-
Notifications
You must be signed in to change notification settings - Fork 148
weighted_ecmp_test
This is to ensure that,
-
Implementations can be configured for weighted equal cost multipath (ECMP) routing for IS-IS neighbors that are one hop away.
-
When WECMP is enabled, traffic destined to an IS-IS route represented by a multipath set of next-hop interfaces will be unequally distributed across the interfaces based on their bandwidth.
Each LAG bundle below is made up of 2x100G ports.
graph LR;
A[ATE1:LAG1] <-- IBGP+IS-IS --> B[LAG1:DUT];
C[DUT:LAG2] <-- IBGP+IS-IS --> D[LAG1:ATE2];
E[DUT:LAG3] <-- IBGP+IS-IS --> F[LAG2:ATE2];
G[DUT:LAG4] <-- IBGP+IS-IS --> H[LAG3:ATE2];
In the topology above,
-
Configure 1xLAG interface between ATE1<->DUT and 3xLAG interfaces between DUT and ATE2. Each LAG interface is expected to be of 2x100Gbps
-
Configure IPv4 and IPv6 L2 adjacencies between DUT and ATE LAG bundles. Therefore, DUT will have 1xIS-IS adjacency with ATE1 i.e. DUT:LAG1<->ATE1:LAG1, and 3xIS-IS adjacencies with ATE2 i.e. DUT:LAG2<->ATE2:LAG1, DUT:LAG3<->ATE2:LAG2 and DUT:LAG4<->ATE2:LAG3
-
/network-instances/network-instance/protocols/protocol/isis/global/afi-safi
-
/network-instances/network-instance/protocols/protocol/isis/global/config/level-capability, set to LEVEL_2
-
/network-instances/network-instance/protocols/protocol/isis/levels/level/config/metric-style set to WIDE_METRIC
-
-
Configure IPv4 and IPv6 IBGP peering between both ATEs and the DUT using their loopback addresses for both IPv4 and IPv6 address families.
- /network-instances/network-instance/protocols/protocol/bgp/peer-groups/peer-group/afi-safis/afi-safi/config
-
Attach a network with an IPv4 and an IPv6 prefix to ATE2 and have it advertise these prefixes over its IBGP peering with the DUT. The DUT in turn should advertise these prefixes over its IBGP peering with ATE1
- Please use
IPv4 prefix = 100.0.1.0/24andIPv6 prefix = 2001:db8:64:64::/64
- Please use
-
Similarly, attach a different network to ATE1 with IPv4 and IPv6 prefixes and advertise the same over its IBGP peering with the DUT.
- Please use
IPv4 prefix = 100.0.2.0/24andIPv6 prefix = 2001:db8:64:65::/64
- Please use
-
On the DUT, enable WECMP loadbalancing for multipath IS-IS routes and set the load-balancing-weight to use LAG bandwidth.
-
/network-instances/network-instance/protocols/protocol/isis/global/config/weighted-ecmp set to Enabled
-
/network-instances/network-instance/protocols/protocol/isis/interfaces/interface/weighted-ecmp/config/load-balancing-weight set to Auto
-
-
Start 1024 flows from IPv4 addresses in 100.0.2.0/24 to 100.0.1.0/24
-
Start 1024 flows from IPv6 addresses in 2001:db8:64:65::/64 to 2001:db8:64:64::/64
-
Ensure that the total traffic of all flows combined is ~20Gbps
-
Ensure that the DUT has learnt the routes for prefixes 100.0.1.0/24 and 2001:db8:64:64::/64 over IBGP. Following paths
- /network-instances/network-instance/afts/next-hops/next-hop/state/ip-address
-
Ensure that the DUT has learnt routes to the IPv4 and IPv6 loopback addresses of ATE2. It is expected that these prefixes are reachable via 3 different Next-Hop addresses corresponding to the LAG1, LAG2 and LAG3 interfaces on ATE2.
-
It is expected that the IS-IS instance in DUT will equally distribute the traffic received on DUT:LAG1 over the LAG bundles corresponding to ATE2:LAG1, ATE2:LAG2 and ATE2:LAG3 when the 3 LAG bundles have the same bandwidth available.
-
Traffic distribution between DUT:LAG2, DUT:LAG3 and DUT:LAG4 is expected to be ~33% each of the total traffic received on DUT:LAG1.
-
Check for the following paths
-
/network-instances/network-instance/protocols/protocol/isis/global/state/weighted-ecmp, should be true
-
/network-instances/network-instance/protocols/protocol/isis/interfaces/interface/weighted-ecmp/state/load-balancing-weight, should be auto
-
/interfaces/interface/state/counters/out-pkts
-
/interfaces/interface/state/counters/in-pkts
-
-
-
Stop traffic from RT-9.1 and introduce a failure by disabling one of the member interfaces in ATE2:LAG1.
