spanning tree and nexus 7k and 9k

[Richard Lucht]

We have 2 nexus chassis our current core is a nexus 7000 running 6.2(12) and a newer nexus 9516 that we just upgraded to 7.0(3)I7(6).  We had to do this to upgrade to replace the fabric and modules.  Prior to the upgrade we had the 9k hanging off the 7k with a couple of switches off the 9k.  After we performed the upgrade the link between the 9k and 7k broke.  Spanning on the 7k is putting all vlans in a block status on the interface to the 9k.  The 9k is trying to be the root despite having a lower priority that the 7k.  Currently we have the 2 nexus switches connected with a cisco 3850 in between.  We have the management interface plugged directly into the 7k so we are able to test the connections.  the int vlan we put on there is for our route out of the 9k for the default vrf, even though it says it is up up and we have an arp entry we are not able to ping out of the 9k to the gateway or anything else.  We get the message that there is no route to destination.  We have verified the ip route on the default vrf.  We have not other switch plugged into the 9k at this time.

I would like to add this in, the cisco switch I have plugged in between the 9k and 7k, I put an IP address on the same vlan that we have in both the 9k and 7k.  I can ping the ip address I put on the cisco switch but I can not ping the 9k IP address

From the 9k

IP Route Table for VRF "default"
'*' denotes best ucast next-hop
'**' denotes best mcast next-hop
'[x/y]' denotes [preference/metric]
'%<string>' in via output denotes VRF <string>

0.0.0.0/0, ubest/mbest: 1/0
*via 172.23.248.57, [1/0], 03:54:36, static
172.23.248.56/29, ubest/mbest: 1/0, attached
*via 172.23.248.58, Vlan3090, [0/0], 03:54:36, direct
172.23.248.58/32, ubest/mbest: 1/0, attached
*via 172.23.248.58, Vlan3090, [0/0], 03:54:36, local

[lovsharm]

Hi Richard, 

i am sorry, i am bit confuse with the connection currently and not working. can to put it topology wise and ip address what is not pingable and what is pingable?

Lovkesh

[Richard Lucht · Accepted solution]

the 9k is trunked to a switch then to the 7k.  no other connections.  We resolved this, I hooked an out of band connection and nuked the configuration.  I then rebuilt most of it and it is working as it should.  We think that when we upgraded the code and replaced the fabric and blades something got screwed up.  We started fresh with it and so far it is working as it should.  Thanks though.

[lovsharm]

Glad to know!

[amer@netsoftit.ae]

Hi team,

we are migrating pair of cisco nexus 7010 with a pair of nexus 9504 chassis, what will be correct road map? we need zero downtime solution.

if anyone has any suggestion, it will be highly appreciated.

currently cisco nexus 7010 chassis are our core switches, configured with VPC with redundant n7010. 

as per my understanding I can do this:

1) take the backup of existing configurations.

2) replicate same configuration on new pair of chassis.

3) shut down redundant chassis first, replace with new 9504.

4)plug back cables from n7010 back to 9504 backup.

5)replace primary n7010 with primary 9504 and then move cables from 7k to 9k chassis.

6)configure VPC on both 9k chassis back for redundancy.

please let me know any other plan of action i could take.