Linux-Networking Digest #81, Volume #10 Mon, 1 Feb 99 20:13:45 EST
Contents:
Connecting to this and that (Duane Klein)
MPI implementation on Debian Linux (Haresh)
Re: get MX record ([EMAIL PROTECTED])
IP Tunnel Setup under 2.0.3[5|6] (Jon Wright)
Re: how to configure cgi with Apache (Doug Goldstein)
Linux to Win95 networking problem ([EMAIL PROTECTED])
IP Masqing (Doug Goldstein)
Re: Redirecting Traffic based on Port Nr. (Greg Weeks)
netatalk serves files not printer ("Joe Doherty")
Re: Linux -> NT RAS + MSCHAP + encryption (Chris Richmond - MD6-FDC ~)
Re: ppp dial-on-demand or Diald ??? ("Daniel")
odd ppp problem (Luke Skywalker)
Load Balancing POP3 (Damon Warren)
Unidentified error: eth0 Transmit timed out??? ([EMAIL PROTECTED])
cpio Help (Allen Wong)
Re: [Q]: Domain Name Hosting (Luca Filipozzi)
Re: Redirecting Traffic based on Port Nr. (Luca Filipozzi)
pop3 via ssh and fetchmail (peter)
----------------------------------------------------------------------------
From: Duane Klein <[EMAIL PROTECTED]>
Subject: Connecting to this and that
Date: Mon, 01 Feb 1999 14:05:20 -0500
Reply-To: [EMAIL PROTECTED]
In my home office I have 3 PC's. One is Win95, one is a dual boot
Win95/RHL 5.2, and the third is a notebook running Slackware w/ a 2.0.34
rev. of Linux. I also use this last PC in my work office where I
connect it up to a LAN running on Novell 4.11.
1. In the home office I am trying to get the RHL and Slackware PC
connected over ethernet. I've read various How-to's but the RHL PC
keeps telling me the network is unreachable. If I boot that PC up to
WIN95, I can talk to the other Win95 PC with no problem. The Slackware
PC, when connected in my work office has no problem pinging other PC's.
2. I'm trying to use the smbclient command to look at either a Win95 or
WinNT machine. using the smbclient -L hostname, on the work offfice
network, I'm able to see servers, workgroups, etc. When I try to
connect to a host, I get an "invalid network name in tree connect."
I've tried the password in upper and lower case, but I get the same msg.
I believe I have the correct sharename, username, and password.
[EMAIL PROTECTED]
------------------------------
From: Haresh <[EMAIL PROTECTED]>
Subject: MPI implementation on Debian Linux
Date: Thu, 28 Jan 1999 14:24:42 -0400
hi, i was wondering if anyone out there has installed the Message
Passing Interface library successfully on a Debian system running KDE?
thanks...
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: get MX record
Date: Thu, 28 Jan 1999 18:29:33 GMT
: I alway use nslookup
: example:
: nslookup <enter>
: set type=mx <enter>
: synapses-it.com. <enter>
: this will give all mx records of the domain synapses-it.com (the dot is
: important).
nslookup up rocks :) But you can also do it in a single commandline call
if you want.
out [greyfoxx] ~$ nslookup -q=mx linuxberg.com
Server: localhost
Address: 127.0.0.1
linuxberg.com preference = 10, mail exchanger = mail.linuxberg.com
linuxberg.com nameserver = cns1.idirect.com
linuxberg.com nameserver = cns2.idirect.com
mail.linuxberg.com internet address = 207.136.66.73
cns1.idirect.com internet address = 207.136.66.20
cns2.idirect.com internet address = 207.136.80.18
------------------------------
From: [EMAIL PROTECTED] (Jon Wright)
Subject: IP Tunnel Setup under 2.0.3[5|6]
Date: 1 Feb 1999 17:26:14 -0500
I have tried to get tunneling configured between two machines. The diagram
below shows my configuration:
10.1.1.1
a 192.168.1.2 192.168.2.2 b----------
\ /
\ /
192.168.1.1 \ 192.168.254.1 / 192.168.2.1
*----------------------------------*
192.168.2.1 / 192.168.254.2
/
/
c 192.168.3.2
Machine Eth0 Eth1 (If Present)
A 192.168.1.2
B 192.168.2.2 10.1.1.1
C 192.168.3.2
I have machine B configured to masquerade all traffic to 192.168.1.0/24
because I cannot route the 10.0.0.0/8 network from end to end (It's a network
that I have no control over). I need to allow all machines on 192.168.3.0/24
to see into 10.0.0.0/8.