-
Restart 1024 flows from IPv4 addresses in 100.0.2.0/24 to 100.0.1.0/24
-
Restart 1024 flows from IPv6 addresses in 2001:db8:64:65::/64 to 2001:db8:64:64::/64
-
Ensure that the total traffic of all flows combined is ~20Gbps
-
It is expected that the IS-IS instance in DUT will unequally distribute the traffic received from ATE1:LAG1 over the LAG bundles corresponding to ATE2:LAG1, ATE2:LAG2 and ATE3:LAG3.
-
Traffic on DUT:LAG2 is expected to be ~20% while traffic on DUT:LAG3 and DUT:LAG4 is expected to be ~40% each of the total traffic received on DUT:LAG1. If the traffic is not unequally shared between the DUT LAG bundles towards ATE2 then this test is a failure.
-
Check for the following paths
-
/interfaces/interface/state/counters/out-pkts
-
/interfaces/interface/state/counters/in-pkts
-
-
-
/network-instances/network-instance/protocols/protocol/bgp/peer-groups/peer-group/config
-
/network-instances/network-instance/protocols/protocol/bgp/peer-groups/peer-group/afi-safis/afi-safi/config
-
/network-instances/network-instance/protocols/protocol/isis/global/afi-safi
-
/network-instances/network-instance/protocols/protocol/isis/global/config/level-capability
-
/network-instances/network-instance/protocols/protocol/isis/levels/level/config/metric-style
-
/network-instances/network-instance/protocols/protocol/isis/global/config/weighted-ecmp
-
/network-instances/network-instance/protocols/protocol/isis/interfaces/interface/weighted-ecmp/config/load-balancing-weight
-
/routing-policy/defined-sets/prefix-sets/prefix-set/
-
/routing-policy/defined-sets/prefix-sets/prefix-set/prefixes/prefix/config/ip-prefix
-
/routing-policy/defined-sets/prefix-sets/prefix-set/prefixes/prefix/config/masklength-range/exact
-
/routing-policy/policy-definitions/policy-definition/config/name
-
/routing-policy/policy-definitions/policy-definition/statements/statement/config/name
-
/routing-policy/policy-definitions/policy-definition/statements/statement/conditions/match-prefix-set/config/prefix-set
-
/routing-policy/policy-definitions/policy-definition/statements/statement/conditions/match-prefix-set/config/match-set-options
-
/routing-policy/policy-definitions/policy-definition/statements/statement/actions/config/policy-result/ACCEPT_ROUTE
-
/network-instances/network-instance/protocols/protocol/bgp/neighbors/peer-group/afi-safis/afi-safi/apply-policy/config/import-policy
-
/network-instances/network-instance/protocols/protocol/bgp/neighbors/peer-group/afi-safis/afi-safi/apply-policy/config/export-policy
- /network-instances/network-instance/protocols/protocol/isis/global/state/weighted-ecmp
- /network-instances/network-instance/protocols/protocol/isis/interfaces/interface/weighted-ecmp/state/load-balancing-weight
- /interfaces/interface/state/counters/out-pkts
- /interfaces/interface/state/counters/in-pkts
-
Home
- Test Plans
- ACCTZ-1.1: Record Subscribe Full
- ACCTZ-2.1: Record Subscribe Partial
- ACCTZ-3.1: Record Subscribe Non-gRPC
- ACCTZ-4.1: Record History Truncation
- ACCTZ-4.2: Record Payload Truncation
- Authz: General Authz (1-4) tests
- CNTR-1: Basic container lifecycle via
gnoi.Containerz. - CNTR-2: Container network connectivity tests
- Credentialz-1: Password console login
- Credentialz-2: SSH Password Login Disallowed
- Credentialz-3: Host Certificates
- Credentialz-4: SSH Public Key Authentication
- Credentialz-5: Hiba Authentication
- DP-1.2: QoS policy feature config
- DP-1.3: QoS ECN feature config
- DP-1.4: QoS Interface Output Queue Counters
- DP-1.7: One strict priority queue traffic test
- DP-1.8: Two strict priority queue traffic test
- DP-1.9: WRR traffic test
- DP-1.10: Mixed strict priority and WRR traffic test
- DP-1.11: Bursty traffic test
- DP-1.14: QoS basic test
