Artem Danilov

https://github.com/tikv/client-go/pull/1525 is landed. Once https://github.com/pingcap/tidb/pull/58440 landed, we would need to bump it one more time for https://github.com/tikv/pd/pull/8936 then I can work on sysvar.

Thanks [cfzjywxk](https://github.com/cfzjywxk) for the comment. I think it is not always possible for PD to reply to provide this information to TiDB in case it is completely overloaded. https://github.com/tikv/pd/issues/8678 proposes...

> The root cause: > The connection context has been cancelled, but the stream context is different from the connection context, so the stream will not cancel the pending request....

This will work only for table which fit a single region along with the index, right?

> Maybe we can change it to [0, 1]? /cc @Tema @niubell @okJiang @rleungx I don't mind the pct/ratio, but I wish you changed the name as well. The `_pct`...

@LykxSassinator how is marking store as busy in PD helps to avoid sending traffic too early? I don't see TiDB region cache or Unified Health Controller interpreting store#is_busy flag set...

> PD will avoid balancing leaders to a TiKV Thanks @LykxSassinator. I see. This make sense for leader reads, but what about follower reads (including stale reads)? Can this flag...

@lhy1024 @bufferflies could you please approve 4 pending workflows to run ^