this could be due to a number of factors, but as a watch client, you should always reconnect and resume from where you previously left off if you get a canceled error

Is there a way to figure out the reason? It was consistently happening every ~1.5mins

if it is constant, it might be the gRPC connection timeout

is that configurable? i tried to look through the params in the watch call but the only thing I could find that looked remotely related was a

deadline

param. I'll look through the docs as well

it would be a gRPC option

ah cool. i'll go through those