[f-nsp] ICX 6610 not learning mac addresses any more

[Frank Bulk]

The first thought that comes to mind is that there is a MAC address table hash issue.

 

This vendor does a good job of explaining it: https://www.calix.com/content/dam/calix/mycalix-misc/lib/iae/exa/m/eanepg/index.htm?toc10152651.htm?101585.htm

 

Frank 

 

From: foundry-nsp <foundry-nsp-bounces at puck.nether.net> On Behalf Of Tudor Capatina
Sent: Thursday, September 14, 2023 8:04 AM
To: foundry-nsp at puck.nether.net
Subject: [f-nsp] ICX 6610 not learning mac addresses any more

 

Hey there

 

got a weird issue on the above mentioned switches and wondering if anyone has seen something similar. 

A server with mac-add MAC1 is connected to an ICX 6610 SW1. The server is generating traffic. SW1 has a mac-add table of around 10k:

Total active entries from all ports = 10368

And it does not learn MAC1. However SW1 is connected to another ICX SW2 with a smaller mac-address table (5k) which learns MAC1 from the link to SW1.

The problem is that SW1 will treat all traffic to MAC1 as unknown unicast and flood it on all it's ports. The issue is not with a specific mac-address but with several.

Mac learning timers are default all over the place. Mac learning is not disabled anywhere. 

 

I was suspecting some cam profile but cannot see such an option on these devices. 

In theory there should be space to allow learning of more mac-addresses:

 

show default values
sys log buffers:50         mac age time:300 sec       telnet sessions:5        

System Parameters    Default    Maximum    Current   
- - - -        
mac                  32768      32768      32768     

 

Any ideas are welcome 

 

Thanks

Tudor

 

 

 

 

 



Links contained in this email have been replaced. If you click on a link in the email above, the link will be analyzed for known threats. If a known threat is found, you will not be able to proceed to the destination. If suspicious content is detected, you will see a warning.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/foundry-nsp/attachments/20230915/8d237625/attachment-0001.htm>