Catalyst 2960 with LACP link down when switching from Member to Master

[iTbaJz]

Hello.

English is not my native language, so please bear with my mistakes.

We have a redundant system with two Catalyst 2960s stacking and two load balancers.

Last week, because L2SW(#1) on the Master side lost power, L2SW(#2) transitioned from Member to Master.

At that time, after the link down of L2SW(#1) was detected on the load balancer, the link down of L2SW(#2) was also detected for about 5 seconds.

In Catalyst 2960 stacking, when the status transitions from Member to Master, does the link down occur on the device?

Thanks for the help.

[Leo Laohoo]

---

@iTbaJz wrote:
when the status transitions from Member to Master, does the link down occur on the device?

---

No, the link to the secondary remains up.

[iTbaJz]

@Leo Laohoo 

Thanks for your reply.

Sorry, I have one more thing to ask.

We checked the logs of the L2SW(#2), only the port on which LACP was configured was linked down when transitioned from Member to Master.
Is there any possibility that LACP may have affected it?

Thanks.

[Leo Laohoo]

It should not. 

Depends on the load balancer and how it switches traffic from one NIC to another.  One method is to bounce that said NIC and all traffic goes automatically to the secondary.

[iTbaJz]

I'm sorry for replying late and thanks for answering.
But even now I am not sure what’s causing this problem.

I'll send system configuration diagram (Fig1.png) and L2SW(#2)'s log file (L2SW_log.txt). Could you think of anything that might have caused it?

And L2SW(#2) logs were outputting "administratively down" before the link went down.
L2SW(#2) appears to be intentionally down. I would like to know the cause of the "administratively down".

Thanks.