I've been having a read of the following document:
And to be honest, it just is not working....
I have an SRX340 that will have ethernet as it's primary link, but I want the DSL to be back up if the primary fails....
Anyone got a basic config I can use for ppp connectivity over ADSL to an LNS?
I probably need to explain the requirments a little better:
The VDSL2 MPIM will be acting like an ADSL CPE (as a backup if the primary Ethernet Circuit fails).
So, I need to ensure it is configured as per a CPE, automated RADIUS assigned (IPCP) IP address. authentication via PPP etc etc...
I have a working Cisco one and here is the Config from that device:
interface ATM0 no ip address no atm ilmi-keepalive bridge-group 1 pvc 0/38 encapsulation aal5mux ppp dialer dialer pool-member 1
interface Dialer0 ip address negotiated ip nat outside ip virtual-reassembly in encapsulation ppp dialer pool 1 dialer-group 1 ppp authentication chap callin ppp chap hostname email@example.com ppp chap password 7 helpmeifyoucan
So, I need something similar to this in Juniper..... Any help would be great?
Oh, almost forgot, I have the Sync light as Green and solid....
I have configured the following, so I think I am almost there, but missing something:
set interfaces at-4/0/0 encapsulation atm-pvcset interfaces at-4/0/0 atm-options vpi 0set interfaces at-4/0/0 dsl-options operating-mode autoset interfaces at-4/0/0 unit 0 encapsulation atm-ppp-vc-muxset interfaces at-4/0/0 unit 0 vci 0.33set interfaces at-4/0/0 unit 0 ppp-options chap default-chap-secret "$9$w4soGDjqfQnHqIclMN-HqmP5F"set interfaces at-4/0/0 unit 0 ppp-options chap local-name "firstname.lastname@example.org"set interfaces at-4/0/0 unit 0 ppp-options chap passiveset interfaces at-4/0/0 unit 0 ppp-options pap local-name "email@example.com"set interfaces at-4/0/0 unit 0 ppp-options pap local-password "$9$Web8NbsYoGjqgo/tORlegoJZUH"set interfaces at-4/0/0 unit 0 ppp-options pap passiveset interfaces at-4/0/0 unit 0 family inet negotiate-address
When I look at interfaces terse I see the following:
at-4/0/0 up upat-4/0/0.0 up down inetat-4/0/0.32767 up up
And if I complete the following command:
run show ppp statistics - to see where the LCP process is at....
Session statistics from PPP process Total sessions: 1 Sessions in disabled phase : 0 Sessions in establish phase : 1 Sessions in authenticate phase: 0 Sessions in network phase : 0 Bundles in pending phase : 0
Not exactly sure of the readout......
Any help please?
Okay, I have it working. This ADSL is to be used as a backup in case the ethernet circuit goes down, so I have also configured a default route with preference.... here is the working config:
I have installed the VDSL module into slot 4... for reference:
set interfaces at-4/0/0 encapsulation atm-pvcset interfaces at-4/0/0 atm-options vpi 0set interfaces at-4/0/0 dsl-options operating-mode autoset interfaces at-4/0/0 unit 0 encapsulation atm-ppp-vc-muxset interfaces at-4/0/0 unit 0 vci 0.38set interfaces at-4/0/0 unit 0 ppp-options chap default-chap-secret "$9$w4soGDjqfQnHqIclMN-HqmP5F"set interfaces at-4/0/0 unit 0 ppp-options chap local-name "firstname.lastname@example.org"set interfaces at-4/0/0 unit 0 ppp-options chap passiveset interfaces at-4/0/0 unit 0 ppp-options pap local-name "email@example.com"set interfaces at-4/0/0 unit 0 ppp-options pap local-password "$9$Web8NbsYoGjqgo/tORlegoJZUH"set interfaces at-4/0/0 unit 0 ppp-options pap passiveset interfaces at-4/0/0 unit 0 family inet negotiate-address
set routing-options static route 0.0.0.0/0 next-hop x.x.x.xset routing-options static route 0.0.0.0/0 qualified-next-hop at-4/0/0.0 preference 25
The above post shows the working ADSL from the NTE perspective. If I pull the ethernet connection it does route through the backup ADSL, exactly as I want.
The problem I still have is the other end..... Let's have a closer look at the connection through the downstream ISP between my Core PE and the Customer NTE device:
ge-0/0/1 (192.168.1.1/30) NTE ge-0/0/15.10 (10.20.30.2/30) --> S-Tag (Downstream ISP) --> xe-1/2/4.10 (10.20.30.1/30) Core
So, although the NTE recognises the ethernet cable has been pulled, the Core does NOT see the route as being down and still tries to forward the traffic out of the xe-1/2/4.10 interface instead of routing to the LNS and ADSL.....
This must be because of the layer 2 connectivity between the downstream ISP and the Core...... it never see's the route go down so will never fail over...... I will discuss options with the ISP but wanted to get your thoughts on this?
Okay. The issue could be a little more complicated than originally thought.
Given that we do not know for sure what the customer would order, we are looking at the following as a scenario:
NTE device = SRX340 - Ethernet / VDSL
The NTE device has the following ports configured:
ge-0/0/1 - towards the Customer CPE
ge-0/0/15.10 - Customer VLAN to the PE Core
ge-0/0/15.99 - Management VLAN to NTE (Only between PE core and NTE)
at-4/0/0 - DSL Failover
If ge-0/0/15 fails then all traffic from and to the customer will go through at-4/0/0 interface.
Possible issues with scenario requirement:
1: How the PE Core will detect that the ge-0/0/15 interface has failed?
2: The LCP (IPCP) negotiated address on the at-4/0/0 interface will have to be different than the interface on the CPE and then routing has to be in place for that?
3: How do ISPs already get around the above issues?
I note the following forum post has some information, but it does not explain it from an ISP perspective:
Any help would be greatly appreciated.
This is more to keep people updated so they can try too.....
I have completed the following:
Moved the at-4/0/0 interface into the untrust zone.
Now I have completed source NAT. This overcomes both the usage of the same address (CPE and AT-4/0/0 interfaces) and the core not recognising the loss of the interface/route.....
Whether this is an end resolution or not remains to be seen and tested.....
The NAT is between the untrust and trust zone and as the ONLY interface in the untrust is at-4/0/0 then there is no effect when the ethernet is up.
I will close this discussion as I believe the issue with configuring the VDSL2 MPIM is complete. It is working as expected. I do have another issue but it is not with the module, so I will open a new discussion.