Gleb Podkolzin

We're experiencing something similar in our code; the general way to reproduce appears to be this: * Find a slow request you can reliably trigger. Typically it should be slow...

1) Our backend is using `grpc-java` with `netty` which as I understand has no specific "max number of streams" setting currently implemented (which was my original theory - requests get...

I also took some time off, but now am dedicating more time to this. So far I have had no luck reproducing outside of our stack even after changing the...

Will look into getting a TCP dump - need to do some more research to figure that out. However, this might provide another clue - turns out this condition is...

> Is your server application code purposely shutting down the server when handling the bad request? If not, I don't know why it would send a GOAWAY like that. No,...

> If you could create and share a completely isolated reproduction of this error, that would be really helpful. Still working on this; I have a suspicion the GOAWAY in...

Success - I can reproduce our specific condition outside of our backend. Probably final update for this thread as it makes sense to me to open a separate issue to...