11:02:06 GMT <provenzano> Hi! Is there anything similar to sentinels in Redis Cluster? Or the cluster integrity is only managed by the cluster itself?
11:07:43 GMT <badboy_> provenzano: Cluster is completely self-managed
11:17:08 GMT <provenzano> First I was planning to run 2 redis servers with 3+ sentinels in different servers to achieve quorum.
11:17:08 GMT <provenzano> With redis cluster I MUST have 3 redis servers in order to achieve cluster integrity, right?
11:36:35 GMT <badboy_> 3+
12:26:55 GMT <provenzano> Thanks! :)
15:03:02 GMT <RedisUser> Hello! Have a question about ZRANGEBYLEX command. According original documentation, there is following information: if store keys in ordered set with zero score, later keys can be retrieved with lexicographical order. And operation complexity will be O(log(N)+M), where N is total elements count and M is result set size.
15:03:06 GMT <RedisUser> But after some experiments and reading source code, it's probably what ZRANGEBYLEX operation has a linear time search, when every element in `ziplist` will be matched against request. If so, complexity will be more larger than described above - about O(N), because every element in `ziplist` will be scanned. Where is truth? :)
15:04:26 GMT <RedisUser> Thanks in advance
15:22:01 GMT <RedisUser> After debugging with gdb, it's clean that ZRANGEBYLEX command is implemented in [genericZrangebylexCommand](https://github.com/antirez/redis/blob/unstable/src/t_zset.c#L2827) function. Control flow continues at `eptr = zzlFirstInLexRange(zl,&range);`, so major work for element retrieveng will be performed at [zzlFirstInLexRange](https://github.com/antirez/redis/blob/unstable/src/t_zset.c#L895-L916) function.
15:22:05 GMT <RedisUser> All namings and following control flow consider that `ziplist` structure is used, and all comparation with input operands are done sequentially element by element.
15:22:08 GMT <RedisUser> Inspecting memory with analysis after inserting well-known keys in redis store, it seems that ZSET elements are really stored in `ziplist` - byte-per-byte comparation with [gauge](https://github.com/antirez/redis/blob/unstable/src/ziplist.c#L118) confirm it.
15:22:15 GMT <RedisUser> So question - how can documentation be wrong and propagate logarithmic complexity where linear one appears? Or maybe ZRANGEBYLEX command works slightly different? Thanks in advance.
17:32:53 GMT <Diplomat> Hey guys, is it possible to increment a counter so it resets it's TTL ?
17:33:38 GMT <Diplomat> Maybe send expire command right after incr ?
19:15:45 GMT <codeyman> It seems that redis only supports upgrade. Are there any scripts that can help with a downgrade? (i.e. convert a newer rdb file to an older one)
19:17:06 GMT <zpojqwfejwfhiunz> codeyman: no
19:18:23 GMT <zpojqwfejwfhiunz> you can try writing a script that goes through all the keys you are interested in and does some ETL to transload them to an older version of the server which can then be saved to disk
19:23:24 GMT <zpojqwfejwfhiunz> although I doubt that solves your underlying issue
22:26:44 GMT <Lewd_> Hi, anyone available to help with an issue I'm having when attempting to restore from a backup on a new vps?
23:09:29 GMT <rduke2> if I wanted a very fast store for stateful data (high read/write throughput) but I can't guarantee my dataset will fit in memory is redis still suitable?  is there any way to just throttle the memory to LRU say?
23:19:25 GMT <minus> rduke2: it has to fit into memory, but there are eviction mechanisms like lru
23:32:36 GMT <rduke2> minus: but what if I have a 2GB dataset but only 1GB of RAM?  Can't I just keep the "hot" data in RAM?