[j-nsp] unreachable indirect next-hop-static

amin amin amiensda at gmail.com
Mon Nov 30 00:04:50 EST 2009


Stefan,

Thanks for your explanation,

let me say if I configure MPLS between R3 and R2, and I install lo0 at R1,
could R4 reach that lo0 ?
as far as I know, I doubt it if using indirect static.

Regards,

Samin

On Mon, Nov 30, 2009 at 10:34 AM, Stefan Fouant <
sfouant at shortestpathfirst.net> wrote:

> Yep, as I mentioned before, the packet is getting to R3, but R3 has no
> route to the 10.10.10/30 subnet. This is the fundamental premise behing
> "hop-by-hop" routing. Each router along the path needs to consult it's
> routing table to find a match. If you don't like this behavior, use MPLS ;)
> The resolve keyword on R4 only allows R4 to recursively look in it's routing
> table for a path to reach an indirectly connected next-hop, it doesn't
> instill any intelligence on R3. When the packet gets to R3 it needs to know
> where to forward the packet. That is why your traceroute is returning ICMP
> unreachables after the first hop.
>
> Sorry for the top posting. I'm on BB.
>
>
> Stefan Fouant
> www.shortestpathfirst.net
>
> Sent from my Verizon Wireless BlackBerry
> ------------------------------
> *From: * amin amin <amiensda at gmail.com>
> *Date: *Mon, 30 Nov 2009 09:48:35 +0700
> *To: *Stefan Fouant<sfouant at shortestpathfirst.net>
> *Cc: *<juniper-nsp at puck.nether.net>
> *Subject: *Re: [j-nsp] unreachable indirect next-hop-static
>
> In indirect next-hop should it need route define at r3 for prefix
> 10.10.10/30?
> here is the configuration:
> logical-systems {
>     r1 {
>         interfaces {
>             ge-2/1/6 {
>                 unit 1 {
>                     vlan-id 1;
>                     ##
>                     ## invalid path element 'aggregated-ether-options'
>                     ##
>                     family inet {
>                         address 10.10.10.1/30;
>                     }
>                 }
>             }
>         }
>         routing-options {
>             static {
>                 route 0.0.0.0/0 next-hop 10.10.10.2;
>             }
>         }
>     }
>     r2 {
>         interfaces {
>             ge-2/1/7 {
>                 unit 1 {
>                     vlan-id 1;
>                     ##
>                     ## invalid path element 'aggregated-ether-options'
>                     ##
>                     family inet {
>                         address 10.10.10.2/30;
>                     }
>                 }
>                 unit 2 {
>                     vlan-id 2;
>                     ##
>                     ## invalid path element 'aggregated-ether-options'
>                     ##
>                     family inet {
>                         address 100.100.100.1/30;
>                     }
>                 }
>             }
>         }
>         protocols {
>             ospf {
>                 area 0.0.0.0 {
>                     interface ge-2/1/7.2;
>                 }
>             }
>         }
>     }
>     r3 {
>         interfaces {
>             ge-2/1/6 {
>                 unit 2 {
>                     vlan-id 2;
>                     ##
>                     ## invalid path element 'aggregated-ether-options'
>                     ##
>                     family inet {
>                         address 100.100.100.2/30;
>                     }
>                 }
>                 unit 3 {
>                     vlan-id 3;
>                     ##
>                     ## invalid path element 'aggregated-ether-options'
>                     ##
>                     family inet {
>                         address 20.20.20.1/30;
>                     }
>                 }
>             }
>         }
>         protocols {
>             ospf {
>                 area 0.0.0.0 {
>                     interface all;
>                 }
>             }
>         }
>     }
>     r4 {
>         interfaces {
>             ge-2/1/7 {
>                 unit 3 {
>                     vlan-id 3;
>                     ##
>                     ## invalid path element 'aggregated-ether-options'
>                     ##