- example-0.1: Topology Test
- FP-1.1: Power admin DOWN/UP Test
- gNMI-1.1: cli Origin
- gNMI-1.2: Benchmarking: Full Configuration Replace
- gNMI-1.3: Benchmarking: Drained Configuration Convergence Time
- gNMI-1.4: Telemetry: Inventory
- gNMI-1.5: Telemetry: Port Speed Test
- gNMI-1.8: Configuration Metadata-only Retrieve and Replace
- gNMI-1.9: Get requests
- gNMI-1.10: Telemetry: Basic Check
- gNMI-1.11: Telemetry: Interface Packet Counters
- gNMI-1.12: Mixed OpenConfig/CLI Origin
- gNMI-1.13: Optics Telemetry, Instant, threshold, and miscellaneous static info
- gNMI-1.14: OpenConfig metadata consistency during large config push
- gNMI-1.15: Set Requests
- gNMI-1.16: fabric redundancy test
- gNMI-1.17: Controller Card redundancy test
- gNMI-1.18: gNMI subscribe with sample mode for backplane capacity counters
- gNMI-1.19: ConfigPush after Control Card switchover
- gNMI-1.20: Telemetry: Optics Thresholds
- gNMI-1.21: Integrated Circuit Hardware Resource Utilization Test
- gNMI-1.22: Controller card port attributes
- gNMI-1.27: gNMI Sample Mode Test
- GNMI-2: gnmi_subscriptionlist_test
- gNOI-2.1: Packet-based Link Qualification
- gNOI-3.1: Complete Chassis Reboot
- gNOI-3.2: Per-Component Reboot
- gNOI-3.3: Supervisor Switchover
- gNOI-3.4: Chassis Reboot Status and Reboot Cancellation
- gNOI-4.1: Software Upgrade
- gNOI-5.1: Ping Test
- gNOI-5.2: Traceroute Test
- gNOI-5.3: Copying Debug Files
- gNOI-6.1: Factory Reset
- Health-1.1: Generic Health Check
- Health-1.2: Healthz component status paths
- MGT-1: Management HA solution test
- MTU-1.3: Large IP Packet Transmission
- OC-1.2: Default Address Families
- OC-26.1: Network Time Protocol (NTP)
- P4RT-1.1: Base P4RT Functionality
- P4RT-1.2: P4RT Daemon Failure
- P4RT-2.1: P4RT Election
- P4RT-2.2: P4RT Metadata Validation
- P4RT-3.1: Google Discovery Protocol: PacketIn
- P4RT-3.2: Google Discovery Protocol: PacketOut
- P4RT-3.21: Google Discovery Protocol: PacketOut with LAG
- P4RT-5.1: Traceroute: PacketIn
- P4RT-5.2: Traceroute Packetout
- P4RT-5.3: Traceroute: PacketIn With VRF Selection
- P4RT-6.1: Required Packet I/O rate: Performance
- P4RT-7.1: LLDP: PacketIn
- P4RT-7.2: LLDP: PacketOut
- Replay-1.0: Record/replay presession test
- Replay-1.1: Record/replay diff command trees test
- Replay-1.2: P4RT Replay Test
- RT-1.1: Base BGP Session Parameters
- RT-1.2: BGP Policy & Route Installation
- RT-1.3: BGP Route Propagation
- RT-1.4: BGP Graceful Restart
- RT-1.5: BGP Prefix Limit
- RT-1.7: Local BGP Test
- RT-1.10: BGP Keepalive and HoldTimer Configuration Test
- RT-1.11: BGP remove private AS
- RT-1.12: BGP always compare MED
- RT-1.14: BGP Long-Lived Graceful Restart
- RT-1.19: BGP 2-Byte and 4-Byte ASN support
- RT-1.21: BGP TCP MSS and PMTUD
- RT-1.23: BGP AFI SAFI OC DEFAULTS
- RT-1.24: BGP 2-Byte and 4-Byte ASN support with policy
- RT-1.25: Management network-instance default static route
- RT-1.26: Basic static route support
- RT-1.27: Static route to BGP redistribution
- RT-1.28: BGP to IS-IS redistribution
- RT-1.29: BGP chained import/export policy attachment
- RT-1.30: BGP nested import/export policy attachment
- RT-1.32: BGP policy actions - MED, LocPref, prepend, flow-control
- RT-1.33: BGP Policy with prefix-set matching
- RT-1.34: BGP route-distance configuration
- RT-1.51: BGP multipath ECMP
- RT-1.52: BGP multipath UCMP support with Link Bandwidth Community
- RT-1.53: prefix-list test
- RT-1.55: BGP session mode (active/passive)
- RT-2.1: Base IS-IS Process and Adjacencies
- RT-2.2: IS-IS LSP Updates
- RT-2.6: IS-IS Hello-Padding enabled at interface level
- RT-2.7: IS-IS Passive is enabled at interface level
- RT-2.8: IS-IS metric style wide not enabled
