At 01:14 PM 11/6/98 -0500, Edward Doolittle wrote:
>
>On Fri, 6 Nov 1998, Marvin George wrote:
>
>> For over 9 months, we have successfully used Linux (RH 5.0) with diald
>> and IP masquerading to put our LAN on the Internet. Our former ISP used
>> a version of Unix (unknown which one). We recently switched to a
>> different ISP who uses (UGH!!!) NT. The dynamically assigned local IP
>> is not present in the output of ifconfig -- it is using the same IP as
>> the ethernet card, which is from the private set of network numbers
>> (192.168.1.x). When I log on using dip, it works correctly, assigning
>> the dynamic IP to the local (ppp0) interface. With both dip and diald,
>> the remote IP is correctly assigned.
>
>So what's the problem?
Diald, according to the documentation I have found, can not use dip. Dip
requires manual intervention to dial. Diald can log onto the NT ISP OK,
but will not communicate once pppd is started, I assume due to the
incorrect local IP number. Diald works great with the old Unix ISP.
>> Also, when logging on with dip, IP masquerading does not work.
>
>Odd. Where is your IP masquerading set up? Please post your
>configuration files, output from ifconfig, ipfwadm, etc. so we can see the
>details.
See the end of this message. (Sorry about the length....)
>> Any ideas about what I might need to change for either diald or IP
>> masquerading?
>
>Umm ... switch to an ISP that uses Unix? :-)
We hopefully will BE an ISP before too long, and we WILL use Linux.
The ISP we just switched FROM (running Unix) has high prices and terrible
tech support...we have co-located our own RH 5.1 Linux server onto the
premises of the new ISP, but we (and our customers) must still access it
through the ISP's dialup NT-based servers, or from elswhere on the
Internet. Why does dip pick up the new ISP's dynamically-assigned inet
address, but diald doesn't? And why does diald pick it up from the old ISP
but not the new one? Both dip and diald pick up the P-t-P address OK,
regardless of which ISP is being connected to. I think that, when the inet
address is wrong but the P-t-P address is OK, our Linux box sends out
packets OK but does not respond to received packets at all.
>Ed
Marv (requested info follows)
+++++++++++++++++++
Here's the output of ifconfig (edited to remove some eth0:? aliases) with
diald running, not connected:
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
RX packets:508 errors:0 dropped:0 overruns:0
TX packets:508 errors:0 dropped:0 overruns:0
eth0 Link encap:Ethernet HWaddr 00:20:18:25:18:C9
inet addr:192.168.1.57 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:7981 errors:0 dropped:0 overruns:0
TX packets:697 errors:0 dropped:0 overruns:0
Interrupt:9 Base address:0x300
sl0 Link encap:Serial Line IP
inet addr:192.168.1.57 P-t-P:192.168.0.2 Mask:255.255.255.0
UP POINTOPOINT RUNNING MTU:552 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0
TX packets:8319 errors:0 dropped:0 overruns:0
++++++++++++++++++++
And when connected to the old (Unix) ISP via diald:
++++++++++++++++++++
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
RX packets:597 errors:0 dropped:0 overruns:0
TX packets:597 errors:0 dropped:0 overruns:0
eth0 Link encap:Ethernet HWaddr 00:20:18:25:18:C9
inet addr:192.168.1.57 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:10048 errors:0 dropped:0 overruns:0
TX packets:1091 errors:0 dropped:0 overruns:0
Interrupt:9 Base address:0x300
sl0 Link encap:Serial Line IP
inet addr:204.134.118.109 P-t-P:204.134.75.23 Mask:255.255.255.0
UP POINTOPOINT RUNNING MTU:552 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0
TX packets:10221 errors:0 dropped:0 overruns:0
ppp0 Link encap:Point-to-Point Protocol
inet addr:204.134.118.109 P-t-P:204.134.75.23 Mask:255.255.255.0
UP POINTOPOINT RUNNING MTU:552 Metric:1
RX packets:144 errors:2 dropped:2 overruns:0