>                     family inet {
>                         address 20.20.20.2/30;
>                     }
>                 }
>             }
>         }
>         protocols {
>             ospf {
>                 area 0.0.0.0 {
>                     interface all;
>                 }
>             }
>         }
>         routing-options {
>             static {
>                 route 10.10.10.0/30 {
>                     next-hop 100.100.100.1;
>                     resolve;
>                 }
>             }
>         }
>     }
> }
> interfaces {
>     ge-2/1/6 {
>         vlan-tagging;
>     }
>     ge-2/1/7 {
>         vlan-tagging;
>     }
> }
>
> [edit]
> lab# run show route 10.10.10.1 logical-system r3
>
> [edit]
> lab# run show route 10.10.10.1 logical-system r4
>
> inet.0: 5 destinations, 5 routes (5 active, 0 holddown, 0 hidden)
> + = Active Route, - = Last Active, * = Both
>
> 10.10.10.0/30      *[Static/5] 00:21:17, metric2 2
>                     > to 20.20.20.1 via ge-2/1/7.3
>
> [edit]
> lab# run show route 20.20.20.2 logical-system r1
>
> inet.0: 3 destinations, 3 routes (3 active, 0 holddown, 0 hidden)
> + = Active Route, - = Last Active, * = Both
>
> 0.0.0.0/0          *[Static/5] 00:20:40
>                     > to 10.10.10.2 via ge-2/1/6.1
>
> [edit]
> lab# run traceroute 10.10.10.1 logical-system r4
> traceroute to 10.10.10.1 (10.10.10.1), 30 hops max, 40 byte packets
>  1  20.20.20.1 (20.20.20.1)  0.388 ms !N  0.298 ms !N  0.273 ms !N
>
> [edit]
> lab# run traceroute 20.20.20.1 logical-system r1
> traceroute to 20.20.20.1 (20.20.20.1), 30 hops max, 40 byte packets
>  1  10.10.10.2 (10.10.10.2)  0.388 ms  0.289 ms  0.276 ms
> ^C
> [edit]
>
> rgds,
> samin
>
> On Sat, Nov 28, 2009 at 11:05 AM, Stefan Fouant <
> sfouant at shortestpathfirst.net> wrote:
>
>> > -----Original Message-----
>> > From: juniper-nsp-bounces at puck.nether.net [mailto:juniper-nsp-
>> > bounces at puck.nether.net] On Behalf Of amin amin
>> > Sent: Friday, November 27, 2009 7:38 PM
>> >
>> > Can Junos do indirect next-hop static, let me say
>> > r1-r2-r3-r4
>> > r4 should reach r1 using static route with gateway 100.100.100.1
>> > r1 using default route static
>> > r2-r3-r4 are using ospf protocol area 0
>> > r1's IP to r2 = 10.10.10.1/30
>> > r2's IP to r1 = 10.10.10.2/30
>> > r2's IP to r3 = 100.100.100.1/30
>> > r3's IP to r2 = 100.100.100.2/30
>> > r3's IP to r4 = 20.20.20.1/30
>> > r4's IP to r3 = 20.20.20.2/30
>> >
>> > when I check availble route 10.10.10/30 at r4, it's always shown up.
>> > but it
>> > never be reached
>> > the configuration at r4 for static route
>> > set routing-option static route 10.10.10/30 next-hop 100.100.100.1
>> > resolve
>> > set routing-option forwarding-table indirect-next-hop
>> >
>> >
>> > the configuration static route at r1
>> > set routing-option static route 0/0 next-hop 10.10.10.2
>>
>> What happens when you do a traceroute?  Based on your configs, I imagine
>> the
>> packets would get dropped on r3 as there is no route defined on r3 to
>> reach
>> the 10.10.10/30 subnet.
>>
>> Also, you do not need to do 'set routing-options forwarding-table
>> indirect-next-hop'.  That has nothing to do with the 'resolve' keyword
>> under
>> [edit routing-options static route].  Rather, it is used to improve route
>> convergence time by abstracting a level of recursion so that fewer route
>> to
>> forwarding next-hop bindings need to be updated when underlying next-hop
>> paths are changed.  More details on that can be found via the link below:
>>
>>
>> http://www.juniper.net/techpubs/software/junos/junos92/swconfig-routing/enab
>> ling-an-indirect-next-hop.html<http://www.juniper.net/techpubs/software/junos/junos92/swconfig-routing/enab%0Aling-an-indirect-next-hop.html>
>>
>> Stefan Fouant
>> www.shortestpathfirst.net
>> GPG Key ID: 0xB5E3803D
>>
>>
>
>
> --
> -------------------------------------------
>    http://aministi.web.id/
> -------------------------------------------
>



-- 
-------------------------------------------
   http://aministi.web.id/
-------------------------------------------


More information about the juniper-nsp mailing list