02-14-2022 05:53 AM
Hi
I wanted to share an issue I've had with OSPF and Catalyst 9300 switches. I would like to know if anyone else has seen this or not?
We recently purchased 2 * Catalyst 9300-48UXM switches. Switch1 has C9300-NM-8X card installed and network advantage. Switch2 has network essentials. Both switches are running v.17.03.04.
Switch2 is connect to switch1 with a single uplink configured as a trunk port. Switch1 is configued with OSPF routing and has an uplinks to 2 ASR routers - Interface Te1/0/48 goes to RTR1 and Te1/0/47 goes to RTR2. It is expected that the default equal-cost routing is suffice for this setup, as the 2 swtiches have replaced 2 older 3560 switches, with the same layout and configuration.
When Switch1 had both router uplinks connected, we started seeing some intermittent packet loss and discards on the interface stats to RTR2. If the uplink to RTR2 was shut down, the issue went away. When the RTR2 uplink came back up we saw the issue and if the RTR1 uplink was shut down, the issue remained. We replaced cables and sfp modules in the router with no effect. We swapped the 2 switches around but we still saw the issue. CDP was established with both switches and routers seeing each other. Also, OSPF adjacency was established over both links
Eventually, I moved the uplink to RTR 2 from Te1/0/47 to Te1/0/46 and the issue disappeared. All traffic is flowing correctly and there is no packet loss and we are happy with this solution. However, it would be good to know if there is something underlying on the Catalyst 9000 series that we have missed. So, does anyone know if there is some reason why port 47 on both switches would cause packet loss? Has anyone seen this sort of issue before or have we just been unlucky with 2 bad interfaces on 2 switches?
For reference the switch configs are:
RTR1 Interface Configuration
------------------------------------------------
description LINK to Switch1
ip address X.X.X.X 255.255.255.252
ip ospf message-digest-key 1 md5 7 XXXXXXXXXXXXXXXXXXXXXX
speed 1000
no negotiation auto
cdp enable
RTR2 Interface Configuration
------------------------------------------------
description LINK to Swtich1
ip address X.X.X.X 255.255.255.252
ip ospf message-digest-key 1 md5 7 XXXXXXXXXXXXXXXXXXXXXX
speed 1000
no negotiation auto
cdp enable
end
Switch1 Configuration
------------------------------------------------
interface TenGigabitEthernet1/0/47
description LINK to RTR2
no switchport
ip address X.X.X.X 255.255.255.252
ip ospf message-digest-key 1 md5 7 XXXXXXXXXXXXXXXXXXXXXX
speed 1000
duplex full
interface TenGigabitEthernet1/0/48
description LINK to RTR1
no switchport
ip address X.X.X.X 255.255.255.252
ip ospf message-digest-key 1 md5 7 XXXXXXXXXXXXXXXXXXXXXX
speed 1000
duplex full
Many Thanks
Roy
02-14-2022 06:52 AM
Hi Roy,
The config looks correct and pretty simple. Since this is happening with the same port (1/0/47) on 2 different switches, it could be a software bug. You may want to open a ticket with Cisco and see if they have any info regarding this issue.
HTH
02-14-2022 06:57 AM
Hi,
you better open a TAC and get the troubleshoot from cisco team.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide