[juneau-lug] Re: ACS DSL linux setup
- From: James Zuelow <e5z8652@xxxxxxxxxx>
- To: juneau-lug@xxxxxxxxxxxxx
- Date: Sat, 27 May 2006 08:51:13 -0800
Yah, here's one of my typical long-winded and wordy replies.
On Saturday 27 May 2006 02:23, East Wind wrote:
> Hello Jamie,
>
>
> J> Used the wizard and selected PPP over ethernet and set his user name and
> J> password. ifconfig showed eth0 with an address at 192.168.1.47 and ppp0
> J> had an address of 209.x.x.x. But couldn't ping anything.
>
OK, eth0 needs to be up, but it doesn't need to have an IP address. Mine
doesn't. All eth0 does is provide a layer 1 link to the DSL Modem. All of
your traffic, default route, everything goes through ppp0.
> Do an adsl-start.
^^^ If it doesn't work try `pon dsl-provider` or `pon acs` (use pon PEER to
start ppp, poff to stop it.) There is also pppoe-start.
>
> route delete default
^^^ use this one
>
> route add default ppp0
^^^ use this one. This is the right syntax for my setup, which doesn't have a
gateway specified:
default * 0.0.0.0 U 0 0 0 ppp0
>
> Then ping.
>
Should work.
>
> J> ACS help was not. I haven't set up DSL before, but didn't expect a
> J> problem. What have I missed/messed?
>
Probably not the correct default route, or it is pointing at eth0 (which
doesn't go anywhere).
> Roaring Penguin PPPoE website if this is fedora.
> I've never used the wizard - I edit the rp-pppoe.conf
> file.
>
> J> 1) My LAN connection left some config files set? But the clean Knoppix
> J> boot had the same problem.
> J> 2) Bad username / password? We called ACS to double check, but is there
> J> a way to tell? See an error message or monitor the line.
> J> 3) ADSL choice was wrong? What then.
>
/var/log/messages should show you any ppp errors (Debian). If not messages,
try /var/log/syslog. I've not run Mandriva so don't know their log splits.
> [root@somebox ~]# adsl-status
> adsl-status: Link is up and running on interface ppp0
If that doesn't work, try pppoe-status, but on my box it doesn't work - but
I'm tracking Debian testing, so things break from time to time.
> ppp0 Link encap:Point-to-Point Protocol
> inet addr:209.193.14.103 P-t-P:209.193.14.254
> Mask:255.255.255.255 UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1492
> Metric:1 RX packets:233714 errors:0 dropped:0 overruns:0 frame:0
> TX packets:208378 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0 txqueuelen:3
> RX bytes:197798423 (188.6 MiB) TX bytes:28814064 (27.4 MiB)
>
>
>
> J> 4) What does a working DSL ifconfig show - should I expect 2 ports eth0
> J> & ppp0?
Yes. Remember eth0 is just there to provide ppp0 a physical connection to the
DSL modem, it doesn't do any layer 2 stuff. You'll see bits coming in and
out if you look at ifconfig, but that is mostly ppp0 traffic.
---> side note. "Mostly ppp0 traffic"? The rest is coming from your
neighbors. You know how DSL providers always tell you you have your own
private connection when you use DSL, not "shared with your neighbors" like
cable modem? Set up eth0 with no IP address and point snort at it. You'll
see all of your non-existent neighbors with their non-existent viruses or
chatty Windows boxes scanning away for UPnP, broadcasting their workgroups
and shares, etc. So don't forget to specifically mention eth0 in your
firewall, even if it doesn't have an IPv4 address - it probably has an IPv6
address!
