Richard Cooper

Craaaap i don't know how to fix it... Maybe we need to add additional mutex to tcp connection?

@aofei thanks a lot, i'll try to experiment with this at this weekends (come back to this thread in near future, i promise!) @Lonami if you have a little free...

@aofei: I tried to copy messages handling from telethon (thanks Lonami for hint) but no progress at al... Someone whispered in my ear a small inside that you're not required...

@aofei oooow i got it! problem is that multiple requests getting seqno in parallel but it's illegal for telegram service. F.e.: ``` req1: get me seqno mtproto: locks req2: get...

@aofei i'll refactor full method, and commit today or tommorow, thanks for help! @Lonami you too 😎

@innovative-sol @TibebeJS @aofei looks like we fixed it. Try it out with the [latest commit](https://github.com/xelaj/mtproto/commit/df86cba1ad26d714e1569b5be9486f4f32be14b4) and give us a little feedback, please. Hope that for you it works as well...

@aofei 1) [`MTProto.sendPacket`](https://github.com/xelaj/mtproto/blob/df86cba1ad26d714e1569b5be9486f4f32be14b4/network.go#L21) not thread safe by design (each request can be prepared asynchronously and then sent synchronously). So, all stuff _behind_ `m.seqNoMutext` is actual preparation stuff. Btw, glad that...

@hookzof do you have any patch which i can merge? @MJDrgn is right about unixnano, it is rounding, so maybe it's better to add to nanoseconds random value from 0...

Hey @blez! Thanks for asking and ping for inactivity! I researched this issue, Here is whats happen: So, TL schema has preeeety few types, including only bytestream and a few...

@blez i'll tried to fix this bug, tests passed out, but in real life result ma be different, so i'll not close this issue yet. If you have a few...