When you say 'outage' do you mean the FI uplinks didn't come back up for 10 minutes?
Having unbalanced throughput capabilities to your upstream switches could cause a bottleneck during failback scenario, especially if you have hyperflex environment.
Also, incorrect spanning tree settings on your upstream switches can also cause traffic to get blackholed when an FI has just come back online.
This will probably need a TAC case to sort through logs and time line, and to understand what the outage actually entails.
Thanks,
Kirk...