Skip to content

route_addition_during_failover_test

Github Action edited this page Mar 24, 2023 · 4 revisions

TE-13.1: gRIBI route ADD during Failover

Summary

Validate gRIBI route persistence during SSO

Procedure

  • Connect ATE port-1 to DUT port-1, and ATE port-2 to DUT port-2.

  • Create 64 L3 sub-interfaces under DUT port-2 and corresponding 64 L3 sub-interfaces on ATE port-2.

  • Establish gRIBI client connection with DUT negotiating FIB_ACK as the requested ack_type.

  • Install 64 L3 sub-interfaces IP to NextHopGroup(NHGID: 1) pointing to ATE port-2 as the nexthop.

  • Inject 1000 IPv4Entries(IPBlock1: 198.18.196.1/22) in default VRF with NHGID: 1.

  • Validate that the entries are installed as FIB_PROGRAMMED using getRPC.

  • Send traffic from ATE port-1 to prefixes in IPBlock1 and ensure traffic flows 100% and reaches ATE port-2.

  • Start injecting another 1000 IPv4Entries(IPBlock2: 198.18.100.1/22) in default VRF with NHGID: #1.

  • Check for coredumps in the DUT and validate that none are present.

  • Concurrently, trigger a supervisor switchover using gNOI SwitchControlProcessor  while IPBlock2 entries are only partially installed.

  • Check for coredumps in the DUT and validate that none are present post failover

  • Following reconnection of the gRIBI client to a new master supervisor, validate if partially ACKed entries of IPBlock2 are present as FIB_PROGRAMMED using a get RPC.

  • Re-inject IPBlock2 in default VRF with NHGID: #1.

  • Send traffic from ATE port-1 to prefixes in IPBlock2 and ensure traffic flows 100% and reaches ATE port-2.

Protocol/RPC Parameter coverage

  • gNOI:
    • System
      • SwitchControlProcessor

Config parameter coverage

Telemery parameter coverage

Clone this wiki locally