Linux-Networking Digest #816, Volume #10 Sun, 11 Apr 99 00:13:53 EDT
Contents:
Re: PPP and chat script problem (ts)
Re: New install questions ("Eugene")
Interesting Mail Server Setup ("Smith")
Re: Cable Modem? Distributing over small network. ("Eugene")
Linux as router? (Meade)
pppd 2.3.5 idle timeout: (Adrian Chiang)
Is current network load info available? ("Dr. Douglas C. MacKenzie")
Re: 100baseT 3Com adapter problem - works fine under Windoz but not (Peter Eddy)
Re: Free CD of ur choice for curing my PPP blues (jason)
Re: IP Masq + Windows Networking Angst (Luca Filipozzi)
----------------------------------------------------------------------------
From: ts <[EMAIL PROTECTED]>
Subject: Re: PPP and chat script problem
Date: Sat, 10 Apr 1999 22:11:09 -0400
Ryan,
Your problem appears to be PAP/CHAP related. Check out the url below and read
this ppp set-up. If you follow it straight through (don't skip anything!!) you
should be connected in no time.
http://axion.physics.ubc.ca/ppp-linux.html
Regards,
ts
Lee Howes wrote:
> Please E-mail replies to [EMAIL PROTECTED], not reply to this message.
>
> This is my log of the communication with virgin net for my ppp connection.
> This clearly does not work, do you have any idea what addition could make
> this work. (NOTE: password and username converted to dashes)
> Apr 10 09:41:53 localhost pppd[668]: pppd 2.3.5 started by root, uid 0
> Apr 10 09:41:54 localhost chat[669]: send (ATDT0645505400^M)
> Apr 10 09:41:54 localhost chat[669]: expect (505400)
> Apr 10 09:42:13 localhost chat[669]: ATDT0645505400
> Apr 10 09:42:13 localhost chat[669]: -- got it
> Apr 10 09:42:13 localhost chat[669]: send (^M)
> Apr 10 09:42:13 localhost chat[669]: expect (ogin:)
> Apr 10 09:42:33 localhost chat[669]: ^M^M
> Apr 10 09:42:33 localhost chat[669]: login:
> Apr 10 09:42:33 localhost chat[669]: -- got it
> Apr 10 09:42:33 localhost chat[669]: send (--user--^M)
> Apr 10 09:42:33 localhost chat[669]: expect (assword:)
> Apr 10 09:42:33 localhost chat[669]: ^M
> Apr 10 09:42:53 localhost chat[669]: login: --user--^M
> Apr 10 09:43:03 localhost chat[669]: Password:
> Apr 10 09:43:03 localhost chat[669]: -- got it
> Apr 10 09:43:03 localhost chat[669]: send (--password--^M)
> Apr 10 09:43:03 localhost pppd[668]: Serial connection established.
> Apr 10 09:43:13 localhost pppd[668]: Using interface ppp0
> Apr 10 09:43:13 localhost pppd[668]: Connect: ppp0 <--> /dev/ttyS2
> Apr 10 09:43:13 localhost pppd[668]: sent [LCP ConfReq id=0x1 <magic
> 0x9a42a177> <pcomp> <accomp>]
> Apr 10 09:43:40 localhost last message repeated 9 times
> Apr 10 09:43:43 localhost pppd[668]: LCP: timeout sending Config-Requests
> Apr 10 09:43:43 localhost pppd[668]: Connection terminated.
> Apr 10 09:43:43 localhost pppd[668]: Receive serial link is not 8-bit
> clean:
> Apr 10 09:43:43 localhost pppd[668]: Problem: all had bit 7 set to 0
> Apr 10 09:44:03 localhost pppd[668]: Hangup (SIGHUP)
> Apr 10 09:44:03 localhost pppd[668]: ioctl(TIOCSETD, N_TTY): Interrupted
> system call
> Apr 10 09:44:04 localhost pppd[668]: Exit.
