Nexus 7000 failover question

[Marius Gunnerud]

Hi,

We have a datasenter which spans over two locations and HSRP is configured for all VLAN interfaces between the sites.  Each Nexus 7K has 3 VDCs (one for MPLS/routing, one for FabricPath / L2 connectivity, and the third for OTV).

When we pull cables out of the active data center location we have subsecond failover....no issues.  The firewalls also do subsecond failover no issues when there is a link failure.

When we take the whole active data center off line (pull the power cable out) we loose 8 pings / packets, which is acceptable in such a situation.

The issue comes when we bring the data center back online.  When we do this we loose 80 pings / packets.

The VDCs all have default boot order configuration so they al boot simultaniously.  I am speculating that the packet loss comes because of the order in which the line cards come online, but i do not know this for sure.  Has anyone come across this  before and know if there is a way to minimize the packet loss when powering up the Nexus after a powerfailure?

Thanks

[Richard Michael]

you need to put more information on the following,

1) Detailed network topology

2) Is the ping loss happening while the vpc+ peer was coming up? Do you have also M2 modules in place? modules list please

3) From where to where you are pinging and seeing losss?