> [root@somebox ~]# ifconfig
>
> eth0 Link encap:Ethernet HWaddr 00:C0:9F:4A:CA:90
> inet addr:192.168.0.2 Bcast:192.168.0.255 Mask:255.255.255.0
> inet6 addr: fe80::2c0:9fff:fe4a:ca90/64 Scope:Link
> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> RX packets:212173 errors:0 dropped:0 overruns:0 frame:0
> TX packets:237328 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0 txqueuelen:1000
> RX bytes:31070849 (29.6 MiB) TX bytes:199976334 (190.7 MiB)
> Interrupt:11 Base address:0xa000
>
> eth1 Link encap:Ethernet HWaddr 00:A0:24:AA:00:D6
> inet6 addr: fe80::2a0:24ff:feaa:d6/64 Scope:Link
> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> RX packets:252213 errors:0 dropped:0 overruns:0 frame:0
> TX packets:226898 errors:0 dropped:0 overruns:0 carrier:0
> collisions:43 txqueuelen:1000
> RX bytes:204166516 (194.7 MiB) TX bytes:33965688 (32.3 MiB)
> Interrupt:5 Base address:0x300
See no IP address on eth1? EastWind has his DSL here.
>
> lo Link encap:Local Loopback
> inet addr:127.0.0.1 Mask:255.0.0.0
> inet6 addr: ::1/128 Scope:Host
> UP LOOPBACK RUNNING MTU:16436 Metric:1
> RX packets:2594 errors:0 dropped:0 overruns:0 frame:0
> TX packets:2594 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0 txqueuelen:0
> RX bytes:2982012 (2.8 MiB) TX bytes:2982012 (2.8 MiB)
>
> ppp0 Link encap:Point-to-Point Protocol
> inet addr:209.193.14.103 P-t-P:209.193.14.254
> Mask:255.255.255.255 UP POINTOPOINT RUNNING NOARP MULTICAST MTU:1492
> Metric:1 RX packets:233871 errors:0 dropped:0 overruns:0 frame:0
> TX packets:208542 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0 txqueuelen:3
> RX bytes:197918703 (188.7 MiB) TX bytes:28824872 (27.4 MiB)
>
>
> J> 5) Some additional command needed to "up" it?
>
> adsl-start I believe, if you're doing it manually.
>
> eth1 is not brought up at boot time.
> ppp0 attaches to eth1 when you adsl-start.
Once again, try `pon dsl-provider` or `pppoe-start`
(dsl-provider is the default name used for Debian's pppoeconf. Your mileage
may vary on Mandriva, or you may have changed the provider name to something
like ACS.)
My system is set up to start ppp0 at start, which was a configuration option
from Debian's pppoeconf. I did set up my eth0 to come up at boot even though
it doesn't have any IPv4 configuration.
When I was in Iraq nobody could get ACS DLS set up on this box, but I found
that was only because the wrong username was written down on the paper they
were using. (That's why you were beating your head on the monitor, Matt!)
Remember ACS wants "username@e-mail" as a username, not just the name part.
(Which is dumb, since anyone actually using their e-mail address is telling
the world their DSL account name, and they don't enforce particularly strong
password policies.) Once I put the right username into the system it "Just
Worked" (TM).
Another thing to remember is that you can start a ppp connection multiple
times, giving you ppp0, ppp1, ppp2, etc. etc. So if you set up ppp0 and it
fails, you try again and get ppp1 and it works, but your default route points
to ppp0 (or your firewall denies ppp1 traffic) it will seem to be broken.
`poff -a` will kill them all and let you start over.
And finally, if it still wants to use an old username or wrong password and
going through the config utility over and over again won't fix it -- hit
the /etc/ppp/peers directory and delete everything named ACS, acs, provider,
dsl-provider, etc. before you start over. (The only file I would leave on my
box is kppp-options, but since I don't use kppp deleting that too wouldn't
hurt anything.)
It should work fine, no problems as long as you start fresh every time you try
to reconfigure it. (And it's not black magic if Linux is involved, so ACS
might want to educate their helpdesk just a wee bit.)
------------------------------------
The Juneau Linux Users Group -- http://www.juneau-lug.org
This is the Juneau-LUG mailing list.
To unsubscribe, send an e-mail to juneau-lug-request@xxxxxxxxxxxxx with the
word unsubscribe in the subject header.
Other related posts: