Upgrade to Vault 1.17 causing follows to rpc error trying to talk to the leader

Vault
1

After Updating Vault from 1.16.3 to 1.17.1 I found my cluster basically fell over and the passive/follower nodes output lots of errors of the form:

error during forwarded RPC request: error="rpc error: code = Unavailable desc = connection error: desc = \"transport: Error while dialing: dial tcp <active instance ip>:8201: connect: connection refused\""

Is anyone aware of a change between 1.16 and 1.17 that might cause this behaviour?

2

I worked this out. I’d accidentally set the ‘addr’ config to http rather than https so https client connections being forwarded were erroring due to the stepdown from https on the original connection to http on the forward.