Oran Agra

the file seems corrupted. the crash in redis-rdb-tools: ``` Traceback (most recent call last): File "rdbtools/cli/rdb.py", line 107, in main() File "rdbtools/cli/rdb.py", line 101, in main parser.parse(dump_file) File "/home/oran/work/redis-rdb-tools/rdbtools/parser.py", line...

sripathikrishnan isn't active here anymore. i'm currently maintaining this repo, but i don't have much time to contribute either. looks like a good idea, it can be a subclass of...

Hi, your question is very unclear, there's no snappy compression in redis, and i'm not sure what you want to achieve. i can only assume that you mean that the...

ohh. maybe implement a new callback class, that derives from the json callback class. then on each callback, do your decompression, and forward the decompressed data to the json class...

In a nut shell, in case the write traffic during FLUSHALL ASYNC reaches OOM condition, we'll get the CLIENT PAUSE WRITE behavior (postponing / throttling) instead of an OOM error...

Redis will behave correctly, but it's not useful for users. Users want a FLUSHALL command that doesn't block the traffic for a long duration, ideally it should release all the...

@madolson AFAIK the reason replication buffers aren't included in eviction limits is to avoid a feedback loop. that's already a bad fact (same as lazy free eviction), and doing the...

Looking at an older redis, before CLIENT PAUSE WRITE was added, even then, during a client pause, we also suspended active expire, etc, and it seems it was very much...

I think the first option is preferred since it is inline with other commands like ZRANGE. speaking of ZRANGE, isn't it plain and simple to use `ZRANGE WITHSCORES` to get...

we are not likely to ever add a ZMRANK that works on multiple keys. but letting ZRANK work on multiple members (like HSET does today), is a good point, and...