Linux-Networking Digest #3, Volume #11 Sat, 1 May 99 07:13:43 EDT
Contents:
Re: Beowulf and Nuclear Testing Simulation (Jim Richardson)
ipchains help please (Anjan Sen)
10Base2 Connector/Cable Quality (Glenn Watson)
Re: Urgent... trust account problems (Benoit Gerrienne)
Re: Need help with PPP ("IBL")
Re: KPPP & IBM Global Network ISP problems (John Thompson)
Possible to use Linux for small firewall? (Miguel Cruz)
Dynamic PPP hostname changes ! (Neil Benson)
Re: DEC 21143 Chipset (Trever Adams)
Re: LPRng and a local printer (Holger Eitzenberger)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Jim Richardson)
Subject: Re: Beowulf and Nuclear Testing Simulation
Date: 1 May 1999 05:26:04 GMT
Reply-To: [EMAIL PROTECTED]
On Thu, 29 Apr 1999 21:33:32 GMT,
PG, in the persona of <[EMAIL PROTECTED]>,
brought forth the following words...:
>Why not ask the Chinese ? They seem to be interested in that sort of
>thing...
>(sorry... ;^) - just had to comment...)
>
>pg
>
>Jonathan Bazemore <[EMAIL PROTECTED]> wrote in message
>news:[EMAIL PROTECTED]...
>> Can the beowulf supercomputer system be used to run nuclear bomb
>> simulations?
>>
>> I was thinking about a 16 node (PII) cluster with a workstation, with an
>> estimated processing power of 1.4 gigaflops.
>>
>> Any comments or tips are welcome, thanks.
>>
>
>
nah, the chinese don't do beowulf, they just get Billy boy to
send 'em hardware, no, the _other_ billy boy, you know, the preside.. no,
the _other_ president, the one who was so embarassing on the witness stand,
no, the one who tries to bury his opponants under legal word games... no,
the other one, the one who buys rigged polls. nonono the _other_ one, the
one with... oh fergit it...
:)
--
Jim Richardson
www.eskimo.com/~warlock
All hail Eris
"Linux, where do you want to go tomorrow?"
------------------------------
From: [EMAIL PROTECTED] (Anjan Sen)
Subject: ipchains help please
Date: Sat, 1 May 1999 10:50:08 +0100
Reply-To: [EMAIL PROTECTED]
Hello,
I have a system that is Red Hat 5.2, with the kernel newly upgraded to
2.2.7. I have a small private network with 192.168.*.* addreses, and this
Linux box that acts as a gateway. It's on dialup and has a static ip
address.
Prior to the upgrade I was using ipfwadm to masquerade my private
addresses like this, and it worked pretty well
ipfwadm -F -p deny
ipfwadm -F -a m -S 192.168.1.1/32 -D 0.0.0.0/0
Now, after the upgrade I attempted to get ipchains running in a very
simple manner. I thought I'd start off by using Rusty's Three-Line guide
To Masquerading found at www.rustcorp.com. My problem is that although I
have the correct version of ipchains - 1.3.8 - even when i cut and
paste the lines from my web browser, I get syntax errors from ipchains.
For example
root@sealion /usr/src/linux
> echo 1 > /proc/sys/net/ipv4/ip_forward
root@sealion /usr/src/linux
> ipchains -P forward -j DENY
ipchains: -P requires a chain and a policy
Try `ipchains -h' or 'ipchains --help' for more information.
