Kevin, MPLS VPN label must be unique per each vm/lxc on a given compute node. The forwarding based on that label is done *only* on a given compute node (SAFI 128 next hop).
So you can have the same labels across compute nodes as the next hop will be different. But on a given compute node same label would result in switching incoming packets to the same vm/lxc. Cheers, R. PS. Label is 20 bits so I am not sure if 2^20 makes you any close to reach the max VM limit per compute node :)) On Mon, Jun 27, 2016 at 8:31 PM, kevin parrikar <[email protected]> wrote: > Hello All, > In a hypothetical scenario running 5120 vms from one tenant,spread > across 50 compute running 100 vms each ,won't it exhaust MPLS labels and > prevent creation of new vms? > > Can MPLS label be duplicated for another tenant so that it can also have > 5120 vms > > Can you please help me understand this. > > Regards, > Kevin > > _______________________________________________ > Users mailing list > [email protected] > http://lists.opencontrail.org/mailman/listinfo/users_lists.opencontrail.org > >
_______________________________________________ Dev mailing list [email protected] http://lists.opencontrail.org/mailman/listinfo/dev_lists.opencontrail.org
