0

The following is what is used to benchmark "write to all members".

# write to all members
benchmark --endpoints=${HOST_1},${HOST_2},${HOST_3} --conns=100 --clients=1000
put --key-size=8 --sequential-keys --total=100000 --val-size=256

My understanding is that only the leader listens to write requests. So how is it possible to write to all members of an etcd cluster?

foo
  • 166
  • 9

1 Answers1

0

The secondary nodes just forward the write requests they receive to the leader.

foo
  • 166
  • 9