A Real VXLAN Packet, Header by Header
A genuine VXLAN packet captured between a spine and a leaf, taken apart byte by byte from the outer MAC down to the ICMP echo request inside.
A genuine VXLAN packet captured between a spine and a leaf, taken apart byte by byte from the outer MAC down to the ICMP echo request inside.
An L2VNI build on a real Nexus fabric, including the vPC anycast VTEP mistake that produced an empty peer table and how it was fixed.
EVPN route types stop being abstract the moment you read them off a real fabric. Here are Types 2, 3 and 5 from a working Nexus lab.
BUM traffic still has to get everywhere in a VXLAN fabric. Here is how ingress replication handles it, and what you give up by not running PIM.
A spine-leaf underlay should be the least interesting part of your fabric. Here is the OSPF build that gets you there, with real Nexus output.
The vPC failure matrix, with the Type-1 mismatch demonstrated on real hardware and the peer-link and keepalive rows explained rather than faked.
A real NX-OS vPC build with the warnings it emits, the health output decoded field by field, and an honest look at orphan ports.
Why NX-OS rejects channel-group before feature lacp, how to read the port-channel flags legend, and what a suspended member actually means.
A practical orientation for your first console session on a Cisco Nexus: what the box is, what is turned on, which VRF you are in, and how to save it properly.
If you know IOS XE cold, most of NX-OS will feel familiar until it suddenly does not. Here are the six differences that bite on day one, with real Nexus 9000v output.
%CDP-4-NATIVE_VLAN_MISMATCH hands you both interfaces and both native VLANs in one line. The real damage is underneath it: untagged frames get re-homed at the trunk boundary and two VLANs quietly become one. Captured live on IOS XE 17.18.2 in CML.
An err-disabled port stays down forever by default, because nothing auto-recovers until you arm it. The full cause list from IOS XE 17.18.2, how to identify which one fired in ten seconds, and the captured proof that auto-recovery is a retry loop rather than a fix.