Vladislav

I wanted to simplify journaling so that stub transactions report directly to the main transaction, this way we don't need both `ReportWritesSquashedMulti` and `FIX_ConcludeJournalExec` at all, but with a single...

Will do. This PR will become much easier because we don't need to use ScheduleSingleHop as it's no longer special

A bit rough around the edges but good enough to be looked on 🔬

We have only one alternative option - remove the replica if it lags behind too much

`BufferedStreamerBase` from `io_utils` has clearly defined limits on what gap we can keep before stopping

Actually it might be that we need just to switch those two lines 🤔 https://github.com/dragonflydb/dragonfly/blob/3dd6c4959c3056d6f007cc4001e0923302a566a2/src/server/replica.cc#L144-L145 With this order, the loop continues working and blocks because ENABLE is still set With...

we already had an issue for this around a year ago

https://github.com/hibiken/asynq/blob/cc777ebdaa62b69bd6e985fa97117b854e7d1cd6/internal/rdb/rdb.go#L219 ```go // Input: // KEYS[1] -> asynq:{}:pending // KEYS[2] -> asynq:{}:paused // KEYS[3] -> asynq:{}:active // KEYS[4] -> asynq:{}:lease // -- // ARGV[1] -> initial lease expiration Unix time...

1. We provide global and non atomic mode 2. It will help with latency because we don't need to dispatch, but running on a single shard doesn't mean we can...

True.... Slot locks nice feature for DF in general