root@sealion /usr/src/linux
> ipchains -A forward -i ppp0 -j MASQ
ipchains: No chain by that name
root@seYalion /usr/src/linux
> ipchains --version
ipchains 1.3.8, 27-Oct-1998
This must be me doing something daft, but I can't see what. I have
compiled the necessary support into my kernel
root@sealion /usr/src/linux
> grep CONFIG_IP_FIREWALL .config
CONFIG_IP_FIREWALL=y
root@sealion /usr/src/linux
> grep CONFIG_FIREWALL .config
CONFIG_FIREWALL=y
and ifconfig shows me this
ifconfig
eth0 Link encap:Ethernet HWaddr 00:A0:24:7E:BB:87
inet addr:192.168.1.2 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:2097 errors:0 dropped:0 overruns:0 frame:0
TX packets:1757 errors:0 dropped:0 overruns:0 carrier:0
collisions:4 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:194 errors:0 dropped:0 overruns:0 frame:0
TX packets:194 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
ppp0 Link encap:Point-to-Point Protocol
inet addr:158.152.160.123 P-t-P:158.152.1.222
Mask:255.255.255.255
UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1500 Metric:1
RX packets:116 errors:0 dropped:0 overruns:0 frame:0
TX packets:96 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:10
Please can someone help me? I've searched everywhere but haven't found an
answer.
Thanks.
------------------------------
From: Glenn Watson <[EMAIL PROTECTED]>
Subject: 10Base2 Connector/Cable Quality
Date: Sat, 01 May 1999 22:08:28 +1200
Hi all - I realise this isn't the most appropriate newsgroup for this
question, but anyway....I will soon be setting up a small network at
home.
The network will be between my Linux box and my brother's windoze box -
it is intended to allow us to use one modem/dial up connection at the
same time...I am planning to use IP masquerade for this purpose, and the
length of the cable will be 25 - 35 metres....
I figure that for a simple network like this, network cards which use
the BNC style 10Base2 cable will be fine (I'm a cheap student ok!...:)
Being a student I was just wondering what the difference in speed would
be between buying pre-made 10Base2 cable, and buying the coax/connectors
and making the cable myself...
Obviously, making it myself is a lot cheaper, but what will the
performance be like compared to a professionally made cable...
------------------------------
From: Benoit Gerrienne <[EMAIL PROTECTED]>
Crossposted-To: linux.samba
Subject: Re: Urgent... trust account problems
Date: Sat, 01 May 1999 12:03:51 +0200
Cyriac REMY wrote:
>
> Well.... I need help to to smthing with Samba 2.0.3-8
>
> Let me explain... Ive two servers : one NT and one Linux... Each servers
> control one domain :
> ==> NT server MYNT : control DOM1
> ==> Lin. server MYLIN : control DOM2
>
> when I configure smb.conf like other classics smb.conf files, I can see
> my server on network
> explorer on my NT workstation named NTWKS (global network ==> two
> domains DOM1 and
> DOM2 and when I clic on DOM2, I can see MYLIN and I can connect and see
> shared directories)
>
> Ok but now I want to switch NTWKS from DOM1 to DOM2, ssssoooo I go to
> network config in
> control panel and clic on domain radio button and type DOM2 and I see a
> message like :
> be sure your computer has an account on this domain (Im french, so Ive
> french message but it's
> smthing like that) but Ive created an account for NTWKS on MYLIN with
> adduser NTWKS$ and smbpasswd -a -m NTWKS.... ?!?!?
>
> And I see on my log file :
>
> [1999/04/27 15:17:28, 5] passdb/smbpass.c:getsmbfilepwent(258)
> getsmbfilepwent: returning passwd entry for user NTWKS$, uid 506
> [1999/04/27 15:17:28, 10] passdb/passdb.c:iterate_getsmbpwnam(156)
> found by name: NTWKS$
> [1999/04/27 15:17:28, 7] passdb/smbpass.c:endsmbfilepwent(81)
> endsmbfilepwent: closed password file.
> [1999/04/27 15:17:28, 4] smbd/password.c:smb_password_ok(400)
> Checking SMB password for user NTWKS$
> [1999/04/27 15:17:28, 5] smbd/password.c:smb_password_ok(410)
> use last SMBnegprot challenge
> [1999/04/27 15:17:28, 4] smbd/password.c:smb_password_ok(427)
> smb_password_ok: Checking NT MD4 password
> [1999/04/27 15:17:28, 4] smbd/password.c:smb_password_ok(431)
> NT MD4 password check succeeded
> [1999/04/27 15:17:28, 4] smbd/reply.c:session_trust_account(439)
> session_trust_account: Wksta trust account NTWKS$ denied by server
> [1999/04/27 15:17:28, 3] smbd/error.c:error_packet(127)
>
> So what happend ?????
