debing.sun

from your ouput we can see than `redis-cli` just consume a littlt memory. it doesn't get stuck, but rather that it can't receive a reply(i don't know why it doesn't...

> As said: > > * removing the forwarding port in VSCode fixes the problem, it dumps the GBs in my main memory > * quiting redis-cli also dumps the...

@zygisa do you still have the same problem? is there reproducible steps? i tried serveral time locally but can't reproduce it.

we need to update db->hexpire when executing writable hash command or other commands that may affect hash. 1. flushdb, del, move, .etc 2. defrag 3. hset, hdel, and other may...

I'm wondering why we introduced list ebucket which is only one per db. It doesn't save a lot of memory overhead, but introduced move code complexity.

> * I think that I took care to handle flush a/sync, del, move and cover with tests. Pls, lmk if something is missing. > * defrag is something that...

yes, you can copy this config file, but you need to remove all the dynamically generated configs, like `sentinel myid`, `known-sentinel`, .etc.

### Benchmark ``` taskset -c 10 ./src/redis-server ./redis.conf --save "" ``` ``` taskset -c 16-20 memtier_benchmark --pipeline 30 --test-time 60 -c 1000 -t 2 -x 1 --hide-histogram ``` unstable: ```...

Update the benchmark 1) a large number of arguments ``` taskset -c 16-20 memtier_benchmark --hide-histogram --test-time 180 --pipeline=10 --command="HSET h a 1 b 2 c 3 d 4 e 5...

please note that I no longer automatically release argv in cron because it may still be used elsewhere, such as by blocked clients. Since I have avoided excessive waste when...