A large number of nodes on the lhg73 subnet are offline

I don’t see any incident response chatter. My best guess is that it’s a dashboard error. I pinged the dashboard folks to look into it

The subnet was upgraded beginning at 9:46 UTC. You should find an upgrade proposal on the NNS that would have been approved just before that.

For whatever reason, it took 5 minutes before the subnet had enough running replicas to continue replicated execution. You must have been looking at it during those 5 minutes.

1 Like

https://dashboard.internetcomputer.org/proposal/133855

Nodes will be temporarily offline when a subnet is being upgraded to the new replica release. As @free states, the roll out schedule is reflected in the NNS proposals (one per subnet). lhg73 is the subnet with the largest state size so that is probably the reason it took 5 minutes for enough replica nodes to become online again.