Bruno

I am also interested in this. For instance I have a server that accepts streaming rpcs, but, a single user could spawn multiple connections and kill one of my tasks...

Thanks for replying! With atomic: ```cpp #include #include #include #include std::atomic counter = 0; void GetConnection(boost::mysql::connection_pool& pool, int index) { pool.async_get_connection(boost::asio::cancel_after( std::chrono::seconds(6), [index](boost::mysql::error_code error, boost::mysql::pooled_connection /*connection*/) { if (error) {...

Thank you so much! It worked, I can't reproduce a deadlock now. What is odd is that I compiled it with -fsanitize=thread and it didn't accuse any warnings. I think...

To add to the above, when using the connection with async_execute, I was able to reproduce the deadlock again: ```cpp #include #include #include #include #include std::atomic counter = 0; void...

Dang, thanks for pointing that out! Still wrapping my head around asio and completion tokens :D Would this also be an issue if I were using coroutines? This is the...

Thanks Marcelo! It makes sense now. Do you think the API docs should state that or maybe the API should be modified to make it clearer? It isn't immediately obvious...

+1

Maybe related: https://github.com/RedisInsight/RedisInsight/issues/3022

If you are using a local redis running on a container: ``` docker run --restart=unless-stopped -p 6379:6379 -p 8001:8001 --name=server_local_redis \ -v redis_data:/data \ -e REDIS_ARGS="--appendonly yes --save 60 1...

> Hi [@zlausd](https://github.com/zlausd), this [PR](https://github.com/boostorg/redis/pull/217) has some improvements that I think will fix this problem. I couldn't test it in the container because it uses vcpkg to pull boost so...