>
> At last, I want to add DOM2 in the login NT Box to let users choose
> between the
> two domains...
>
> Can you help me ??? :)
>
> Cyriac REMY
> [EMAIL PROTECTED]
Hi,
have you added the "domain logons = yes" line into your smb.conf file ?
Best regards,
Benoit Gerrienne
Telindus Belgium
MHS & X.400 Support Engineer
My samba help site : http://www.ping.be/linux-and-samba ( under construction
)
Professional web site : http://www.telindus.com
------------------------------
From: "IBL" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.dial-up,alt.os.linux,comp.os.linux.hardware
Subject: Re: Need help with PPP
Date: Sat, 1 May 1999 12:21:46 +0200
hi there,
Sandeep Singh heeft geschreven in bericht
<[EMAIL PROTECTED]>...
>Hello all!
>
>I am trying to set-up PPP on a linux box (Redhat 5.2 kernel 2.2.5). Here
>is the syslog:
>Can anybody tell me what I am missing here? Thanks in advance for your
>time...
>
>Sandeep
>
>Apr 14 22:27:36 zobar chat[459]: Please Sign-on: -- got it
>Apr 14 22:27:36 zobar chat[459]: send ([EMAIL PROTECTED]^M)
>Apr 14 22:27:37 zobar pppd[445]: Serial connection established.
>Apr 14 22:27:38 zobar pppd[445]: Using interface ppp0
>Apr 14 22:27:38 zobar pppd[445]: Connect: ppp0 <--> /dev/ttyS0
>Apr 14 22:27:45 zobar pppd[445]: Modem hangup
>Apr 14 22:27:45 zobar PAM_pwdb[441]: (login) session closed for user
>root
>Apr 14 22:27:45 zobar pppd[445]: Connection terminated.
this resembles alot of what my logs show.
lucky me, i have a friend at my isp's helpdesk. their logs showed an error
41 wich means that pap or chap isn't working properly OR not the right
protocol is used to transmit.
you'll have to check with your isp.
few possibilities: the papa format <user>@domain * pap secret isn't
supported by many isp's
instead old pap-auth is using <user> * papsecret
another thing that could trigger this behaviour is on the transport
protocol.
for isdn with my isp i can only use hdlc / assync to sync converrtion. in a
win-box this (or the x75 ) protocol is set by default. but this may vary
from telephone standards accross the world.
good luck.
------------------------------
From: John Thompson <[EMAIL PROTECTED]>
Subject: Re: KPPP & IBM Global Network ISP problems
Date: Fri, 30 Apr 1999 17:46:39 -0600
J. Peterson wrote:
>
> No. I have defaultroute & noipdefault set in the "options" file.
> While connected, I more'd the file to make sure.
>
> On 29 Apr 1999 21:54:53 GMT, "LJ" <[EMAIL PROTECTED]> wrote:
>
> >Is Kppp adding a defaultroute ?
It's probably better to continue this in
ibmnet.dialers.other, so I've directed follow-ups there.
I have:
115200
name internet.usinet.<my_user_name>
noauth
defaultroute
noipdefault
in /etc/ppp/options and:
"internet.usinet.<my_user_name>" *
"<my_password>"
in /etc/ppp/pap-secrets
--
-John ([EMAIL PROTECTED])
------------------------------
Crossposted-To: comp.security.unix
Subject: Possible to use Linux for small firewall?
From: [EMAIL PROTECTED] (Miguel Cruz)
Date: Fri, 23 Apr 1999 12:53:54 GMT
For some reason this is eluding me. I posted a similar question a year ago
but never came up with a great solution. Perhaps with the changes in the
newer kernel something is possible now.
