[rbak-nsp] SE-400 IP Pool Help

Marcin Kuczera marcin at leon.pl
Mon Dec 8 07:09:10 EST 2008


Jeff Tantsura wrote:
> Hi Sherwin,
> 
> If you don't mind answering, why didn't you contact Redback's TAC in first
> place?

well, I may answer

The answers from that group are much quicker than answers via 
tech-support, in my case Ericsson, because of all the procedures.

Regards,
Marcin



> 
>> -----Original Message-----
>> From: redback-nsp-bounces at puck.nether.net [mailto:redback-nsp-
>> bounces at puck.nether.net] On Behalf Of Sherwin Ang
>> Sent: maandag 8 december 2008 12:04
>> To: David Freedman
>> Cc: redback-nsp at puck.nether.net
>> Subject: Re: [rbak-nsp] SE-400 IP Pool Help
>>
>> thank you so much, that works!  i was about to lose faith in Redback.
>>
>> On Sun, Dec 7, 2008 at 9:23 PM, David Freedman
>> <david.freedman at uk.clara.net> wrote:
>>> Yes, I've had stupid amounts of problems with the "summary-only" keyword
>> in
>>> the past,
>>> it is most certainly broken in my release (6.1.3.1)
>>>
>>> Since you are doing this in a VPN content (and hence can't filter iBGP),
>>> Remove the "redistribute subscriber" and all the aggregate addresses
>>> and only have "redistribute static" then maintain static supernets
>>> routed to null0 (ip route x.x.x.x/y null0)
>>>
>>> like:
>>>
>>> router bgp vpn
>>> redistribute static
>>> no redistribute subscriber address
>>> no aggregate-address 200.96.96.0/19 summary-only
>>> no aggregate-address 200.96.204.0/22 summary-only
>>> no aggregate-address 200.96.252.0/22 summary-only
>>> ip route 200.96.96.0/19 null0
>>> ip route 200.96.204.0/22 null0
>>> ip route 200.96.252.0/22 null0
>>>
>>> Dave.
>>>
>>> ------------------------------------------------
>>> David Freedman
>>> Group Network Engineering
>>> Claranet Limited
>>> http://www.clara.net
>>>
>>>
>>>
>>> -----Original Message-----
>>> From: redback-nsp-bounces at puck.nether.net on behalf of Sherwin Ang
>>> Sent: Sun 12/7/2008 11:46
>>> To: redback-nsp at puck.nether.net
>>> Subject: [rbak-nsp] SE-400 IP Pool Help
>>>
>>> Hello Redback List,
>>>
>>> i've added a new IP Pool to our SE-400 using this configuration:
>>>
>>>  interface IPPOOL
>>>   ip address 200.96.0.1/30
>>> !
>>>  interface POOL multibind lastresort
>>>   ip unnumbered IPPOOL
>>>   ip pool 200.96.96.0/19
>>>   ip pool 200.96.204.0/22
>>>   ip pool 200.96.252.0/22
>>>
>>>  subscriber default
>>>    ip address pool name POOL
>>>    timeout idle 10
>>>    ppp mtu 1492
>>>
>>> Everything works great with regards to the Pool on our PPPoE but the
>>> problem now is how the routes are redistributed and aggregated on our
>>> BGP.
>>>
>>> I have this configuration to aggregate and redistribute the routes.
>>>
>>>  router bgp vpn
>>>   address-family ipv4 unicast
>>>    export route-target 999:4
>>>    import route-target 999:4
>>>    redistribute connected
>>>    redistribute subscriber address
>>>    redistribute static
>>>    aggregate-address 200.96.96.0/19 summary-only
>>>    aggregate-address 200.96.204.0/22 summary-only
>>>    aggregate-address 200.96.252.0/22 summary-only
>>>
>>> it looks correct but when i take a look at the BGP peer's routing
>>> table, it's getting a lot of /32's from the subscriber pool.  i now
>>> have like 8,000+ of them and growing.  The routes are not advertised
>>> without redistribute subscriber.  I tried putting a route-map and a
>>> prefix-filter to match only the supernets (/19's,/22's) but since it's
>>> still sending /32's, it won't match.
>>>
>>> I hope someone can point me to the right direction.
>>>
>>> Thank you.
>>>
>>> -Sherwin
>>> _______________________________________________
>>> redback-nsp mailing list
>>> redback-nsp at puck.nether.net
>>> https://puck.nether.net/mailman/listinfo/redback-nsp
>>>
>>>
>> _______________________________________________
>> redback-nsp mailing list
>> redback-nsp at puck.nether.net
>> https://puck.nether.net/mailman/listinfo/redback-nsp
> 
> _______________________________________________
> redback-nsp mailing list
> redback-nsp at puck.nether.net
> https://puck.nether.net/mailman/listinfo/redback-nsp
> 



More information about the redback-nsp mailing list