I have tried the following:
Machine B:
route add 192.168.3.2 gw 192.168.254.2
ifconfig tunl0 10.1.1.1 pointopoint 192.168.3.2
route add -net 192.168.3.0 gw 192.168.3.2 tunl0
Machine C:
route add 192.168.2.2 gw 192.168.3.1
ifconfig tunl0 192.168.3.2 pointopoint 10.1.1.1
route add -net 10.0.0.0 gw 192.168.2.2 tunl0
This does NOT work. I cannot ping|traceroute|telnet over this configuration.
Can anyone offer a solution to this?
Thanks,
Jon
--
Jon Wright, Network Administrator | 5700 SW 34th St. Suite 1235
Info Tech, Inc. | Gainesville, FL 32608
[EMAIL PROTECTED] | (352)375-7624 Voice
[EMAIL PROTECTED] | (352)373-9586 Fax
------------------------------
From: Doug Goldstein <[EMAIL PROTECTED]>
Subject: Re: how to configure cgi with Apache
Date: Mon, 01 Feb 1999 23:53:40 GMT
Get the new Apache 1.3.4 and read the INSTALL and README.configure files. What
you need to do is setup suEXEC.
------------------------------
From: [EMAIL PROTECTED]
Subject: Linux to Win95 networking problem
Date: Mon, 01 Feb 1999 21:22:02 GMT
I know that similar problems have been posted before, but I've yet to see a
good answer so far.
I have a two Win95 machines running TCP/IP hooked together through a 10 port
Linksys hub. The two win95 machines see each other fine, all tcp apps work
without a hitch. I recently have installed Redhat 5.2 running kernel 2.0.36 on
a third system and I'm attempting to attach it to the network.
The Linux box has an SMC Ultra NIC in it configured at 192.168.0.3, the
other two machines are 192.168.0.1 and 192.168.0.2. Subnet on all three is
255.255.255.0. Neither of the win95 machines can ping the linux box and the
linux box can't ping out to either of the win95 machines. The linux box says
network unreachable. The Win95 machines say 'request timed out' 4 times.
Interestingly, the RX packets for eth0 on ifconfig increments by 4 each time
I ping the linux box from a Win95 machine. So it appears something is
reaching the Linux machine but for some reason an error is occurring. I'm not
using DNS at all, but I have put both machines in my host files and they
resolve the names, the IP itself won't ping though. I've used ifconfig,
route, and netstat and can't see anything wrong configuration wise. The
/proc/interrupts file doesn't show any conflict for the NIC which is being
detected at boot time fine as much as I can tell. I've seen people suggesting
reinstalling Win95 networking including the NIC itself which I was desperate
enough to do, but that also didn't work.
I've read relavant portions of Linux Unleashed, the Net-3 and Networking
HOWTOs, and countless newsgroup posts regarding similar circumstances. The
only thing I haven't ruled out yet is a hardware problem with the SMC card
which I will swap tonight when I am home from work, but I'm fairly sure this
card is good. Is there any other place i can look configuration wise to try
and see if TCP/IP is configured correctly on the Linux box. I can ping
loopback and my own ip on the linux machine alright, but as I said, trying to
ping out of the linux box anywhere returns network unreachable. Also worth
noting, pinging out from the linux machine shows no activity on the hub. If I
ping to the linux machine from a win95 box, I can see some activity on the
hub and on the linux NIC.
Sorry for the longwindedness, but I wanted to be thorough. Does anyone think
this is anything other than a hardware conflict? Any advice or input is
appreciated greatly. Thanks in advance,
Dan Friedman
Cambridge Computer Services
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: Doug Goldstein <[EMAIL PROTECTED]>
Subject: IP Masqing
Date: Mon, 01 Feb 1999 23:57:45 GMT
I'm havin a little problem with my IP Masqing if someone can help. I
have my Linux box plugged directly to the net on a static IP. Then my
Win98 box connected to the Linux box on an network. Standard stuff. But
once in a while the Win98 box will lose connection to the internet.
Usually openning a telnet session into the Linux box will start that up
again. Any suggestions? Also I've been having a problem with Diablo. It
tells me my connection is too poor to play a game but I can chat. Also
says I'm not processing any UDP packets. I was told I needed this setup
for Diablo in /etc/rc.d/rc.local
/usr/sbin/ipautofw -A -r udp 6112 6112 -c tcp 116
/usr/sbin/ipautofw -A -r udp 6112 6112 -c tcp 118
Any help? Suggestions? Comments?