TX packets:183 errors:0 dropped:0 overruns:0
+++++++++++++++++++++
And via diald to the new (NT) ISP:
+++++++++++++++++++++
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
RX packets:627 errors:0 dropped:0 overruns:0
TX packets:627 errors:0 dropped:0 overruns:0
eth0 Link encap:Ethernet HWaddr 00:20:18:25:18:C9
inet addr:192.168.1.57 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:10164 errors:0 dropped:0 overruns:0
TX packets:1201 errors:0 dropped:0 overruns:0
Interrupt:9 Base address:0x300
ppp0 Link encap:Point-to-Point Protocol
inet addr:192.168.1.57 P-t-P:209.181.112.35 Mask:255.255.255.0
UP POINTOPOINT RUNNING MTU:552 Metric:1
RX packets:12 errors:1 dropped:1 overruns:0
TX packets:24 errors:0 dropped:0 overruns:0
sl0 Link encap:Serial Line IP
inet addr:192.168.1.57 P-t-P:209.181.112.35 Mask:255.255.255.0
UP POINTOPOINT RUNNING MTU:552 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0
TX packets:21 errors:0 dropped:0 overruns:0
++++++++++++++++++++++++
And, finally, when connected via dip to the new ISP:
++++++++++++++++++++++++
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
RX packets:639 errors:0 dropped:0 overruns:0
TX packets:639 errors:0 dropped:0 overruns:0
eth0 Link encap:Ethernet HWaddr 00:20:18:25:18:C9
inet addr:192.168.1.57 Bcast:192.168.1.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:10209 errors:0 dropped:0 overruns:0
TX packets:1241 errors:0 dropped:0 overruns:0
Interrupt:9 Base address:0x300
ppp0 Link encap:Point-to-Point Protocol
inet addr:209.181.112.82 P-t-P:209.181.112.70 Mask:255.255.255.0
UP POINTOPOINT RUNNING MTU:1524 Metric:1
RX packets:10 errors:0 dropped:0 overruns:0
TX packets:10 errors:0 dropped:0 overruns:0
----------------
Here's the connect script used by diald. Note that I have prompt strings,
usernames, and passwords for both ISPs, and comment out the ones for the
ISP which is NOT being called. (Also note I changed the
usernames/passwords to protect the guilty. :-))
#!/bin/sh
# Copyright (c) 1996, Eric Schenk.
#
# This script is intended to give an example of a connection script that
# uses the "message" facility of diald to communicate progress through
# the dialing process to a diald monitoring program such as dctrl or
diald-top.
# It also reports progress to the system logs. This can be useful if you
# are seeing failed attempts to connect and you want to know when and why
# they are failing.
#
# This script requires the use of chat-1.9 or greater for full
# functionality. It should work with older versions of chat,
# but it will not be able to report the reason for a connection failure.
# Configuration parameters
# The initialization string for your modem
MODEM_INIT="ATZ&C1&D2%C0"
# The phone number to dial
# for cyberport
PHONE_NUMBER="3246500"
# for outerbounds
#PHONE_NUMBER="5648030"
# The chat sequence to recognize that the remote system
# is asking for your user name.
# for cyberport
USER_CHAT_SEQ="name:--name:"
# for outerbounds
#USER_CHAT_SEQ="ogin:--ogin:"
# The string to send in response to the request for your user name.
# for cyberport
USER_NAME="wwwww"
# for outerbounds
#USER_NAME="xxxxx"
# The chat sequence to recongnize that the remote system
# is asking for your password.
PASSWD_CHAT_SEQ="assword:"
# The string to send in response to the request for your password.
# for cyberport
PASSWORD="yyyyy"
# for outerbounds
#PASSWORD="zzzzz"
# The prompt the remote system will give once you are logged in
# If you do not define this then the script will assume that
# there is no command to be issued to start up the remote protocol.
#PROMPT="annex:"
# The command to issue to start up the remote protocol
#PROTOCOL_START="ppp"
# The string to wait for to see that the protocol on the remote
# end started OK. If this is empty then no check will be performed.
#START_ACK="Entering PPP mode."
