Warning: If this step is completed when NCNs are offline or shutdown, the information compared here will not match the actual connections. Therefore, this step should be re-run again once the whole system is up.
To validate the cabling you can run command similar to below:
canu validate shcd-cabling --ips 10.252.0.2,10.252.0.3 --tabs 40G_10G --corners J12,T36 --shcd ./SHCD.xlsx
NOTE
Modify the command to use the correct SHCD file and correct--tabs
,--corners
, and IP addresses.
The output should look as follows:
sw-spine-001
Rack: x3000 Elevation: u24
--------------------------------------------------------------
Port | SHCD | Cabling
--------------------------------------------------------------
1 ncn-w001:pcie-slot1:1 ncn-w001:pcie-slot1:2
2 ncn-w002:pcie-slot1:1 ncn-w002:pcie-slot1:1
3 ncn-w003:pcie-slot1:1 ncn-w003:pcie-slot1:1
sw-spine-002
Rack: x3000 Elevation: u24
--------------------------------------------------------------
Port | SHCD | Cabling
--------------------------------------------------------------
1 ncn-w001:pcie-slot1:2 ncn-w001:pcie-slot1:2
2 ncn-w002:pcie-slot1:2 ncn-w002:pcie-slot1:2
3 ncn-w003:pcie-slot1:2 ncn-w003:pcie-slot1:2
In the returned output, look for differences between the SHCD and actual network configuration.
In the above example, incorrect cabling was detected on the sw-spine-001
switch on port 1:
1 ncn-w001:pcie-slot1:1 ncn-w001:pcie-slot1:2*
The SHCD has the correct information on port 1 but the actual switch configuration is mismatched.
To fix this issue, re-cable ncn-w001
so that it is correctly connected to sw-spine-001
.
Return to deploying the management nodes.