Thanks,
Doug Goldstein
------------------------------
Reply-To: [EMAIL PROTECTED]
From: [EMAIL PROTECTED] (Greg Weeks)
Subject: Re: Redirecting Traffic based on Port Nr.
Date: Mon, 1 Feb 1999 17:55:49 -0600
In article <7942sj$ods$[EMAIL PROTECTED]>,
"Volker Kalthaus" <[EMAIL PROTECTED]> writes:
> Hello all,
>
> i need help dealing with the following problem:
>
> i have one linux box servin as router & firewall. This Linux Box has one
> official internet IP Adress. My Problem is now, that i want to redirect all
> incoming traffic on port 80 to a specific IP Adr. Same for port 25.
>
> So how can i redirect Traffic on Linux based on Port Nr. to other IP Adrs
> ???????
Question:
I want to run a web server inside my Linux IP Masquerade firewall that
can be accessed from the internet. How do I do this.
Standard Answer:
A pin hole allows incoming connection to go through a firewall to an
internal machine for a specific service. There are two ways that I know
of to open a pin hole in an IP Masq Linux firewall without proxying.
Both are mentioned at the IP Masq resource page at
http://ipmasq.cjb.net/ or http://www.tor.shaw.wave.ca/~ambrose/ They are
redir and ipportfw. I tried redir first, and while it works and has the
advantage that you can test it from inside the firewall it has the
disadvantage that the logs on the web server show all connections as
coming from the firewall. ipportfw is a kernel patch and a utility to
change the kernel tables. It's advantages are it's faster and the logs
show the correct source. It's biggest disadvantage is that it's only
testable from outside the firewall. ipportfw information is at
http://www.ox.compsoc.org.uk/~steve/portforwarding.html and redir is at
http://www.geocities.com/SiliconValley/Heights/2288/redir_0.7.orig.tar.gz
Any pin hole poses a security risk as it bypasses your firewall. Use
them with care.
=======================================
This will work for mail on port 25 also, but it's not very secure. The
proper way to do mail is with smap from the tis toolkit. After saying
that, I just use port forwarding.
Greg Weeks
--
http://durendal.tzo.com/greg/
------------------------------
From: "Joe Doherty" <[EMAIL PROTECTED]>
Subject: netatalk serves files not printer
Date: Mon, 1 Feb 1999 13:37:37 -0800
I've been fooling around with this for too many hours . . . All I want to
do is print from my wife's Mac through the linux box.
I'm running RH5.1 (kernel 2.0.32) on an old AST Advantage! (486/33) 16mb
ram.
I installed netatalk (1.4b2+asun2.1.1) using RPM.
I've also got Samba running for my win95 machine.
The printer is generic postscript (HPIII with Postscript + cartridge) hooked
up to my linux box. It prints from my win95 box using Samba and from the
linux box.
My wife's machine is a Mac G3 OS 8.0.
The machines are using AppleTalk over ethernet.
Here's the problem (which I've seen so many times before on this and other
lists). My wife's Mac can see the server in the chooser under Appleshare,
and I can serve files from the linux box to her Mac; but I can't see any
printers in the chooser. I've fiddled with papd.conf in all the ways
recommended by man, by the how-tos and by every message I've ever read.
Every time I change papd.conf I restart atalk (which restarts papd). Here's
the latest configuration of papd.conf
poekie:\
:pr=lp:op=velda:\
:pd=/var/spool/lpd/lp/APLW81F1.PPD: (I've also used the HPIII .ppd
file)
poekie is one of the names specified in printcap as the postscript printer
(lp|poekie|hpiii postscript)
I've also tried it with an empty papd.conf,
and by changing pr to read :pr=poekie: and pr= -P poekie:
Nothing; no printer in the chooser. The chooser has HPIII and Laserwriter
8, which my wife can use when she dials into her office network. Shouldn't
poekie appear in the chooser window when I choose either of these printers
through appletalk?
The only anomaly I can think of (and it's in the messages below at Jan 30
13:11:54) is that I get an eth0 delayed initialization at boot, which I
remedy by entering the following after logging in as root:
insmod 3c509
ifconfig eth0 192.168.1.254
route add -net 192.168.1.0
and then I start samba and atalk. (This is probably unrelated, but any help
in solving this problem would be appreciated, also)
AFPD is simply
eth0
and I let the machine choose the rest of the parameters.
