You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We use ProgressNotify in most of our etcd watchers and notice that the etcd servers have a spike of etcd_network_client_grpc_sent_bytes_total every exact 10 minutes. We think it's because we created a lot of watchers at the same time so every 10 minutes the servers need to send out progress update for every watcher. Is it true? If so, is it possible to add some jittering for the progress notify on the server side to avoid this kind of network spike?
I vaguely recall that the progress notification is supposed to be approximately every 10 minutes in the past, was it changed to exact every 10 minutes recently?
The text was updated successfully, but these errors were encountered:
We use ProgressNotify in most of our etcd watchers and notice that the etcd servers have a spike of etcd_network_client_grpc_sent_bytes_total every exact 10 minutes. We think it's because we created a lot of watchers at the same time so every 10 minutes the servers need to send out progress update for every watcher. Is it true? If so, is it possible to add some jittering for the progress notify on the server side to avoid this kind of network spike?
I vaguely recall that the progress notification is supposed to be approximately every 10 minutes in the past, was it changed to exact every 10 minutes recently?
The text was updated successfully, but these errors were encountered: