: I am having the same problem with mgetty. RH 5.2, I've been trying to solve
: this problem for approx 2 weeks, I have read every howto, man page, and news
: group and still don't have the solution.
Y'all need to post more information than "I have it set up correctly, but it
still doesn't work". It works for me, so therefore y'all don't have it set up
correctly!
Post the line you added to /etc/inittab. Here's mine:
# Start a getty on the modem line
m1:2345:respawn:/sbin/mgetty ttyS1 DT19200
The "DT19200" references a line in /etc/gettydefs.
If you still can't get it to work, look in /var/log/messages for relevant
messages. And, if you don't know what the messages mean, post 'em.
Stu
------------------------------
From: [EMAIL PROTECTED] (Jim Roberts)
Subject: Re: Why FTP/Telnet connection to Linux box is very slow ?
Date: Fri, 09 Apr 1999 02:49:14 GMT
> Hi,
>
> I had the same problem and resolved it the way you said (RH5.2). Is
> there anyway to get linux to use DNS instead of the hosts file? I'm
> using M$ NT4.0 DNS, but it resolves the names ok when I ping hosts.
>
> TIA,
>
SNIP
Yes. The file /etc/host.conf and resolv.conf control what is used for
name look-up and in what order. Check the man pages.
If you use NIS then the /etc/nsswitch.conf and resolv.conf control
lookup functions. (Unless you are using Solaris in which case, the
/etc/nsswitch.conf file is used in either case.)
--
Jim Roberts Never enough time!
[EMAIL PROTECTED]
------------------------------
From: "Snoopy :-))" <[EMAIL PROTECTED]>
Crossposted-To:
alt.linux,alt.linux.sux,alt.os.linux,alt.os.linux.caldera,comp.os.linux.hardware,comp.protocols.smb
Subject: Re: new to linux how do I read howto/FAQ under linux
Date: Fri, 09 Apr 1999 00:02:14 -0400
It's very easy. Use your Browser. (You don't have to be Connected to do
this)
I am assuming, that you have Netscape.
1. Open Netscape Navigaator +(Click) File Menu + Open Page
2. Type ( Case-Sensitive) "/usr/doc/HOWTO" <enter>
Note: if you have installed the "HOWTO in .html - Format then at the
bottom of the "Page" open the "other-formats" directory. OR type after
"/HOWTO/other-formats/"
You can read All the other Documents with your Browser, Just make sure you
type in the Corect PATH.
BTW: In the future avoid "Cross-Posting" to so many Groups, or you are asking
to be Flamed.
Good luck
Snoopy :-))
Tony Langdon wrote:
> It's 06 Apr 99 23:21,
> We'll return to [EMAIL PROTECTED] and All's
> discussion of new to linux how do I read howto/FAQ under linux
>
> ni> I'm new to Linux - how do I read the Howto/FAQ's under Linux ?
> ni> Please specify application & path (ie less /usr/doc/howto/xxxx.yyyy)
> ni> as I'm really raw & even if you don't know my system I'll work it out.
> ni> I installed a workstation version of Red Hat 5.2 in a machine with
> ni> Dos/win95 running on the primary partition.
>
> They are installed in many formats. Probably the most useful are text
> (may be gzipped text) and HTML. The HTML versions can be read with any
> web browser, which is probably the easiest way to go. Make sure you've
> installed the "Extra Documentation" package.
>
> ni> How do I check how my Linux partition has been sud divided ? ie I re
> ni> partitioned my Dos C drive & installed Linux in the new partition but
> ni> would like to see how RH set it up for my own interest/understanding.
>
> use fdisk and (p)rint the partition table. It will be displayed on the
> screen.
>
> .. Resistance Is Useless! (If < 1 ohm)
> --
> |Fidonet: Tony Langdon 3:633/284.18
> |Internet: [EMAIL PROTECTED]
> |
> | Standard disclaimer: The views of this user are strictly his own.
------------------------------
** 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
******************************
