00:26:25 GMT <madsage> i found my problem
00:26:28 GMT <madsage> heh
00:27:18 GMT <madsage> sentinel auth-pass needed set
00:54:15 GMT <madsage> how do I make my my master node master again after failover to a slave
00:54:34 GMT <madsage> looks like it worked. but its staying there
00:54:57 GMT <madsage> i need a script for that?
01:00:02 GMT <madsage> wow lots of chatter here
01:00:03 GMT <madsage> lol
09:38:17 GMT <alkuzad> hi redis users, is it possible to use list commands with master/slave configuration or redis-cluster ?
09:38:31 GMT <alkuzad> technically lpop is read + pop
09:55:59 GMT <King_DuckZ> hi, anyone's got the time to check my cpp redis lib code? any advice/opinion on it is welcome https://github.com/KingDuckZ/dindexer/tree/dev/src/backends/redis
16:28:07 GMT <King_DuckZ> regarding redis+lua, the documentation says "all the keys that the script uses should be passed using the KEYS array", but what if I don't know the key name at the moment I call the script?
16:29:02 GMT <King_DuckZ> like, I'm going to set foo:3, which is the result of "foo" .. ":" .. id
16:57:51 GMT <King_DuckZ> nobody around? I can't find any info on the internet
16:59:44 GMT <King_DuckZ> oh I think I just found my answer... http://stackoverflow.com/questions/32089406/why-there-are-keys-and-argv-arrays-when-calling-redis-lua-scripts
17:00:42 GMT <King_DuckZ> so this makes it impossible to have a compliant implementation of this script https://github.com/soveran/ohm-scripts/blob/master/save.lua
17:29:02 GMT <King_DuckZ> everybody seems to be just generating keys regardless of what the documentation says http://tjholowaychuk.tumblr.com/post/19321054250/redis-lua-scripting-is-badass
19:40:57 GMT <iamchrisf> Is there a way to get the total system memory of the instance redis is running on with a call to redis?  I am checking a remote redis instance(elasticache) and the sensu redis scripts are borked at they get total memory from /proc/meminfo … So I need to fix the script to grab the remote total_memory
21:03:01 GMT <minus> iamchrisf: redis-cli info memory?
21:03:14 GMT <minus> is that what you're looking for?
22:49:09 GMT <steeze> im a bit confused. how are strings when used like pages:about different from a hash?
22:49:40 GMT <steeze> because then you can have pages:home pages:information and it's that like having pages: {home: '', information: '', about: '' }
22:50:59 GMT <steeze> looks like the colon syntax with strings was a way to namespace, but wouldnt it kind of be unnecessary since we have hashes?
22:51:35 GMT <steeze> since we have hashes now*
22:54:45 GMT <steeze> im trying to pick the best structure for what im doing but having some issues understanding everything clearly. i have an array of users and an array of stores. i want to store # of items bought at each store for each user
22:55:32 GMT <steeze> would it be best to do a namespaced hash maybe? like user_item_count:<user_id>:<store_id> and my value be an array of product ids purchased by that user at that store
22:55:43 GMT <steeze> but wait that's just a key/val with a fancy string right?
22:59:34 GMT <minus> hashes are more space efficient for one
23:02:41 GMT <steeze> minus, disregarding implementation, what are the applicable differences?
23:03:09 GMT <steeze> or i guess is there a reason for me to not use a hash?
23:04:07 GMT <minus> can't be nested, have a limit wrt size and amount of keys (above which they don't perform as well anymore)
23:04:29 GMT <minus> if nesting isn't a problem, then why not
23:10:18 GMT <steeze> so can you also namespace your hash key's?
23:14:05 GMT <minus> namespaces aren't a thing really
23:14:20 GMT <minus> redis just interprets keys as data
23:14:31 GMT <minus> (yes, also binary data)
23:14:54 GMT <minus> it has no concept of namespaces, those are purely for the humans
23:16:11 GMT <steeze> i think what's confusing me is looking at my boss's code and seeing him set a key/val then using that key/val to later look up hash values
23:16:17 GMT <steeze> but im so confused why we've done it that way
23:18:02 GMT <minus> magic!
23:18:21 GMT <steeze> haha no i just dont understand why we would need to do that weirdness if we have hashes
23:18:36 GMT <minus> if anyone ever tried to understand the redis stuff i built at work, oh boy
23:18:40 GMT <steeze> lol
23:18:48 GMT <steeze> okay that makes me feel a bit better i guess
23:19:18 GMT <steeze> idk why this is so hard for me to grok
23:19:23 GMT <steeze> i keep thinking in circles
23:19:46 GMT <minus> actually it's just using a zset for range lookups (like ip to ip network resolution) and using hashes to cluster/bucket info
23:20:39 GMT <steeze> that doesn't sound terrible
23:20:43 GMT <minus> sounds like you have a 2-stage lookup
23:20:50 GMT <steeze> but why
23:20:59 GMT <minus> idk
23:21:11 GMT <minus> i tried to avoid that very much
23:23:00 GMT <steeze> https://gist.github.com/steezeburger/c348bdb946b0e75911f0da477ab075cc#file-sample-php
23:23:11 GMT <steeze> like i get what's happening, but not why
23:24:23 GMT <minus> oh
23:24:57 GMT <minus> you need to get the pids and uids you want to look into from somewhere of course
23:25:46 GMT <steeze> ohh is this because you cant nest?
23:25:47 GMT <minus> redis does have a KEYS command that you could use to find all matching keys, but it's discouraged because it's got linear complexity
23:26:05 GMT <minus> yes, you cannot nest things
23:26:07 GMT <steeze> so i have to have some structure to store some info to use to query my hashes?
23:26:14 GMT <steeze> okay that makes sense
23:26:20 GMT <steeze> i think
23:26:21 GMT <steeze> haha
23:26:54 GMT <minus> i mean, you could've probably packed the data into the hash as well
23:28:20 GMT <steeze> i think im going to use a hash like ` HSET user_event_ids:<user_id>  <store_id> [array of product ids]`
23:29:19 GMT <minus> sounds decent enough
23:30:22 GMT <minus> good night and good luck with your redis adventures
23:30:33 GMT <steeze> thanks for the help!
23:30:35 GMT <steeze> gnight
23:30:41 GMT <minus> you're welcome