- RT-2.9: IS-IS metric style wide enabled
- RT-2.10: IS-IS change LSP lifetime
- RT-2.11: IS-IS Passive is enabled at the area level
- RT-2.12: Static route to IS-IS redistribution
- RT-2.13: Weighted-ECMP for IS-IS
- RT-2.14: IS-IS Drain Test
- RT-2.15: IS-IS Graceful Restart Helper
- RT-3.1: Policy based VRF selection
- RT-3.2: Multiple <Protocol, DSCP> Rules for VRF Selection
- RT-4.10: AFTs Route Summary
- RT-5.1: Singleton Interface
- RT-5.2: Aggregate Interfaces
- RT-5.3: Aggregate Balancing
- RT-5.4: Aggregate Forwarding Viable
- RT-5.5: Interface hold-time
- RT-5.6: Interface Loopback mode
- RT-5.7: Aggregate Not Viable All
- RT-5.8: IPv6 Link Local
- RT-5.9: Disable IPv6 ND Router Arvetisment
- RT-5.10: IPv6 Link Local generated by SLAAC
- RT-6.1: Core LLDP TLV Population
- RT-7.1: BGP default policies
- RT-7.2: BGP Policy Community Set
- RT-7.3: BGP Policy AS Path Set
- RT-7.4: BGP Policy AS Path Set and Community Set
- RT-7.5: BGP Policy - Match and Set Link Bandwidth Community
- RT-7.8: BGP Policy Match Standard Community and Add Community Import/Export Policy
- RT-7.11: BGP Policy - Import/Export Policy Action Using Multiple Criteria
- RT-14.2: GRIBI Route Test
- SEC-3.1: Authentication
- SFLOW-1: sFlow Configuration and Sampling
- System-1: System testing
- TE-1.1: Static ARP
- TE-1.2: My Station MAC
- TE-2.1: gRIBI IPv4 Entry
- TE-2.2: gRIBI IPv4 Entry With Aggregate Ports
- TE-3.1: Base Hierarchical Route Installation
- TE-3.2: Traffic Balancing According to Weights
- TE-3.3: Hierarchical weight resolution
- TE-3.5: Ordering: ACK Received
- TE-3.6: ACK in the Presence of Other Routes
- TE-3.7: Base Hierarchical NHG Update
- TE-3.31: Hierarchical weight resolution with PBF
- TE-4.1: Base Leader Election
- TE-4.2: Persistence Mode
- TE-5.1: gRIBI Get RPC
- TE-6.1: Route Removal via Flush
- TE-6.2: Route Removal In Non Default VRF
- TE-8.1: DUT Daemon Failure
- TE-8.2: Supervisor Failure
- TE-9.1: FIB FAILURE DUE TO HARDWARE RESOURCE EXHAUST
- TE-9.2: MPLS based forwarding Static LSP
- TE-9: gRIBI MPLS Compliance
- TE-10: gRIBI MPLS Forwarding
- TE-11.1: Backup NHG: Single NH
- TE-11.2: Backup NHG: Multiple NH
- TE-11.3: Backup NHG: Actions
- TE-11.21: Backup NHG: Multiple NH with PBF
- TE-11.31: Backup NHG: Actions with PBF
- TE-13.1: gRIBI route ADD during Failover
- TE-13.2: gRIBI route DELETE during Failover
- TE-14.1: gRIBI Scaling
- TE-14.2: encap and decap scale
- TE-15.1: gRIBI Compliance
- TE-16.1: basic encapsulation tests
- TE-16.2: encapsulation FRR scenarios
- TE-16.3: encapsulation FRR scenarios
- TE-17.1: VRF selection policy driven TE
- TR-6.1: Remote Syslog feature config
- TRANSCEIVER-1: Telemetry: 400ZR Chromatic Dispersion(CD) telemetry values streaming
- TRANSCEIVER-3: Telemetry: 400ZR Optics firmware version streaming
- TRANSCEIVER-4: Telemetry: 400ZR RX input and TX output power telemetry values streaming.
- TRANSCEIVER-5: Configuration: 400ZR channel frequency, output TX launch power and operational mode setting.
- TRANSCEIVER-6: Telemetry: 400ZR Optics performance metrics (pm) streaming.
- TRANSCEIVER-7: Telemetry: 400ZR Optics inventory info streaming
- TRANSCEIVER-8: Telemetry: 400ZR Optics module temperature streaming.
- TRANSCEIVER-9: Telemetry: 400ZR TX laser bias current telemetry values streaming.
- TRANSCEIVER-10: Telemetry: 400ZR Optics FEC(Forward Error Correction) Uncorrectable Frames Streaming.
- TRANSCEIVER-11: Telemetry: 400ZR Optics logical channels provisioning and related telemetry.
- TRANSCEIVER-12: Telemetry: 400ZR Transceiver Supply Voltage streaming.
- TRANSCEIVER-13: Configuration: 400ZR Transceiver Low Power Mode Setting.
- TUN-1.4: Interface based IPv6 GRE Encapsulation
- TUN-1.9: GRE inner packet DSCP
- Test Plans