Lugia
Lugia
@kyessenov I think I found the reason. Before the wasm encounters an "Uncaught RangeError", there are many "too_many_premature_resets" errors. I tried searching the code and saw that it would close...
Is there any progress? @kyessenov @PiotrSikora
> @LugiaChang is your Wasm plugin written in Javascript by any chance? no, by rust and tinygo
> Hm, in that case, I think v8 is throwing the error; in my experience, RangeError comes from Javascript world. I think what is happening is that your stress test...
> IMO, I wouldn't worry about this too much from a production perspective. Unless someone is trying to DoS you, you probably won't trigger the conditions for this. We should...
> Well the issue is that your stress tester is sending too many resets too quickly; regular service traffic shouldn't be resetting that soon When I create multiple grpc conns,...
> reuse如果为false的话，看起来是创建了一次性conn，用完之后会close掉。如果reuse为true，那么会从已创建的conn里选出一个，即使它已经达到了MaxCurrentStream
> I used an `Arc` @LugiaChang Is Wasm single thread? I'm not quite sure.
> the panic may be It happens elsewhere It may be wasm runtime error. see https://github.com/istio/istio/issues/49972