I need to run a firewall for just two NT machines. It is not going to be
possible to create a subnet for them.
Let's say the firewall is 10.0.0.1 and the machines are 10.0.0.2 and
10.0.0.3.
Here's what I tried (it did not work).
An NT machine needs to be available to outside world as 10.0.0.2
eth0 on firewall: 10.0.0.1
eth1 on firewall: 192.168.1.1
NT machine: 2 IP addresses: 10.0.0.2 and 192.168.1.2. Set gateway to
10.0.0.1. So far, so good.
On firewall: static route to 10.0.0.2 on eth1.
On firewall: proxy arp for 10.0.0.2 on eth0.
arp -s 10.0.0.2 et:he:rn:et:ad:dr pub
But that doesn't seem to work. Other machines on the network will not send
anything to the firewall machine when I try to ping 10.0.0.2. In their arp
tables (these are WinNT machines) the entry shows up as "invalid" or not at
all.
So...............
Is this the right approach? Is there a better way to provide firewall for
two machines without setting up a subnet? Is there ANY way to do it?
miguel
------------------------------
From: Neil Benson <[EMAIL PROTECTED]>
Subject: Dynamic PPP hostname changes !
Date: Sat, 01 May 1999 11:53:22 +0100
Hi
I have redhat 5.2 and im able to form a ppp connection to my ISP with no
problems. My ISP has dynamic IP allocation. And here is where the
problem starts...
My host name is localhost, when ive conencted to my ISP my hostname
alters to some string of random letters. so I got from nbenson@localhost
to nbenson@P4fsdhJuis !!
This is A) annouying and B) screws up my gnome session managment.
Does anyone the reason behind this ? My /stc/hosts only conatins the
127.0.0.1 entry for the local loop; i dont know what ip my isp will give
me.
Anyone else having this problem ?
Thanks
Neil Benson
[EMAIL PROTECTED]
http://www.users.globalnet.co.uk/~nbenson/
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
------------------------------
From: Trever Adams <[EMAIL PROTECTED]>
Subject: Re: DEC 21143 Chipset
Date: Sat, 01 May 1999 03:30:41 -0600
Roger wrote:
>
> Did anyone get a DEC 21143 Chipset Ethernet Card working properly? An older
> 21140 Card works. I'm using S.u.S.E. 6.0 on ASUS P5-A 300MHz K6-2
>
> Roger
Something changed in the tulip driver that made it not work in stock
2.2.x. I dont know about 2.0.x. I suggest you go to the web page listed
in /usr/src/linux/drivers/net/tulip.c and get a newer version. I didn't
use the latest test, but did use one from the devel page. It works
great with my 21143 based PCMCIA card.
Good luck,
Trever
------------------------------
From: Holger Eitzenberger <[EMAIL PROTECTED]>
Subject: Re: LPRng and a local printer
Date: 1 May 1999 10:58:10 GMT
Kai-Uwe Reichert <[EMAIL PROTECTED]> wrote:
> :lp=/dev/lp0
> :sd=/var/spool/lpd/lp0
> :lf=log:af=acct
> :if=/usr/sbin/lpf
> :mx#0:sh:sf
This is what my printcap says:
ljet6p|HP Laserjet 6P:\
:lp=/dev/null:
:sd=/var/spool/lpd/remote
:rp=<NAME>
:rm=<HOST>
:sh:pw#80:pl#72:px#1440:mx#0
:af=/var/log/lp-acct:lf=/var/log/lp-errs
Of course you have to change both <NAME> and <HOST> to your needs.
<NAME> is the name of the printer you gave it on that particular
<HOST>. So look in the /etc/printcap on that host for the name.
Regards.
Holger
--
+ PGP || GnuPG key -> finger [EMAIL PROTECTED] +
+++ Debian/GNU Linux <[EMAIL PROTECTED]> +++ ICQ: 2882018 +++
------------------------------
** 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
******************************