>
> Thank you, Lee Howes
------------------------------
From: "Eugene" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.redhat
Subject: Re: New install questions
Date: Sun, 11 Apr 1999 02:23:47 GMT
> My hardware:
> Celeron 300A oc'd to 450
> 256Meg PC100 RAM
> Adaptec 2940 running:
> 2xQuantum Furball 2.G(~600meg partition awaiting the Mandrake CD)
> Toshiba 6.7x CD
> Philips 40x CD
> I may re-install a TEAC 600meg tape drive, prolly not...
> SB AWE 32/8megs
> USR Sport 33k6 internal, model 1171
> Matrox G200/8megs
> 1-Linksys PCI NIC
> 1-Gallant PCI NIC
> K-board, Logitech marble mouse.
looks fine to me... vvveery fine... (esp. compared to my crappy Cyrix
200...)
(as long as the modem is not a winmodem)
> 1) Will ~600meg partition be enough to carry all the networking ware,
> plus a suitable combo of apps?
depends on how much you want to install... default install is about 300 Mb
> 2) Is all the hardware compatable?
see above
> 3) I currently IP Masquerade the Net to one other system in the house.
I currently IP masquarade using the standard Linux kernel. Read the howto at
www.linux.org/help Oh yeah, also read cable modem howto
(btw, here's my box: www.happypenguin.dhs.org)
> 4) I currently run a mail server app here, Hethmon Bros INETMAIL, by
I currently run smail. There's also sendmail (most popular), qmail, etc.,
etc.
> with that replacement, but can Apache also run SMTP/POP3, and how
> difficult is that setup?
uhhm, Apache is a web server, not a pop/smtp server.
there's plenty of imap, pop and smtp servers for Linux. Just search
freshmeat.net
> 5) I am not a gamer, but I still enjoy some older DOS games, like
> DIGDUG(83), Tristan Pinball(93) and Bycicle Bridge(91) etc. How good
> is Mandrake's DOSEMU these days, and will it be able to handle DOS
> apps on HPFS drives?
never tried that, so I've no idea.
> I have lots other questions, of course, and I'd very much appreciate
> if somone could forward the SUB address for a good List that handles
> installation/setup questions.
www.linux.org/help
read installation howto.
--
"Ein Volk, ein Reich, ein Fuhrer" - Adolf Hitler
"One World, one Web, one Program" - Microsoft's slogan
------------------------------
From: "Smith" <[EMAIL PROTECTED]>
Subject: Interesting Mail Server Setup
Date: Sun, 11 Apr 1999 10:21:10 +0800
Hi there,
I am trying to setup 2 linux servers at different locations to perform the
below mentioned function. Windows 98 will be use as clients to connect to
the linux server at each end and Outlook express (configured as IMAP and
SMTP or POP and SMTP) will be run on the Windows 98 boxes.
SETUP :-
Windows 98 A =====Linux Server A -------PPP Link---------Linux Server
B=======Windows 98 B
|-------------Location A----------------------| Dial thru telephone
|---------------Location B ---------------------|
I would like Users at Windows 98 A to send mails out to Users at
Windows 98 B and vice versa. Both Linux server must be able to accept mails
from the other users (meaning Server A must receive mails from Users B) and
store them onto the appropriate user mailbox or to pre-defined user at
either end so that they can receive it.
I know I need to setup both the Linux Server as PPP server and Dial On
Demand server. What else is required and where can I find these information
??
Do I need to run a DNS ??
Other than Sendmail, what else can I use?
I'm running Redhat 5.1, is there anything else I must take care of ??
Thanks to everyone who answered in advance.
------------------------------
From: "Eugene" <[EMAIL PROTECTED]>
Subject: Re: Cable Modem? Distributing over small network.
Date: Sun, 11 Apr 1999 02:00:13 GMT
go to www.linux.org/help and read cable modem howto and IP masquerading
howto.
here's my cable modem box: www.happypenguin.dhs.org
--
"Ein Volk, ein Reich, ein Fuhrer" - Adolf Hitler
"One World, one Web, one Program" - Microsoft's slogan
Peter Eacmen <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> If I were to get a cable modem would it be compatible with Linux? I
> have a small network, all of nodes have an 10Mbps ISA Ethernet card. I
> have a 5-port Ethernet HUB and Cat-5 cabling. I'm running a Samba
> server with two Win95 nodes. How would I be able to set up the cable
> modem so that all nodes on the network could access the Internet. Would
> I need to buy a more expensive switch. Could my Linux box act as a
> gateway to the Cable network?
