Date: Thu, 6 Jul 1995 15:50:10 -0500 (CDT) From: Rob Snow <rsnow@legend.txdirect.net> To: ED WOLPERT <wolperte@knox.pcec.philips.com> Cc: Nik.Clayton@brunel.ac.uk, questions@freebsd.org Subject: Re: PPP and demand start Message-ID: <Pine.BSF.3.91.950706151303.491A-100000@oasis> In-Reply-To: <9507051819.AA13800@eis16.philips.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 5 Jul 1995, ED WOLPERT wrote:
> >>>>> "Nik" == Nik Clayton <Nik.Clayton@brunel.ac.uk> writes:
>
> -Nik> and PPP will then dial out as and when necessary. If I do this
> -Nik> however, PPP starts up, announces that it is in auto mode and
> -Nik> becomes a daemon (preety much what I expected). But if I then
> -Nik> try and ping, telnet or netscape out I get told that the network
> -Nik> is unreachable. PPP does not attempt to dial out.
>
> I had the same problem... it would only attempt to dial out when I
> pinged the gateway (ppp provider). However, after the initial
> session, and after the ppp daemon closed the connection due to
> timeout, pinging a host in my /etc/hosts file (and others) would
> re-open the connection.
>
> A different problem I have: The timeout would occur even with an
> active connection. It seems that if I don't keep pinging something
> while I'm connected (Either through an 'slurp' or long ftp session) it
> will timeout. Most strange. Ideas anyone?
>
GREAT!!! Others having problems with iij-ppp! Now we've got some people
to work with.
I'll tell you what I've done if you'll help me with my problems.
What I've done so far. (works with -auto ondemand)
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
#define ISP 204.57.120.3 - domain txdirect.net
#define oasis MyMachine
#define lost MyMachine connected via lp0
***************************************************
in /etc/hosts:
***************************************************
127.0.0.1 localhost
204.57.120.200 oasis ; he doesn't use the 2nd half of his
; class C so I use a non existant
; part of it internally
Note: I know this is wrong... see the end of message.
***************************************************
in /etc/ppp/ppp.conf
***************************************************
default:
set device /dev/cuaa1
set speed 115200
disable lqr ; what is this ???
deny lqr ; ditto
set dial .... stuff deleted ...
ondemand:
... dialing stuff deleted ...
set timeout 180 30 ; I dont know what the 30 does
set ifaddr 204.57.120.200/24 204.57.120.3/24
add 0 0 204.57.120.3
***************************************************
in /etc/ppp/ppp.linkup
***************************************************
MYADDR:
# delete ALL ; Commented out (Yes/No ?)
add 0 0 HISADDR
Note: I dont even know if this is used with -auto (anyone???)
If so should the delete ALL be there or not?
***************************************************
in /etc/resolv.conf
***************************************************
domain txdirect.net
nameserver 204.57.120.6
Note: Of course you mileage will vary here
***************************************************
in /etc/host.conf
***************************************************
moved hosts before bind
***************************************************
in /etc/sysconfig
***************************************************
defuaultdomainname="txdirect.net" ; My ISP domainname
network_interfaces="lo0 tun0 lp0"
ifconfig_lo0="inet localhost oasis"
ifconfig_tun0="inet oasis 204.57.120.3" ; My ISP
ifconfig_lp0="inet oasis lost" ; to my laptop
static_routes="loopback"
route_loopback="${hostname} localhost"
routedflags=-q ; docs say maybe -s but that causes a dial
; every 30 seconds. :-(
Note: I deleted all the static routes but this one and it made things better
Note: NO DEFAULT ROUTE IS SET. ppp sets it.
***************************************************
Invoke with ppp -auto ondemand in rc.local
***************************************************
MY PROBLEMS
***************************************************
It seems to work with a couple of problems.
1) It dials every 30 minutes and does something.
(turns out to be a DNS lookup from sendmail (queue time is 30 minutes))
(how to turn nameserver lookups off in sendmail tomorrow :-) )
2) I've got to open the connection with a ping before I can start Netscape
or it hangs awhile and then says DNS lookup failure. I assume this is
because 204.57.120.200 get changed to a valid address on connection and
Netscape started with .200 and wants it to be that from then on. ???
(for instance I did this message on oasis with PINE and pinged my ISP to
open the connection before mailing it so PINE wouldn't hang saying it was
sending mail then puke with couldn't send mail or some suck^Hh)
3) I REALLY dont like using his domain internally! I dont want to be
txdirect.net or 204.57.120.XXX. I want to be g3.com and 192.168.X.X
or some such. (until I get mine officially)
I realize that tun0 will HAVE to be 204.57.120.XXX, but I shouldn't
have to make my whole site 204.57.120.XXX, right???
Hope this helps...
---
Rob Snow
rsnow@txdirect.net
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.91.950706151303.491A-100000>