Path visibility, packet drop, and latency measurement with service chaining data flows

A system comprising:

• one or more processors;
• and
• one or more non-transitory computer-readable media storing instructions that, when executed, cause the one or more processors to perform operations comprising:
• determining, based at least in part on flow data and service chain configuration data associated with network switches of a data center fabric, a path that packets of a data flow are to traverse, the path including at least:
• first hop between a first network switch of the data center fabric and a service node disposed outside of the data center fabric, the first network switch configured to egress the packets from the data center fabric to the service node;
• and
• a second hop between a second network switch of the data center fabric and the service node, the second network switch configured to ingress the packets into the data center fabric from the service node;
• receiving, from the first network switch, header data that was included with a packet of the data flow when the packet was sent, to the service node, by the first network switch;
• receiving, from the second network switch, different header data that was included with the packet of the data flow when the packet was received, from the service node, by the second network switch;
• determining, based at least in part on the path, that the packet received at the second network switch is the packet that was sent to the service node by the first network switch;
• and
• calculating, based at least in part on the header data and the different header data, a latency associated with the service node.

One or more non-transitory computer-readable media storing instructions that, when executed, cause one or more processors to perform operations comprising:

• determining, based at least in part on flow data and service chain configuration data associated with network switches of a data center fabric, a path that packets of a data flow are to traverse, the path including at least:
• a first hop between a first network switch of the data center fabric and a service node disposed outside of the data center fabric, the first network switch configured to egress the packets from the data center fabric to the service node;
• and
• a second hop between a second network switch of the data center fabric and the service node, the second network switch configured to ingress the packets into the data center fabric from the service node;
• receiving, from the first network switch, header data that was included with a packet of the data flow when the packet was sent, to the service node, by the first network switch;
• receiving, from the second network switch, different header data that was included with the packet of the data flow when the packet was received, from the service node, by the second network switch;
• determining, based at least in part on the path, that the packet received at the second network switch is the packet that was sent to the service node by the first network switch;
• and
• calculating, based at least in part on the header data and the different header data, a latency associated with the service node.