Mailing List Archive

Hi Yu-lin,

Please see in line:

> I use the APS on two sonet interfaces (so-1/1/1 and so-1/1/2)
> to protect
> the link between my routers and the SONET ADM.
>
> [diagram]
>
> +------(working-circuit)-------+ +
> -----(working-circuit)---------+
> R1 SONET ADM
> R2
> +-------(protection-circuit)------+
> +-------(protection-circuit)------+
>
> Where the so-1/1/1 as the "working-circuit" and so-1/1/2 as the
> "protection-circuit".
>
> The protection works fine except two behaviors that I cannot
> understand.
> First, when I commit the config, following warning messages appear,
> ----------------------------------
> noc@R1# commit
> [edit interfaces so-1/1/2 unit 0 family inet]
> 'address xxx.xxx.xxx.xxx/32'
> warning: identical local address is found on different interfaces
> [edit interfaces so-1/1/2 unit 1 family inet]
> 'address ooo.ooo.ooo.ooo/32'
> warning: identical local address is found on different interfaces
> commit complete
> -----------------------------

Any particular reason you're using a /32 on the PtoP interfaces? Just
curious.
Also, are you using channelized SONET interfaces? Just curious. I see
you have two logical interfaces on the same SONET interface.
Are the addresses the same on so-1/1/2 unit 0 and so-1/1/2 unit 1? It
appears as though these two logical interfaces may have the same
address.

> Second, when I unplug the fiber on so-1/1/1 (the working-circuit),
> the protection-circuit does take over and the traffic been switched
> to protection circuit. BUT, the ospf session between R1 and
> remote router
> (the R2) always been reset. (a ospf up/down message is logged
> in the log file)
>
> Are those two behaviors normal? ("identical local address
> warning" and "OSPF reset")

The router views each interface as a separate routing interfaces and
therefore warns you whenever two interfaces have the same address. This
is normal behavior.

The OSPF reset is normal. What version of JUNOS are you using? Some
enhancements have been made to the most recent version of JUNOS to
significantly enhance the speed at which recovery takes place.

HTH,
Steve

> Thanks in advance.
>
> Regards,
> Yu-lin
> --
>
> _______________________________________________
> juniper-nsp mailing list juniper-nsp@puck.nether.net
> http://puck.nether.net/mailman/listinfo/juniper-nsp
>

Hi Steve,

There is no paticular reason to use a /32 address on the interface.
Just because we use /32 before on a Cisco box. I try to use a /30
block on the interface, the same warning appears again.

The addresses are different on unit 0 and unit 1. I use the Frame-Relay
encapasulation on the SONET link, because DLCI can help me to have
two "virtual circuits" on one link.

The SONET interface is not channelized. The parts number of the
PIC is "750-003034" on R1 and "750-002781" on R2. They are both
4-port STM-1 SMIR PIC. JUNOS on R1 is 6.0R1.3, and on R2 is 5.4R2.4

Thank you again.


Regards,
Yu-lin

----- Original Message -----
From: "Steve Holman" <sholman@juniper.net>
To: "Yu-lin Chang" <ylchang@ascc.net>; <juniper-nsp@puck.nether.net>
Cc: <juniper@groupstudy.com>
Sent: Sunday, September 28, 2003 10:39 AM
Subject: RE: [j-nsp] SONET APS warning?


Hi Yu-lin,

Please see in line:

Any particular reason you're using a /32 on the PtoP interfaces? Just
curious.
Also, are you using channelized SONET interfaces? Just curious. I see
you have two logical interfaces on the same SONET interface.
Are the addresses the same on so-1/1/2 unit 0 and so-1/1/2 unit 1? It
appears as though these two logical interfaces may have the same
address.

> Second, when I unplug the fiber on so-1/1/1 (the working-circuit),
> the protection-circuit does take over and the traffic been switched
> to protection circuit. BUT, the ospf session between R1 and
> remote router
> (the R2) always been reset. (a ospf up/down message is logged
> in the log file)
>
> Are those two behaviors normal? ("identical local address
> warning" and "OSPF reset")

The router views each interface as a separate routing interfaces and
therefore warns you whenever two interfaces have the same address. This
is normal behavior.

The OSPF reset is normal. What version of JUNOS are you using? Some
enhancements have been made to the most recent version of JUNOS to
significantly enhance the speed at which recovery takes place.

HTH,
Steve