Hello,
I recently faced to a strange problem, which i can't understand. Suppose
this network diagram:
MPLS Cloud --- PE (MX80) ---- CE (EX4200) ---- PE (MX80) ---- MPLS Cloud.
As You can see, EX4200 has two uplinks to two MX80, which are part of
MPLS network infrastructure. VPLS Multihoming is configured:
PE1 (gavar)
misak@mx-gavar> show configuration routing-instances Miom_ISP-Inet |
display inheritance | except ##
instance-type vpls;
vlan-id 342;
interface ge-1/1/9.342;
routing-interface irb.342;
route-distinguisher 10.255.255.33:342;
vrf-target target:65500:342;
protocols {
vpls {
site-range 100;
no-tunnel-services;
site martuni {
site-identifier 32;
multi-homing;
site-preference primary;
interface ge-1/1/9.342;
}
site gavar {
site-identifier 33;
}
mac-flush;
connectivity-type irb;
}
}
PE2 (getap)
instance-type vpls;
vlan-id 342;
interface ge-1/1/9.342;
route-distinguisher 10.255.255.10:342;
vrf-target target:65500:342;
protocols {
vpls {
site-range 100;
no-tunnel-services;
site martuni {
site-identifier 32;
multi-homing;
site-preference backup;
interface ge-1/1/9.342;
}
site getap {
site-identifier 10;
}
mac-flush;
}
}
CE site is called martuni, and EX4200 on that site is configured with
vlan-id all on both uplinks.
Now the problem - when one uplinks on EX4200 go down, or for example one
router (getap) goes down, I shouldn't have any problem with this routing
instance, but in this case interface ge-1/1/9.342 goes to CCC-Down state
and stops passing traffic.
There are lot of other VPLS configured on router and that links, and i
noticed that only VPLS-es that exist only on these two PEs show this
behavior. If You have VPLS configured on other PE router throughout MPLS
cloud, links stays up and work as expected.
I tried lot of knobs and options, but no luck. Can anyone explain this
behavior? Any help highly appreciated.
--
Best regards,
Misak Khachatryan,
_______________________________________________
juniper-nsp mailing list [email protected]
https://puck.nether.net/mailman/listinfo/juniper-nsp
