NEXUS 7718 VDC ISSUE, getting letency when pinging any SVI

[Prakash1]

we have 2 vdc in 2 physical box(7718) and vpc between theM but when i am pinging vdc1 l3 vlan getting high latency and while pinging vdc2 l3 vlan ping hop are fine.

 SW01 vdc1
Reply from 10.142.33.2: bytes=32 time=3ms TTL=255
Reply from 10.142.33.2: bytes=32 time=539ms TTL=255
Reply from 10.142.33.2: bytes=32 time=2ms TTL=255
Reply from 10.142.33.2: bytes=32 time=578ms TTL=255
Reply from 10.142.33.2: bytes=32 time=2ms TTL=255
Reply from 10.142.33.2: bytes=32 time=447ms TTL=255
Reply from 10.142.33.2: bytes=32 time=2ms TTL=255
Reply from 10.142.33.2: bytes=32 time=451ms TTL=255

 vdc2
Reply from 10.142.7.3: bytes=32 time=1ms TTL=254
Reply from 10.142.7.3: bytes=32 time=1ms TTL=254
Reply from 10.142.7.3: bytes=32 time=2ms TTL=254
Reply from 10.142.7.3: bytes=32 time=2ms TTL=254
Reply from 10.142.7.3: bytes=32 time=2ms TTL=254
Reply from 10.142.7.3: bytes=32 time=1ms TTL=254
Reply from 10.142.7.3: bytes=32 time=1ms TTL=254
Reply from 10.142.7.3: bytes=32 time=1ms TTL=254

what may be exact issue,any suggestions are welcome.

[Georg Pauwen]

From where to where are you pinging, from VDC1 to VDC2 ? Across the VPC ?

[Prakash1]

Hello Georg,
scenario is like this.
VDC1 have one L3 vlan and same L3 vlan is at VDC2 and hsrp is also configured.
vdc1 vlan111 10.31.2.2
vdc2 vlan111 10.31.2.3
hsrp vip 10.31.2.1

[jkilleda]

Hello Prakash,

NEXUS 7718 VDC ISSUE, getting letency when pinging any SVI
Check if you have Igmp join-group on SVI configured .

[Prakash1]

Hello,

the issue got fixed by implementing one solution on our nexus device.

as previous, we were running Pvst and we have approximately 150 L3 VLANs, and 150 access layer neighbor.

This situation was creating multiple logical ports in the device as per N7K its maximum limit is 16000. so as I observed nexus gives less priority to ICMP packets and ssh packets.

There is command by which check how much logical ports are in the device.

Switch#"sh spanning-tree internal info global"

Total stp_ports*instance:11263

Total ports vlan:11263

Total physical port*vlan :11682

as we have above situation because we did trunk all VLANs toward access layer and vice versa.

after investigating further we did VLAN pruning and amazingly situation got changed.

Switch#"sh spanning-tree internal info global"

Total stp_ports*instance:2000

Total ports vlan:2000

Total physical port*vlan :2145

after this, we are getting 1MS reply to pinging any interface ping.