Catalyst 3850 (3850-NM-4-10G) Module issue

[mmitchell2519]

I'm having an issue with my 3850-NM-4-10G. The module shuts down completely when hot swapping sfp's. Doesn't matter if they are 1G or 10G. The only way to restore connectivity is to reboot the switch. I'm running Denali 16.03.01, WS-3850-48 UPOE. I'm also getting this error message. Please see the output below. 

%OSPF-5-ADJCHG: Process 1, Nbr 0.0.0.101 on TenGigabitEthernet1/1/1 from LOADING to FULL, Loading Done

Has anyone had an issue like this? 

[Reza Sharifi]

%OSPF-5-ADJCHG: Process 1, Nbr 0.0.0.101 on TenGigabitEthernet1/1/1 from LOADING to FULL, Loading Done

The above message is when your OSPF peering comes back up.

You may want to load a different version IOS as this may be a bug. Also, probably a good idea to open a ticket with TAC and have them analyze the logs and the behavior.

HTH

[balaji.bandi]

Can you post the configuration of that port?  ( we have seen this issue another version before we move to cat 9K)

do you have any speed nonegotiation command in the port config?

[mmitchell2519]

Hello,

Here is the config. I have one port configured on the module. Interface tengigabitethernet 1/1/1. 

interface TenGigabitEthernet1/1/1
no mvrp timer periodic
no mvrp
no switchport
ip address  x.x.x.x 255.255.255.252
ip ospf authentication message-digest
ip ospf message-digest-key 1 md5 
ip ospf network point-to-point
ip ospf resync-timeout 40
ip ospf dead-interval 40
ip ospf mtu-ignore
ip ospf 1 area 0.0.0.0
ip ospf cost 64025
ipv6 enable
ospfv3 network point-to-point
ospfv3 dead-interval 40
ospfv3 cost 64025
ospfv3 bfd
ospfv3 1 ipv6 area 0.0.0.0
hold-queue 40 out
end

[Leo Laohoo]

---

@mmitchell2519 wrote:

I'm running Denali 16.03.01

---

Upgrade the firmware of the stack (to something more recent) and this bug should be fixed.