> Has anyone done this?
>
> Thanks,
> Peter
>
------------------------------
From: [EMAIL PROTECTED] (Meade)
Subject: Linux as router?
Date: Sun, 11 Apr 1999 02:38:05 GMT
I've got a small lan, 3 or 4 windows pcs, that need to share
one "Mediaone Express" connection. A Cisco 2500 would be nice,
but the only thing that's available is an old 486 and Slackware
3.2
The question: Before I waste my time trying, does this machine
realistically have enough computing power to do the job of a
router?
Specifics:
486 dx2-66, 12 mb ram, 540 mb hdd
isa slots only;
aha-1522 scsi
nics/ dec etherworks-3 &
amd 1500/2100t
Job req'd/
pick up an ip adr from mediaone's DHCP server and act
as gateway for the other machines, which will all be
using fixed addresses on 10 mbits. Firewall functionality
a plus.
So, any opinions on whether this is enough machine for the
job?
------------------------------
Date: Sun, 11 Apr 1999 13:22:06 +1000
From: Adrian Chiang <[EMAIL PROTECTED]>
Crossposted-To: aus.computers.linux,comp.os.linux.setup
Subject: pppd 2.3.5 idle timeout:
Hello,
I just got a new dial up account with One-Tel, and am using pppd's
dial-on-demand
as described in this document:
http://klug.armintl.com/tech/pppd-demand-dialing.txt
I got it connecting and IP masquerading working happily but since i have
limited time
before getting charged xtra I would like to set an idle timeout.
The problem is that I set the idle timeout to 50 (seconds) but it never
disconnects. I tried turning off
sendmail, all the networked computers to prevent anything keeping the
link alive. Still doesn't disconnect.
I ran tcpdump on eth0 & ppp0 and no packets were going through them.
Here is my settings for /etc/ppp/options:
demand
idle 50
/dev/ttyS1 115200
connect '/usr/sbin/chat -v -f /sbin/unicom.chat'
defaultroute
ipcp-accept-remote
ipcp-accept-local
holdoff 10
and my /sbin/unicom.chat script:
TIMEOUT 5
"" ATZ
"" ATL0
OK ATDT92710000
ABORT "NO CARRIER"
ABORT BUSY
ABORT "NO DIALTONE"
ABORT WAITING
TIMEOUT 45
CONNECT ""
TIMEOUT 5
"sername:" mylogin
"assword:" mypasswd
"" "\d\d\d\d1"
Has anyone had similar trouble?
Thankyou for reading,
Adrian.
------------------------------
From: "Dr. Douglas C. MacKenzie" <[EMAIL PROTECTED]>
Subject: Is current network load info available?
Date: Sat, 10 Apr 1999 23:08:58 -0400
I'm working on some load balancing code and am wondering
if the current network load is available somewhere.
Something like the processor load average that the "uptime" command
provides, but for specific network adapters.
Thanks,
Doug
--
Douglas C. MacKenzie, Ph.D.
Mobile Intelligence Corporation
7963 Pine Hollow, Alto, Michigan 49302-9724
mailto:[EMAIL PROTECTED]
http://www.mobile-intelligence.com
------------------------------
From: Peter Eddy <[EMAIL PROTECTED]>
Subject: Re: 100baseT 3Com adapter problem - works fine under Windoz but not
Date: Sat, 10 Apr 1999 22:28:04 -0400
I remember seeing a few threads on this subject, the 3c905b cards, in
DejaNews recently. Try a search there.
Renaud wrote:
>
> Can anyone help?
