Stack of 3 CBS350-24XS

[fficcadenti]

Hello everyone,

I have 3 CBS350-24XS switches in stack, first upgraded to 3.3.0.16 and a few days ago to 3.4.0.17.

In addition to an unwanted restart due to a critical error, the main problem is switches going into boot loop, apparently due to a timeout before joining the stack.

After rebooting 4 or 5 times, the stack goes up and everything works normally.

I have attached the error logs.

Furthermore, log time seems to be out of sync before the OS loads.

The switches are stacked in ring topology using SFP-H10GB-CU1M cables.

Any ideas? Thanks

[balaji.bandi]

I may be wrong Looking at the Logs, some device showing 3.0.3.16 and some 3.0.4.17

Personally i would take offline the switches and see all version are same and join them back in to Stack ?

what was the version before and working as stable ?

check troubleshooting guide :

https://www.cisco.com/c/en/us/support/docs/smb/switches/Cisco-Business-Switching/kmgmt-3006-troubleshoot-adding-to-existing-stack-cbs.html

[fficcadenti]

Thanks for your reply.

The 3.3.0.16 version appears in the logging.txt file because there were problems, not related to stacking, even with that version. With the 3.3.0.16 version, the switches worked correctly until, after several hours of being turned on, many "I2C failed to read...component CPLD" logs started to appear and managing the switches became very difficult due to slowness and any command given caused everything to freeze, resulting in a stack reboot. It is also worth noting that, if the management was not connected (ssh,web ui,..) everything worked correctly, as if the data plane was not affected in any way.

The update to 3.4.0.17 was performed precisely in the hope of resolving these problems, but instead, the situation worsened. The switches are in production, this evening i'll verify the firmware of each of them and if the logs problem occurs again with latest firmware.

More information will follow.

[fficcadenti]

Hello again,

i've not yet been able to verify the firmware version of each individual switch, but as I already wrote in the previous post, the I2C errors persist. I have attached some logs.

It is worth noting that these errors only occur after several hours of being turned on and do not compromise the functionality of the network, they only create slowdowns at the management level.

Considering the errors on the second unit id, I am considering removing it from the stack to see if leaving only the first and third in stack the errors persist.

Any suggestion?

[balaji.bandi]

i was thinking some issue with uplink ports - If you have SFP try using in uplinks. - or as you mentioned remove 3rd switch and see 2 switch ok running (and check 3rd switch as standalone get same errors too ?)

May be worth opening TAC case as you are already running Latest firmware :

https://www.cisco.com/c/en/us/support/web/tsd-cisco-small-business-support-center-contacts.html