Doug Fawley

@lev-lb you said: > i'll try to see when i can carve some time out for that. In response to a request for a reproduction. If we can't reproduce bugs,...

We should measure the flakiness w/10k runs and assign priority accordingly. Also consider new-ness -- if the test is new we should treat with much higher priority since we don't...

100/100k = 1/1000 which is pretty low, but taking a look to see if we can find the root cause would be worth it.

Flaked 2x on my PR in a single snapshot: https://github.com/grpc/grpc-go/actions/runs/11634120790/job/32400709681?pr=7798 https://github.com/grpc/grpc-go/actions/runs/11634120790/job/32400708629?pr=7798

gRPC does exponential backoff internally automatically. Is there some reason why the default settings aren't working for you?

> I have use some cases in which I need to change the MaxDelay default value. Can you explain why? > I would like only to state again that at...

> In order to setup an optimized configuration, shouldn't we tune the backoff parameters according to a connection deadline? What is wrong with using the default values?

What contexts are you referring to there? We don't generally talk about contexts at the same time as subchannel connection timeouts.

I don't need to review - the grpc-security team's review (whoever does it) should be sufficient. cc @gtcooke94

There are no plans to stabilize the functionality. gRPC's HTTP/2 server provides more features than the Go HTTP server and we haven't yet heard any critical use cases that require...