>
> I have the following setup:
> - 3Com 3c905 B 100Mb/s ethernet adapter
> - Newly installed RedHat Linux 5.2
> - 4 port 100Mb/s RJ45 hub
> - staright and twisted network cables
> - all network parameters configured properly (TCP/IP, fixed IP
> address, gateway, DNS etc)
>
> The card works properly under win98
> The card works properly in 100Mb/s mode when I connect it to another
> PC using a crossed twisted pairs cable
> but the card does not work when I connect it (with a straight twisted
> pairs cable) to my 100Mb/s (only) hub.
> It seems that the cards sets itself to 10Mb/s (which does not work on
> my hub) when I connect it to it.
>
> Cables have been tested and are fine, the hub always worked and still
> does, but I cannot get the damned 100Mb/s link led on when I connect
> to it.
>
> I don't know if I make any sense, but if someone knows how to force
> the card to put itself in 100Mb/s mode at boot, then I will be happy.
>
> Thank you,
>
> Renaud.
>
> ------------------
> report from ifconfig:
> eth0 Link encap:Ethernet HWaddr 00:10:4B:2A:4B:A6
> inet addr:128.1.0.6 Bcast:128.1.0.255 Mask:255.255.255.0
> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> RX packets:1065 errors:0 dropped:0 overruns:0 frame:0
> TX packets:1328 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0
> Interrupt:9 Base address:0xf600
> ------------------
> exerpt from dmseg:
> 3c59x.c:v0.99E 5/12/98 Donald Becker
> http://cesdis.gsfc.nasa.gov/linux/drivers/vortex.html
> eth0: 3Com 3c905 Boomerang 100baseTx at 0xf600, 00:10:4b:2a:4b:a6,
> IRQ 9
> 8K word-wide RAM 3:5 Rx:Tx split, NWay Autonegotiation interface.
> MII transceiver found at address 24, status 786b.
> Enabling bus-master transmits and whole-frame receives.
>
>
------------------------------
From: jason <[EMAIL PROTECTED]>
Subject: Re: Free CD of ur choice for curing my PPP blues
Date: Sat, 10 Apr 1999 23:15:47 -0400
This won't cure all of your problems, but be aware that some hosts are
configured *not* to respond to pings, e.g. erols.com. From my vantage
point, it appears that redhat.com and www.redhat.com are suchly
configured ( I tried pinging them from my dialup and from my comp sci
account, and got nothing, but www.redhat.com loads just fine).
I'm not much qualified to help otherwise... have you tried posting in
comp.protocols.ppp? They were very helpful over there when I needed it
most.
-jason
MalayJ wrote:
>
> Folks:
>
> I have spent enough time on this that I am willing to buy someone a CD of
> their
> choice to solve it. I have read all the HOW-TO's and tried all suggestions
> offered so far - no luck.
>
> Please see end of the mail for 'reward' conditions.
>
> Summary of my problem
> ---------------------
> - Can successfully ping my dns server but I ping anything else there is no
> response.
> - Name resolution works: if I type in redhat.com, its IP address will show
> up.
> - If i remove the default route option from pppd then, i get "network not
> reachable"
> when I try to ping anything other than my name server.
>
> All relevant information and outputs are given below.
>
> System configuration
> --------------------
> - Direct modem dialup connection.
> - There is an ethernet card on the NT side of my system but it is not
> configured under linux.
> - There is an adaptec Ultra2 scsi card in system which not recognized or
> used
> via linux. I am using an IDE drive.
> - Running Redhat 5.2 out of the box on a Dell NT XP350, booting from a
> floppy.
> - From the NT side I can successfully operate PPP over same modem and IP
> addresses.
>
> /etc/ppp/options file contents: this file is used when I invoke pppd
> --------------------------------------------------------------------
> noipdefault
> /dev/ttyS0
> 38400
> lock
> crtscts
> asyncmap 0
> mtu 576
> mru 576
> defaultroute
>
> /etc/resolv.conf settings
> -------------------------
> domain ns.best.com
> nameserver 204.156.128.1
> nameserver 204.156.128.10
>
> /etc/sysconfig/network file contents
> ------------------------------------
> NETWORKING=yes
> FORWARD_IPV4="no"
> HOSTNAME-localhost.localdomain
> GATEWAY=""
> GATEWAYDEV=""
>
> My logon process
> ----------------
> 1. I run minicom and login into my PPP account via the modem.
> 2. Quit minicom without a reset (q) option
> 3. run pppd
>
> AFTER CONNECTION IS ESTABLISHED I GET THE FOLLOWING OUTPUTS:
>
> ifconfig
> --------
> lo Link encap:Local Loopback
> inet addr: 127.0.0.1 Bcast:127.255.255.255 Mask:255.0.0.0
> UP BROADCAST LOOPBACK RUNNING MTU:3584 Metric: 1
> ...<packet info stuff all set to 0>...
>
> ppp0 Link encap: Point-to-Point Protocol
> inet addr:205.149.171.120 P-t-P:204.156.128.1 Mask: 255.255.255.0
> UP POINTOPOINT RUNNING MTU:576 Metric: 1
> RX packets:5 errors:0 dropped:0 overruns:0 frame:0
> TX packets:5 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0
> Memory:7181830-7181cc4
>
> route -n
> --------
> Kernel IP routing table
> Destination Gateway Genmask Flags Metric Ref Use Iface
> 204.156.128.1 0.0.0.0 255.255.255.255 UH 0 0 1 ppp0
> 127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo
> 0.0.0.0 204.156.128.1 0.0.0.0 UG 0 0 1 ppp0
>
> ping 204.156.128.1 (this is the name server and gateway address supplied by
> ISP)
> ----------------
> PING 204.156.128.1 (204.156.128.1): 56 data bytes
> 64 bytes from 204.156.128.1: icmp_seq=0 tt1=252 time=173.6 ms
> 64 bytes from 204.156.128.1: icmp_seq=1 tt1=252 time=150.0 ms
> 64 bytes from 204.156.128.1: icmp_seq=2 tt1=252 time=150.0 ms
> ...
>
> ping 205.149.171.120 (my local IP address from ifconfig above)
> -------------------------------------------
> PING 205.149.171.120 (205.149.171.120): 56 data bytes
> 64 bytes from 205.149.171.120: icmp_seq=0 tt1=64 time=0.1 ms
> 64 bytes from 205.149.171.120: icmp_seq=1 tt1=64 time=0.0 ms
> ...
>
> ping redhat.com
> ---------------
> PING redhat.com (207.175.42.154): 56 data bytes
>
> ...<<It just hangs here with no output till I hit Ctrl C>>... then it gives
> 65 packets transmitted, 0 packets received, 100% packet loss.
>
> ping to any other named or harcoded address has same hang.
> If I remove "defaultroute" from the options file and remake the connection,
> the ping to other addresses return:
> "Network not reachable"
>
> Please help!
>
> FREE CD conditons:
> ------------------
> I will buy a CD of your choice (<$15 before shipping) from Amazon and send
> it to you as a gift if:
> - You respond with a reply that fixes the problem.
> - If not, you should be willing to enter into a direct e-mail dialogue,
> where
> I can try your suggestions and let you know the results till the problem
> is fixed or you give up. I will try and deal with one person at a time if
> they are responsive.
>
> Thanks
> Malayj
------------------------------
From: [EMAIL PROTECTED] (Luca Filipozzi)
Subject: Re: IP Masq + Windows Networking Angst
Date: Sat, 10 Apr 1999 11:53:36 -0700
[This followup was posted to comp.os.linux.networking and a copy was sent
to the cited author.]
In article <7eo4ek$g7i$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] says...
>
> I have a question regarding IP Masquerading and Windows Networking.
> Let me explain the situation.
> Some of the diagrams below need to be viewed in a fixed-width font.
>
> My old setup was thus: a windows network run using TCP/IP only. I was
> in a domain called 'SCHOOL'
> and could see the other computers in the domain from Network
> Neighbourhood.
> My machine name was 'WS102', and I needed to log into the NT domain
> when windows started.
