sewenew

> can I know why so many fd since no traffic? Once you create an AsyncRedisCluster object, it tries to connect to a random cluster node, and fetch the slot-node...

> Is there misusage in my ping? Have you waited the returned future before destroying the AsyncRedisCluster object? You need to ensure that all commands have been sent and the...

Can you show me the minimum code that can reproduce the problem? If you use MGET command, these keys should locate on the same slot, even if the cluster scales...

So sorry, but somehow, I missed this message. redis-plus-plus lazily creates connections, if it's the first time that these requests are sent to nodes, it needs to create connections to...

I'm not an expert on CMAKE, but you can take [this issue](https://github.com/sewenew/redis-plus-plus/issues/214) for reference. With the solution mentioned in the issue, you should solve the problem. Regards

Thanks for your reporting! > I finally think the memory accumulation may caused by reconnect logic when redis++ actively discovers redis connection issues when sending commands Did you have a...

`subscriber()` method randomly picks a node, and creates a connection to it. It does not keep connections to every master. This is a different behavior from RedisCluster. Regards

It depends. If you create RedisCluster with MASTER role, then `RedisCluster::subscriber` creates connection to a random master. If you create it with SLAVE role, then `RedisCluster::subscriber` creates connection to a...

So far, NO, it won't reconnect. As the doc mentioned, once you get an exception, you need to create a new Subscriber.

You used a very old version of redis-plus-plus. Please update to the latest one, and try again. This problem should have been fixed.