ATT Uverse RG Bypass (0.2 BTC)
-
Is this expected behavior?
Running the netgraph bypass as documented at https://github.com/MonkWho/pfatt . No LANs have been routed to ngeth0 just yet.
I get about about one packet every two-three minutes from the RG: tcpdump -ei em4
10:06:30.887851 f8:2d:c0:yy:yy:yy (oui Unknown) > Broadcast, ethertype 802.1Q (0x8100), length 424: vlan 0, p 3, ethertype IPv4, 0.0.0.0.bootpc > 255.255.255.255.bootps: BOOTP/DHCP, Request from f8:2d:c0:yy:yy:yy (oui Unknown), length 378And I get about 100 per minute from the ONT: tcpdump -ei em5
09:59:03.144906 a0:f3:e4:59:27:94 (oui Unknown) > f8:2d:c0:yy:yy:yy (oui Unknown), ethertype 802.1Q (0x8100), length 60: vlan 0, p 0, ethertype IPv4, 162-224-176-1.lightspeed.stlsmo.sbcglobal.net > zzz-zzz-179-129.lightspeed.stlsmo.sbcglobal.net: ICMP echo reply, id 30739, seq 4885, length 8- $RG_IF = em4
- $ONT_IF = em5
- f8:2d:c0:yy:yy:yy / zzz-zzz-179-129.lightspeed.stlsmo.sbcglobal.net = my RG
- a0:f3:e4:59:27:94 / 162-224-176-1.lightspeed.stlsmo.sbcglobal.net = ATT
-
^^For the first one, I think that might be a byproduct of the rg not getting an ip when using the eap proxy method. That is, it keeps requesting, but because the proxy only passes 802.1x traffic, it never actually receives it.
The 2nd looks like the gateway is responding to a ping request? You have something pinging the gateway ip (162.224.176.1) often?
-
Sounds like the first one is benign, unless, it is an indicator that something else is wrong.
The second - I’m not pinging in on that IP.
Edit: The second thing, with the 100+ pings per minute, was the pfSense gateway monitor. It's now disabled.
-
What gateway box are you using? Maybe time to dump it entirely and go wpa_supplicant method?
-
@GPz1100 Worked on that last night, I’ve got the certs off of the RG and broken into PEMs. Will work on the rest this evening.
Bricked the gateway though. I think I left the file system RW when I rebooted. It’s in a boot loop.
Thankfully, I still have my Charter connection, so I’m not in an outage condition.
-
Can't get the supplicant mode to work.
I had to comment lines 205-231 of pfatt.sh to get the system to boot.
wpa_cli status says:
Supplicant PAE state=HELD
suppPortStatus=Unauthorized
EAP state=FAILUREtcpdump -i ONT_IF -e vlan says:
05:20:52.486546 f8:2d:c0:xx:xx:xx (oui Unknown) > 01:80:c2:00:00:03 (oui Unknown), ethertype 802.1Q (0x8100), length 22: vlan 0, p 0, ethertype EAPOL, EAPOL start (1) v2, len 0/conf/pfatt/bin/* is 755 and /conf/pfatt/wpa/* is 644
Certs import without error into the web configurator, if only to make sure that they're intact. I've since pulled them back out of there.
What could be keeping this thing from being authorized?
-
I just moved into a temporary housing situation and was told this unit has ATT U-Verse. I'm looking at the back of my current RG (BGW210-700) and it's using an RJ11 from "broadband" port to the wall. The wall port doesn't have an RJ45 connection. Are there any workarounds for this using this method?
-
Rj11 = dsl. You need fiber.
-
@neatneat said in ATT Uverse RG Bypass (0.2 BTC):
I just moved into a temporary housing situation and was told this unit has ATT U-Verse. I'm looking at the back of my current RG (BGW210-700) and it's using an RJ11 from "broadband" port to the wall. The wall port doesn't have an RJ45 connection. Are there any workarounds for this using this method?
Uverse is the name of their interwebz service which can be DSL or Fiber. You could get a DSL modem if you don't want to use their equipment. If you have uverse TV service, or ATT home phone service, you'll need to keep their equipment in place.