Help, Please, and TIA.
Joe
Please excuse the verbosity. Something here might be important to someone
much smarter than I am.
Jan 30 13:11:27 fletcher syslogd 1.3-3: restart.
Jan 30 13:11:27 fletcher kernel: klogd 1.3-3, log source = /proc/kmsg
started.
Jan 30 13:11:31 fletcher kernel: Loaded 4202 symbols from /boot/System.map.
Jan 30 13:11:31 fletcher kernel: Symbols match kernel version 2.0.32.
Jan 30 13:11:31 fletcher kernel: No module symbols loaded.
Jan 30 13:11:31 fletcher kernel: Console: 16 point font, 400 scans
Jan 30 13:11:31 fletcher kernel: Console: colour VGA+ 80x25, 1 virtual
console (max 63)
Jan 30 13:11:31 fletcher kernel: pci_init: no BIOS32 detected
Jan 30 13:11:31 fletcher kernel: Calibrating delay loop.. ok - 16.64
BogoMIPS
Jan 30 13:11:31 fletcher kernel: Memory: 14672k/16384k available (736k
kernel code, 384k reserved, 592k data)
Jan 30 13:11:31 fletcher kernel: This processor honours the WP bit even when
in supervisor mode. Good.
Jan 30 13:11:31 fletcher kernel: Swansea University Computer Society
NET3.035 for Linux 2.0
Jan 30 13:11:31 fletcher kernel: NET3: Unix domain sockets 0.13 for Linux
NET3.035.
Jan 30 13:11:31 fletcher kernel: Swansea University Computer Society TCP/IP
for NET3.034
Jan 30 13:11:31 fletcher kernel: IP Protocols: IGMP, ICMP, UDP, TCP
Jan 30 13:11:31 fletcher kernel: VFS: Diskquotas version dquot_5.6.0
initialized
Jan 30 13:11:31 fletcher kernel:
Jan 30 13:11:31 fletcher kernel: Checking 'hlt' instruction... Ok.
Jan 30 13:11:31 fletcher kernel: Linux version 2.0.32
([EMAIL PROTECTED]) (gcc version 2.7.2.3) #1 Wed Nov 19 00:46:45 EST
1997
Jan 30 13:11:31 fletcher kernel: Starting kswapd v 1.4.2.2
Jan 30 13:11:31 fletcher kernel: Serial driver version 4.13 with no serial
options enabled
Jan 30 13:11:31 fletcher kernel: tty00 at 0x03f8 (irq = 4) is a 16450
Jan 30 13:11:31 fletcher kernel: tty01 at 0x02f8 (irq = 3) is a 16450
Jan 30 13:11:31 fletcher kernel: PS/2 auxiliary pointing device detected --
driver installed.
Jan 30 13:11:31 fletcher kernel: Real Time Clock Driver v1.07
Jan 30 13:11:31 fletcher kernel: Ramdisk driver initialized : 16 ramdisks of
4096K size
Jan 30 13:11:31 fletcher kernel: hda: WDC AC26400B, 6149MB w/512kB Cache,
CHS=13328/15/63
Jan 30 13:11:31 fletcher kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
Jan 30 13:11:31 fletcher kernel: Floppy drive(s): fd0 is 1.44M
Jan 30 13:11:31 fletcher kernel: FDC 0 is an 8272A
Jan 30 13:11:31 fletcher kernel: md driver 0.35 MAX_MD_DEV=4, MAX_REAL=8
Jan 30 13:11:31 fletcher kernel: scsi : 0 hosts.
Jan 30 13:11:31 fletcher kernel: scsi : detected total.
Jan 30 13:11:31 fletcher kernel: Partition check:
Jan 30 13:11:31 fletcher kernel: hda: hda1 hda2 < hda5 hda6 hda7 >
Jan 30 13:11:31 fletcher kernel: VFS: Mounted root (ext2 filesystem)
readonly.
Jan 30 13:11:31 fletcher kernel: Adding Swap: 130372k swap-space
(priority -1)
Jan 30 13:11:31 fletcher kernel: sysctl: ip forwarding off
Jan 30 13:11:31 fletcher kernel: Swansea University Computer Society IPX
0.34 for NET3.035
Jan 30 13:11:31 fletcher kernel: IPX Portions Copyright (c) 1995 Caldera,
Inc.
