Solved: How many IPv4 routes can a Nexus 93180YC-EX carry?

SystemAdministrator7710 · ‎03-01-2021

I am using an N9K as an edge router, peering with my ISP's BGP speaker:

edge-rtr# sh ip route summary
IP Route Table for VRF "default"
Total number of routes: 28931
Total number of paths: 28932

[...]

edge-rtr# sh ip route bgp summary
IP Route Table for VRF "default"
Total number of routes: 28866
Total number of paths: 28866

Best paths per protocol: Backup paths per protocol:
bgp-xxxxx: 28866 None

Number of routes per mask-length:
/8 : 2 /9 : 1 /10: 2 /11: 5 /12: 14
/13: 22 /14: 65 /15: 85 /16: 984 /17: 222
/18: 328 /19: 877 /20: 1641 /21: 1650 /22: 2424
/23: 3273 /24: 17271

(1) How many routes can the N9K carry?

(2) I interpret the output below to suggest that this box has consumed ~1/3 (more precisely: 35.97%) of its available TCAM space for IPv4 routes. Is there a way to increase this, perhaps by sacrificing TCAM space for IPv6 or multicast or other functions?

edge-rtr# show hardware internal forwarding table utilization | include v4|V4
IPv4/IPv6 hosts and routes summary on module : 1
Max IPv4 Trie route entries: 458752
Max V4 TCAM table entries : 6144
Max native host route entries (shared v4/v6) : 65536
Total number of IPv4 host trie routes used : 0
Total number of IPv4 host tcam routes used : 1
Total number of IPv4 LPM trie routes used : 26665
Total number of IPv4 LPM tcam routes used : 2209
Total number of IPv4 host native routes used in native tiles : 61
IPv4 Host-as-Route count : 0
Percentage utilization of IPv4 native host routes : 0.09
Percentage utilization of IPv4 trie routes : 5.81
Percentage utilization of IPv4 TCAM routes : 35.97
IPv4/Ipv6 Mcast host entry summary
Max Mcast V4 SA TCAM entries (S/m) = 2048
Max Overflow Mcast v4 SA TCAM entries = 0
Max Overflow Mcast V4 DA TCAM entries = 0
Total number of IPv4 Multicast SA LPM routes used = 2
Percentage utilization of IPv4 Multicast SA LPM routes = 0.09
edge-rtr#

(3) What happens when IPv4 route TCAM fills up? Does TCAM lock, requiring a reboot to recover? Or does it reject new routes (until such time as old routes are withdrawn)?

--sk

Sergiu.Daniluk · ‎03-01-2021

Hi @SystemAdministrator7710

First, I just want to say that N9K is not the best equipment to be placed as an edge router.

The scalability guide gives you the unidimentional scalability values:

https://www.cisco.com/c/en/us/td/docs/switches/datacenter/nexus9000/sw/92x/scalability/guide_923/b_Cisco_Nexus_9000_Series_NX-OS_Verified_Scalability_Guide_923.html (See. Table 13)

If increasing the routing table space (LPM) is what you are looking for, then you should change the system routing mode to internet-peering:

N9K# show system routing mode
N9K# config
N9K(config)# system routing template-internet-peering

Check the scalability guide for the number of routes (LPM or Hosts) in case of internet-peering vs lpm-heavy.

Cheers,

Sergiu

View solution in original post

Sergiu.Daniluk · ‎03-01-2021