Conssider this L2VPN (martini): A-END: unit 101 { encapsulation vlan-ccc; vlan-tags outer 101; output-vlan-map swap; family ccc; }
(SVLAN 101 has CVLAN 42, 69, 666 below it) B-END: phys { encapsulation ethernet-ccc; unit 0 { input-vlan-map { push; vlan-id 101; ## /MUST/ match to A-END SVLAN, otherwise A-END is down } output-vlan-map pop; family ccc; } } } Everything works just fine. Only I find it really strange B-END cannot push arbitrary S-VLAN, considering A-END is going to change it anyhow. If it's not 101, A-END vill be down with 'vlan-mismatch'. Now consider in B-END if some of the SVLANs need to go to the L2VPN circuit and some SVLAN needs to be locally terminated. A-END is unchanged. B-END, attempt1: unit 42 { encapsulation vlan-ccc; vlan-id-list [ 42 69 666 ]; input-vlan-map { push; vlan-id 101; } output-vlan-map pop; family ccc; } ==> B-END is up, A-END is down (vlan-mismatch) B-END, attempt2: native-vlan-id 101; unit 42 { encapsulation vlan-ccc; vlan-tags outer 101 inner-list [ 42 69 666 ]; output-vlan-map swap; family ccc; } ===> Both sides up, does not work (I've not verified non-workiness myself) I do not want to change A-end configuration. Only B-end can be changed. Is there any workable solution, or example with neither attempt works? -- ++ytti _______________________________________________ juniper-nsp mailing list juniper-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/juniper-nsp