Jan 30 13:11:31 fletcher kernel: Appletalk 0.17 for Linux NET3.035
Jan 30 13:11:37 fletcher named[222]: starting. named 4.9.6-REL Thu Nov 6
23:29:57 EST 1997 ^[EMAIL PROTECTED]:/usr/src/bs/BUILD/bind-4.9.6/named
Jan 30 13:11:37 fletcher named[222]: cache zone "" loaded (serial 0)
Jan 30 13:11:37 fletcher named[222]: primary zone "0.0.127.in-addr.arpa"
loaded (serial 1997022700)
Jan 30 13:11:38 fletcher named[223]: Ready to answer queries.
Jan 30 13:11:38 fletcher named[223]: sysquery: sendto([128.9.0.107].53):
Network is unreachable
Jan 30 13:11:54 fletcher insmod: /lib/modules/2.0.32/net/3c509.o: symbol for
parameter io not found
Jan 30 13:11:57 fletcher papd[364]: restart (1.4b2+asun2.1.1)
Jan 30 13:11:58 fletcher kernel: eth0: 3c509 at 0x300 tag 1, 10baseT port,
address 00 10 5a 1c bf 53, IRQ 10.
Jan 30 13:11:58 fletcher kernel: 3c509.c:1.12 6/4/97
[EMAIL PROTECTED]
Jan 30 13:11:59 fletcher kernel: eth0: Setting Rx mode to 1 addresses.
Jan 30 13:11:59 fletcher afpd[382]: main: atp_open: Cannot assign requested
address
Jan 30 13:11:59 fletcher afpd[382]: ASIP started on 192.168.1.254:548(0)
(1.4b2+asun2.1.1)
Jan 30 13:12:02 fletcher kernel: eth0: Setting Rx mode to 2 addresses.
Jan 30 13:12:02 fletcher atalkd[416]: restart (1.4b2+asun2.1.1)
Jan 30 13:12:03 fletcher kernel: eth0: Setting Rx mode to 3 addresses.
Jan 30 13:12:03 fletcher atalkd[416]: zip_getnetinfo for eth0
Jan 30 13:12:22 fletcher last message repeated 2 times
Jan 30 13:12:24 fletcher PAM_pwdb[402]: (login) session opened for user root
by (uid=0)
Jan 30 13:12:24 fletcher PAM_pwdb[402]: ROOT LOGIN ON tty1
Jan 30 13:12:32 fletcher atalkd[416]: config for no router
Jan 30 13:12:33 fletcher atalkd[416]: ready 0/0/0
Jan 30 13:14:42 fletcher kernel: Sony I/F CDROM : SONY CD-ROM CDU33A
Rev 1.0f
Jan 30 13:14:42 fletcher kernel: Capabilities: tray, audio, eject, LED,
elec. Vol, sep. Vol, double speed
------------------------------
From: [EMAIL PROTECTED] (Chris Richmond - MD6-FDC ~)
Crossposted-To: comp.protocols.ppp
Subject: Re: Linux -> NT RAS + MSCHAP + encryption
Date: 1 Feb 1999 23:23:45 GMT
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>, James Carlson
<[EMAIL PROTECTED]> writes:
|> Arun Sharma <[EMAIL PROTECTED]> writes:
|> > 1. Does Linux support data encryption over the line ?
|>
|> No. That's one of the more bizarre bits of Microsquish proprietary
|> stuff. MS negotiates encryption within the Compression Control
|> Protocol (CCP) rather than as part of the standard Encryption Control
|> Protocol (ECP). I don't think anyone supports that.
Is anyone working on this? I hvae the same problem and there is *NO*
way to change the thing that answers the phone at the other end. The
only option I have now is using Win9x for remote login. I don't
have (nor want) NT, even if it would work.
Thanks, Chris
--
Chris Richmond | Internet: [EMAIL PROTECTED]
Intel, Corp. FM5-97 | Phone: (916)-356-6470
1900 Prairie City Rd. | Pager: (916)-857-8979
Folsom, CA 95630 | I don't speak for Intel & visa versa
------------------------------
From: "Daniel" <[EMAIL PROTECTED]>
Subject: Re: ppp dial-on-demand or Diald ???
Date: Mon, 1 Feb 1999 22:40:47 +0100
>Demand dialing is supported by ppp-2.3.5 when the kernel pieces that
>come with an original source package are properly installed and the
>kernel recompiled. The particular kernel piece ppp-2.3.5/linux/ppp.c
>also requires a small patch.
Check out http://www.tartu.customs.ee/linux/index.shtml on how to patch the
kernel.
>I don't use demand dialing but I understand that the ppp-2.3.5 demand
>dialing is not as versatile as diald.
I prefer diald. You can set time restrictions and some other stuff with it.
--
.~. "In A World Without Fences...
/V\ ...Who Needs Gates?"
// \\
/( )\ Dani�l Verlouw <[EMAIL PROTECTED]>
^^-^^
------------------------------
From: Luke Skywalker <[EMAIL PROTECTED]>
Subject: odd ppp problem
Date: Mon, 01 Feb 1999 15:49:58 -0500
I have the installation of RedHat Workstation 5.2. After months of
teching my modem, i found out the problem. Any how. I use the netcfg
utility to connect to the isp I work for. I connect everytime, no
problem, however when I go to ping our primary dns server, i get no
reponse, period. The file looks as if it is in the process of bringing
back outuput, but nothing comes back. To resolve this I continue to
activate the interface until I can get responses. Any ideas????
jgh
------------------------------
Date: Mon, 01 Feb 1999 22:14:51 +0000
From: Damon Warren <[EMAIL PROTECTED]>
Crossposted-To: comp.mail.sendmail
Subject: Load Balancing POP3
I would just like some feedback on a situation I have. I have a mail
server running SMTP and POP3. I have had to up my inetd limits, and my
sendmail limits, and have been exploring ways of load balancing the two
protocols. My biggest problem is gunna be POP3, I know I can load
balance SMTP via DNS, and use the mailhub feature, But POP3 is another
story. I had two Ideas:
1) use DNS and NFS to load balance.
mail.mydomain.com goes to mail1.mydomain.com, and mail2.mydomain.com
mfiles.mydomain.com will share /var/spool/mail via NFS to mail1, and
mail2
2) Split users across two machines, and use a pop proxy to connect to
the correct machine.
I really don't like ether of these Idea's.... If anyone has any
suggestions about how I can efficiently load balance POP3 please let me
know.
-Damon Warren ([EMAIL PROTECTED])
-Network Systems Administrator
-ImageNet, Inc. (www.oneimage.com)
------------------------------
From: [EMAIL PROTECTED]
Subject: Unidentified error: eth0 Transmit timed out???
Date: Mon, 01 Feb 1999 22:07:52 GMT
Hello, I'm having a problem with my new linux server doing DNS, Samba, and
Netatalk. After having set up these services (which mostly seem to be working
- still haven't figured out printing), I now occasionally get the following
error:
eth0: Tansmit timed out: status 0050 command 0000.
eth0: Trying to restart the transmitter...
Now, everything seems to be working normally, but this server is not in wide
use yet, and I'd like to know what the problem is before* I start letting
everyone hammer it. Anyone know what this error means? I haven't found it in
any FAQ/Man page/How-To.
Any help would be appreciated.
-Deano
============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
------------------------------
From: Allen Wong <[EMAIL PROTECTED]>
Subject: cpio Help
Date: Mon, 01 Feb 1999 13:31:54 -0800
Okay, ladies and gentlemen. I give up! I bought an HP T-20i for
SCSI tape backup for my server. It works well, but only for the server
it's installed on. I want to use it to backup all the hard drives in my
network. But I can't seem to get the right syntax to make either "cpio"
or "tar" work. Or it could also be that I haven't set up the services
properly on my server. Here's a typical exchange when I try to use
"tar":
[amiga@Linux2:amiga]$ tar --rsh-command=/usr/local/bin/ssh -cvf
Hal:/dev/nst0 /usr/local
/bin/bash: /etc/rmt: No such file or directory
tar: Cannot open Hal:/dev/nst0: Input/Output error
tar: Error is not recoverable: exiting now
When I try to use "cpio", I get
[amiga@Linux2:amiga]$ find /usr/local -print | cpio -o -O Hal:/dev/nst0
cannot execute remote shell: No such file or directory
cpio Hal:/dev/nst0: Input/output error
Broken pipe
What's the problem? Incorrect syntax? Do I need to set up an .rhosts
file? What is this /etc/rmt file? I'll try anything!!!