-
I removed the gateway from my setup using the supplicant method from MonkWho's fork of pfatt from aus.
https://github.com/MonkWho/pfatt/tree/supplicant
This worked well on a SG-5100 on pfSense 2.4.5 p1 with certificates purchased on eBay from maczrcool. I'm getting full line speed (940/940) with no issues. I've tested that the setup survives expected and unexpected reboots.
-
@bk150 I'm trying to do what you've accomplished, on my PCEngines APU2 box with 2.4.5 p1 using MonkWho's fork. I have the certs extracted from my Arris BGW210. Did you also need Netgraph for the Supplicant approach? I was hoping to avoid it.
-
@lmgcnbzlp said in ATT Uverse RG Bypass (0.2 BTC):
I have the certs extracted from my Arris BGW210
Just wondering what you used to extract the certs.
-
@JonH I used the python script method per:
https://www.reddit.com/r/ATT/comments/g59rwm/bgw210700_root_exploitbypass/
Had to run it twice because the script didn't have a wait period defined to account for telnet server startup leadtime, but on the second run it appears to have worked and was able to save the certs to my local machine in the manner documented in the readme.
-
Did you get this to work? I am trying to enable pfsense to be moved to a different system via vmotion for doing maintenance on the host. Right now I use the netgraph bypass method, but I don't think it will work if I am using PCI device passthrough, which prevents the VM to be migrated.
Does the vlan0 tagging get interfered with by vmware's management of the network interfaces? I can use port mirroring on a switch to make the ONT and even the gateway available on multiple systems.
thx
mike -
I have the netgraph version working well on my system. I am currently running 2.4.4-RELEASE. Will upgrading to 2.4.5_1 break this setup? Will I have to re-do my setup or can I just do a straight update through the web interface of my pfSense box?
-
@DanielJay23 said in ATT Uverse RG Bypass (0.2 BTC):
I have the netgraph version working well on my system. I am currently running 2.4.4-RELEASE. Will upgrading to 2.4.5_1 break this setup? Will I have to re-do my setup or can I just do a straight update through the web interface of my pfSense box?
Before upgrading I would recommend you compare your current pfatt.sh script with the one if the repository here https://github.com/MonkWho/pfatt and see if any big changes are necessary.
One major change in 2.4.5 is that ng_etf module is now part of base pfsense and you don't need to compile and upload it anymore. And it's also loaded by default so we cleaned out those parts of the script as well. If you don't update your current pfatt.sh it might hang with an error and stop the boot process of your box. Depends what version of the script you are running. Early versions required that module to load and didn't have "-nq" flag that allowed system to skip if it's already loaded.
-
I'm trying to use the supplicant version of pfatt.sh, and my pfSense box hangs on 'waiting EAP for authorization...' during boot. I had to manually break and edit the relevant parts of config.xml to be able to boot properly. I think I'm using the latest version of the script, and I did follow the instructions here. I do have the necessary certificates extracted from an unused gateway device.
Any idea on what the problem might be?
Thanks!
-
@andrew_241 I had this issue too at some point. I believe it happened because I had a typo in the MAC field in the script.
Have you tried booting normally and then running the script by hand to see if it creates interfaces and authenticates? You can take the script apart and run one command at a time to try and see where the problem occurs.
I also need to look at the script and figure out a way for some sort of an escape sequence if it this happens on boot. So people are not stuck like you were. Heck there is still an incomplete "TODO" in the code from the original creator specifically for this issue with hanging on boot.
-
I'll try executing the script manually. Just to confirm, I'll have the WAN interface port (igb0) directly connected to the ONT, but with no mac spoofing on it. Am I correct in assuming that the script creates the interface that spoofs the MAC of the gateway device?
-
@andrew_241 when I initially set mine up my .pem files were not named correctly, so you may want to check that.
On mine, I have the mac address for the certs in the script and the mac address for my RG spoofed on my pfsense box. My certs did not come from my RG that's why I use 2 MACs
