Broadcom BES-53248 Cluster switch with cluster 4 nodes (ISL count is incorrect)

[Quake]

Hello,

I added  two AFF250 nodes to a cluster fas2750  existing already switched. all nodes are in ONTAP 9.9.1P1. 
The new cluster 4 nodes is running fine. All cluster interconnect ports are up. the cluster ring show no errors.

All tests of "Config advisor" about the cluster are passed with succesfully. however "Config advisor" return a warning (medium impact) about the switch configuration (see bellow).

My infrastructure :

2 x  BES-53248 cluster switch 

!System Description "BES-53248A1, 3.4.4.6, Linux 4.4.211-28a6fe76, 2016.05.00.05"
!System Software Version "3.4.4.6"
Filename : BES-53248_RCF_v1.6-Cluster-HA.txt
Date : Oct-20-2020
Version : v1.6
Port Usage:
Ports 01 - 04: 10GbE Cluster Node Ports
Ports 55 - 56: 100GbE Cluster ISL Ports

According to the hwu netapp  I can not apply another RFC than 1.6 with this firmware version (3.4.4.6) could it come that the ONTAP version is 9.9.1RC1 and I am in 9.9.1P1 I do not think but I do not see why I have this message? do you perhaps have an answer to give me ?

BES-53248_RCF_v1.6-Cluster-HA.txt /  Cluster InterConnect Switch EFOS 3.4.4.6  / 9.9.1RC1, 9.8RC1, 9.7RC1, 9.6RC1

Thanks you  

Vincent.

[pedro_rocha]

Hi,

According to the config advisor print, it is telling you about a "Management Switch". It is not the cluster switch (BES).

So, I guess this is a false positive. I've got a couple of false positive messages from config advisor in the past. I recommend that you open a ticket with NetApp config advisor team reporting this. You can do that from the config advisor GUI on the upper right corned if I am not wrong.

Regards,

Pedro

[andris · Accepted solution]

Do you have CN1601 management switches, by any chance?
In any case, I would check what the switch health monitor is actively monitoring with the "system cluster-switch show" command.  If you are monitoring management switches and you see unneeded alerts for them with "system health-alert show", it might be simpler/easier to just stop monitoring the management switches.

This KB has information on the commands you can use to delete a switch you no longer wish to monitor:

https://kb.netapp.com/Advice_and_Troubleshooting/Data_Storage_Systems/Fabric%2C_Interconnect_and_Management_Switches/AutoSupport_Message%3A_Call_home_...