------------------------------
From: [EMAIL PROTECTED] (Luca Filipozzi)
Subject: Re: [Q]: Domain Name Hosting
Date: Mon, 1 Feb 1999 16:40:35 -0800
In article <795dem$ohm$[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
> I still need help on domain hosting. You may have seen my previous post.
>
> I want to move my brie.com to my own dedicated machine where I have a
> static IP address. I currently have it virtually located. When I do a
> $whois brie.com I get the following:
>
> Domain servers in listed order:
>
> NS1.REQUEST.NET 206.151.75.9
> NS2.REQUEST.NET 208.211.228.3
> NS3.REQUEST.NET 209.27.13.3
>
> My static IP address of my own machine at home is 207.212.133.10
> Do I have to set up DNS server on my local machine to make this work?
Yes, you need to set up a DNS server. You should set up an authoritative
primary DNS on your machine and an authoritative secondary DNS with
ns1.granitecanyon.com (see http://soa.granitecanyon.com). You could put
the primary at Granite Canyon, too, but I think the primary should be
your own machine so that you don't rely on Granite Canyon (they're up
*most* of the time) and so that they don't get overloaded by everyone
putting their domains on their DNS server.
> When I
> modify internic will I modify the section above to on my Internic
> Registration to look like below? Or can I have it as DNS.BRIE.COM and have
> WWW.BRIE.COM on the same machine?
>
> Domain servers in listed order:
>
> WWW.BRIE.COM 207.212.133.10
Your registration should be something like
dns.brie.com 207.212.133.10
ns1.granitecanyon.com aaa.bbb.ccc.ddd
Then, on dns.brie.com, you set up a DNS server that has the appropriate
records, like
@ IN SOA blah blah ( blah blah blah )
IN NS dns.brie.com.
IN NS ns1.granitecanyon.com.
IN MX 10 dns.brie.com.
dns.brie.com. IN A 207.212.133.10
www.brie.com. IN CNAME dns.brie.com.
mail.brie.com. IN CNAME dns.brie.com.
news.brie.com. IN CNAME dns.brie.com.
This means that as far as internic is concerned, it will ask dns.brie.com
for the address of www.brie.com (which will be the same machine!).
You can read up about DNS servers in the DNS HOWTO (www.linux.org).
--
Luca Filipozzi <[EMAIL PROTECTED]>
------------------------------
From: [EMAIL PROTECTED] (Luca Filipozzi)
Subject: Re: Redirecting Traffic based on Port Nr.
Date: Mon, 1 Feb 1999 16:43:26 -0800
In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED]
says...
> http://ipmasq.cjb.net/ or http://www.tor.shaw.wave.ca/~ambrose/ They are
> redir and ipportfw.
Or you can use ipautofw which is does require a kernel recompilation but
does not require patching it (under debian / linux 2.0.34).
--
Luca Filipozzi <[EMAIL PROTECTED]>
------------------------------
From: [EMAIL PROTECTED] (peter)
Crossposted-To: comp.security.ssh
Subject: pop3 via ssh and fetchmail
Date: Mon, 01 Feb 1999 22:23:46 GMT
when running it "manually" it works, when running it "automated" it
doesnt work :(
when I execute
$ssh -C -f me@popserver -L 11110:popserver:110 sleep 120
$fetchmail
with the following .fetchmailrc
---
poll localhost proto pop3 port 11110 keep pass mypass
---
all is running ok. fetchmail connects to localhost:11110 and ssh
reconnects to popserver and get my mail and I�m happy.
but when I want to use fetchmail in daemon-mode I write the following
.fetchmailrc
----
poll localhost proto pop3 port 11110 keep pass mypass: preconnect "ssh -C
-f me@popserver -L 11110:popserver:110 sleep 20"
----
which should do exactely the same like above.
but:
$fetchmail -v
fetchmail: 4.5.8 querying localhost (protocol POP3) at Mon Feb 1
23:21:28 1999
fetchmail: POP3 connection to localhost failed: local error: Connection
refused
fetchmail: Query status=2
fetchmail: normal termination, status 2
$
the preconnect-command is executed, cause (when executed within the 20
seconds):
$ netstat -an |grep "11110"
tcp 0 0 0.0.0.0:11110 0.0.0.0:* LISTEN
$
I already played around with poll popserver via localhost but this makes
no differ (like one expected).
can anyone tell what is different in this two ways !! different on local
side or on remote side.
thanks in advance,
peter
=================
pilsl@
ANTISPAM
riemann.atat.at
------------------------------
** 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
******************************
