06:18:54 GMT <vishalpathikgupt> help
07:36:51 GMT <bjpenn> can redis hold a dataset larger than its memory
07:37:05 GMT <bjpenn> "its memory" being the server's memory
07:41:51 GMT <Inge-> they stopped supporting that - and it really does break redis' performance if stuff is swapped to disk
07:53:41 GMT <vishalhe> For those working on Redis modules, you may find this hackathon interesting : http://hck.re/5haXL1
08:07:41 GMT <markuman> where to register?
08:08:39 GMT <markuman> ah on top of the page
10:56:52 GMT <mished> hello guys
10:57:06 GMT <mished> I have a question regarding eval with redis-cli
10:57:32 GMT <mished> like... how can I pass an array as ARGV[1] via cli?
10:57:45 GMT <mished> redis-cli --eval script.lua u1 u2 , {1,2}
10:57:49 GMT <mished> does not seem to work
11:17:44 GMT <mished> oh... just found the answer - I can use cjson right inside the lua script itself :)
14:15:32 GMT <byoungb> Anyone have a very good handle on redis replication / sentinal?
14:16:14 GMT <byoungb> I have been using redis for a while with 2 redis servers behind haproxy, and then 3 sentinal servers monitoring
14:17:21 GMT <byoungb> We are using it in production, and it has been working well.... but we had a split brain issue a while back when the master went offline and it failed over to the slave
14:18:13 GMT <byoungb> and then the original master came back on and was also a slave for quite a while since haproxy didn't know any better and it took a long time for sentinal to kick it off
14:18:24 GMT <byoungb> I read this
14:18:25 GMT <byoungb> https://selivan.github.io/2016/06/10/redis-no-splitbrain-on-network-partition.html
14:18:52 GMT <byoungb> which seemed perfect but I can't get that to work with sentinel at all
14:33:15 GMT <similian> hi all
14:33:47 GMT <similian> is there a way to config redis to limit 5 GB data for example and reject write
14:34:02 GMT <similian> until other serives pull / delete their stuff
14:34:25 GMT <similian> ( we are pushing logs into redis -> logstash
14:34:38 GMT <similian> during production everything is superb
14:34:43 GMT <similian> yet when we reindex
14:35:29 GMT <similian> the log-pusher pushes all logs
14:35:47 GMT <similian> infinite GBs
14:36:25 GMT <similian> and death is imminent
14:38:46 GMT <similian> i would like to set maxmemory limit
14:39:38 GMT <similian> we have maxmemory and noeviction
19:29:58 GMT <aqar> can someone explain to me the difference between clustered redis and etcd ?
19:30:17 GMT <aqar> (in comparable setups)
21:02:42 GMT <aqar> nobody knows ?
21:09:26 GMT <minus> isn't etcd rather for configuration data?
21:09:42 GMT <aqar> i'm talking for using redis as an etcd replacment
21:09:46 GMT <aqar> is there any downside ?
21:10:13 GMT <minus> i guess it'd work
21:10:28 GMT <aqar> looking for a more comprehensive answer :)
21:10:31 GMT <minus> but etcd might give you some consistency guarantees redis can't give you, at least not easily
21:10:38 GMT <aqar> which ?
21:11:21 GMT <minus> i don't know etcd, but redis replication (thus failover consistency) is asynchronous