# Pass a message on to diald and the system logs.
function message () {
[ $FIFO ] && echo "message $*" >$FIFO
logger -p local2.info -t connect "$*"
}
# Initialize the modem. Usually this just resets it.
message "Initializing Modem"
chat TIMEOUT 5 "" $MODEM_INIT OK ATS11=55 TIMEOUT 45 OK ""
if [ $? != 0 ]; then
message "Failed to initialize modem"
exit 1
fi
# Dial the remote system.
message "Dialing system"
chat \
TIMEOUT 75 \
ABORT "NO CARRIER" \
ABORT BUSY \
ABORT "NO DIALTONE" \
ABORT ERROR \
"" ATDT$PHONE_NUMBER \
"CONNECT 115200" ""
case $? in
0) message Connected;;
1) message "Chat Error"; exit 1;;
2) message "Chat Script Error"; exit 1;;
3) message "Chat Timeout"; exit 1;;
4) message "No Carrier"; exit 1;;
5) message "Busy"; exit 1;;
6) message "No DialTone"; exit 1;;
7) message "Modem Error"; exit 1;;
*)
esac
# We're connected try to log in.
message "Logging in"
chat \
TIMEOUT 5 \
$USER_CHAT_SEQ \\q$USER_NAME \
TIMEOUT 45 \
$PASSWD_CHAT_SEQ $PASSWORD
if [ $? != 0 ]; then
message "Failed to log in"
exit 1
fi
# We logged in, try to start up the protocol (provided that the
# user has specified how to do this)
if [ $PROMPT ]; then
message "Starting Comm Protocol"
chat TIMEOUT 15 $PROMPT $PROTOCOL_START
if [ $? != 0 ]; then
message "Prompt not received"
exit 1
fi
fi
if [ $START_ACK ]; then
chat TIMEOUT 15 $START_ACK ""
if [ $? != 0 ]; then
message "Failed to start Protocol"
exit 1
fi
fi
# Success!
message "Protocol started"
-----------------------------
The dip script (w/ changed username/password) (this is for the NT ISP; use
same, but with appropriate changes in prompts/username/password for Unix ISP):
#!/usr/sbin/dip
#
# dip script to connect to outerbounds
#
port /dev/modem
speed 115200
reset
flush
send ATQ0V1E1X1\r
wait OK 2
if $errlvl != 0 goto error
send ATS11=55\r
wait OK 2
if $errlvl != 0 goto error
dial 5648030
wait 00 90
if $errlvl != 0 goto error
sleep 3
wait ogin 10
if $errlvl != 0 goto error
send yyyyy\n
wait assword 5
if $errlvl != 0 goto error
send zzzzz\n
wait Mode. 30
if $errlvl != 0 goto error
get $local 0.0.0.0
mode PPP
error:
print failed to connect
quit
---------------------------
ipfwadm rules (loaded by system rc.* startup scripts):
#!/bin/sh
#
# /etc/rc.d/rc.firewall
#
# called from /etc/rc.d/rc.sysinit
#
# 27 mar 98
# Marvin George <[EMAIL PROTECTED]>
#
# tell what's happening
echo $0: Setting firewall rules
# first, flush out any existing rules
ipfwadm -I -f
ipfwadm -O -f
ipfwadm -F -f
# now set default of accept for I & O, deny for F
ipfwadm -I -p accept
ipfwadm -O -p accept
ipfwadm -F -p deny
# and set the rule for masquerading from local net
ipfwadm -F -a masquerade -P all -S 192.168.0.0/16 -D 0.0.0.0/0
ipfwadm -F -a masquerade -P all -S 209.181.112.0/24 -D 0.0.0.0/0
=============================
NOTE: The '-S 209.---' gets changed as appropriate for the ISP being
connected to; for some reason, the above rules don't seem to work with the
NT-based ISP, even with the correct net address in place. Nor does it work
with the Unix-based ISP when I connect via dip.
Marvin George <[EMAIL PROTECTED]>
----------------------------------------------------
Reach me by ICQ. My ICQ# is 17374743 or,
* Page me online through my Personal Communication Center:
http://wwp.mirabilis.com/17374743 (go there and try it!) or,
* Send me E-mail Express directly to my computer screen
[EMAIL PROTECTED]
For downloading ICQ at http://www.icq.com/
For adding similar signatures to your e-mail go to:
http://www.icq.com/emailsig.html
Marvin George
----------------------------------------------------
-
To unsubscribe from this list: send the line "unsubscribe linux-diald" in
the body of a message to [EMAIL PROTECTED]