>
> ,-------------, ,----------------, ,------------------,
> | ws102 |--| rest of domain |--| proxy & internet |
> | [windows] | '----------------' '------------------'
> '-------------'
>
> Now that I have IP Masquerading setup the picture looks like this:
>
> ,-------------, ,-------------, ,----------------,
> ,------------------,
> | ws102 |--| linux box |--| rest of domain |--| proxy &
> internet |
> | [windows] | '-------------' '----------------'
> '------------------'
> '-------------'
>
> I hope those two are clear. The IP Masquerading works fine for ICQ,
> IRC, FTP, web browsing etc.
> The problem is that with the new setup I cannot see anything except
> myself in Network
> Neighbourhood. I used to be able to see about 40 other
> computers/printers. My network admin,
> while not helping me because it's against policy, says that they only
> use TCP/IP not NetBEUI or
> IPX or anything like that. (He's not allowed to help me, but doesn't
> mind if I get it working).
>
> Aside from not seeing the rest of my workgroup I cannot actually
> authenticate via the NT server
> in the domain. When I start up windows, and have to type in my
> username and password there is a
> pause before it says something along the lines of "Could not
> authenticate credentials with NT
> server. Access may be restricted to certain resources.". I wonder if
> these two are connected.
>
> I have wondered whether I need to use VPN or something like that?
>
> Any advice so I can see the rest of my domain, and/or authenticate
> would be most gratefully
> received.
>
> Regards,
> Ollie Cook
>
> Please remove the NXFHBT before replying by email.
>
> ***** Posted via the UK Online online newsreader *****
>
> Go to http://www.ukonline.co.uk to find out
> about other online services we offer our subscribers.
>
>
>
I apologize in advance to the Linux community. This reply is very MS
based.
OK... here's the scoop.
In an NT Domain, there are the following:
DomMB == Domain Master Browser
SegMB == Segment Master Browser
In the first diagram, you were part of the same segment as the rest of
the network. Your Windoze computer, on startup, broadcasts its existence
to the entire segment and the SegMB hears it. The SegMB returns to your
computer a complete list of machines that it knows about and this is what
you see in the Network Neighbourhood. If configured to use WINS, the
SegMB will know who the DomMB is (it's always a domain controller) and
will be able to get the browse lists for other segments (in a multi-
segment domain) by querying their SegMB's. In a single segment NT domain,
the SegMB and the DomMB will be the same computer.
In the second diagram, you have placed yourself on a separate segment
(Ethernet segment) without any other Windoze computers. Your computer, on
startup, broadcasts its existence and on finding no SegMB, makes itself
the SegMB for your segment (forces an "election"). This is why you see
yourself in the Network Neighbourhood. Now, your computer doesn't know
that there are more Windoze machines on the other side of the Linux box.
It can't find the DomMB for the browse list nor the PDC or BDC for
authentication.
You have two choices for rectifying this:
1) Add an LMHOSTS file to your computer that lists the Domain and the
domain controllers. You should be able to find an lmhosts.sam sample file
on your computer.
2) Specify a WINS. If your NT domain is using WINS (ask your admin), then
you can tell your computer to use it. The WINS database will tell your
computer who the domain controllers are for purposes of authentication
and should also tell your computer (who is the SegMB for your
segment) that there are other SegMB's that need to be queried so that
your Network Neighbourhood list fills up.
None of this has to do with Linux. Your linux box is acting as a router
(if you don't set up any firewall rules). Your system administrator
should be able to solve this issue... it appears all the time in multi-
segment or WAN NT networks where IP gets routed.
If you start adding firewalling rules, just make sure that you let ports
137 138 139 through. These are the nebios over TCP/IP ports and you
*must* let this traffic through.
(Which begs the question as to what you are using the linux box for?)
You have one other option (more complicated): you can use PPTP. You have
to recompile your Linux kernel with patches from John Hardin in order to
support this. Then your sysadmin needs to set up RAS (Remote Access
Server) and you need DUN (Dial Up Networking) with PPTP enabled. I won't
go into more details here.
Hope this helps,
Luca
--
Luca Filipozzi <[EMAIL PROTECTED]>
------------------------------
** 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
******************************
