01:39:15 GMT <Sazpaimon> If I'm using elasticache in production (which uses Redis 2.8), is it safe to use a more recent redis docker version for local development?
01:39:41 GMT <Eddy> Hi guys.
01:40:22 GMT <Eddy> I have a weired Redis issue. Could I get any advices?
01:44:45 GMT <Eddy> I use ElastiCache Redis on the AWS. And the connection is between two AWS regions.
01:45:21 GMT <Eddy> The connection is OK, but some particular commands does not work such as 'info' or 'client list'
01:45:38 GMT <Eddy> GET command also is OK.
09:32:14 GMT <djinja> need to say this (noob speaking): I just gained 50% speed up by using one ZADD for an entire set rather than a for loop of ZADDs … wow!
09:37:20 GMT <djinja> if a zadd is executed with multiple scores,item pairs, is that whole atomic? I mean the server will wait for the whole command to run before accepting another command, right?
10:50:07 GMT <chxd> hello, what can I do if I have from time to time GET commands very slow - 1-3 seconds, give me a direction to go, please :)
11:14:05 GMT <debjeetsarkar> Do we need to specify something in the conf file of the slave or master to make the lsave propagate itself to master automatically when the master fails
11:26:08 GMT <minus> chxd: check the slowlog, monitor your system load (CPU, IO), i guess
11:26:48 GMT <minus> debjeetsarkar: to get failover you need redis sentinel
12:54:32 GMT <debjeetsarkar> @minus , How does redis-cluster works then in terms of slave propagation ?
12:59:16 GMT <minus> that does failover too of course
13:19:11 GMT <debjeetsarkar>  @minus So redis sentinel is required anyways you are saying for the slave to propagate ?
13:20:28 GMT <minus> if you're using redis-cluster you don't need sentinel, cluster does the failover itself in this case
13:21:15 GMT <minus> if you want failover you either need sentinel or in the case where you already have a cluster you don't need anything additional
13:21:48 GMT <debjeetsarkar> @minus yaa , thats what
15:44:00 GMT <xep> topic is out of date -- Redis 3.2.3 is out? :)
16:36:33 GMT <kutenai> I've had an m1.small EC2 server running for about 3 years now. I have a redis instance I use as a part of a local pump control. Lately, I'm having a lot of issues with redis, and I have no idea why. The main symptom is extremely high latency, and regular connection timeouts.
16:36:55 GMT <kutenai> I worked on this till 3am.. still can't figure out why redis is performing so poorly right now.
16:37:07 GMT <kutenai> Any tips?
17:35:08 GMT <kutenai> Just in case someone can help..
17:36:09 GMT <kutenai> My real problem seems to be connections. If I just do this: echo "keys * " | redis-cli -h <host> -p 6279 -n 5  ;; I find that this command fails to connect about 2 out of 3 times..
17:36:30 GMT <kutenai> The latency isn't as big of an issue now that I look at it. My worst --intrinsic-latency is Max latency so far: 32623 microseconds.
17:36:51 GMT <kutenai> This isn't that high of a throughput issue..
17:37:16 GMT <kutenai> But, failing to connect half the time, that is the issue. I just can't figure out why the connections would be so flaky.
17:42:16 GMT <TechIsCool> So Redis is overriding my data_dir to '/' even though I set it to /var/lib/redis
17:42:33 GMT <TechIsCool> why and how does that happen
17:54:07 GMT <sknebel> kutenai: only happening with redis, or also with other things? Maybe the instance itself has some issue
17:55:12 GMT <kutenai> I've rebooted it a couple of times.. I've been looking at what else could be wrong with it
17:58:15 GMT <minus> kutenai: maybe take a look at it with wireshark. if you're going over the internet (which probably is a bad idea with redis since there's no encryption) it could very well be that some part drops stuff
17:59:36 GMT <kutenai> good suggestion.
18:00:24 GMT <kutenai> In this case, security isn't a concern. A) the data needs no security, just ditch levels, B) I have it IP locked on AWS.
18:00:52 GMT <kutenai> I didn't realize wireshark had an OSX version..
18:03:25 GMT <kutenai> The only thing I can tell for sure is that it's an old, m1.small instance. This is 'reported' to be less than idea for redis..
18:03:47 GMT <kutenai> but, this always worked before.. so, why it started failing, I don't know.
18:27:29 GMT <kutenai> minus: it really is not redis, it's the instance.. mysql has the same issue..
19:14:15 GMT <aycaramba> hi
19:15:36 GMT <aycaramba> I have a question regarding the use of hiredis. I am seeing some possibly loss memory leaks in valgrind. But i do not know if i am doing something wrong or not
19:16:49 GMT <aycaramba> here is the part that valgrind shows were the bytes could be lost : https://gist.github.com/pablin87/fee0736507142dec089423de880051ac
19:17:41 GMT <aycaramba> But the part of code there is not from my code... so maybe i am not freeing something that i am supposed to free
19:18:08 GMT <aycaramba> i do not know i someone has encountered this same problem....
19:47:52 GMT <rhizome> hi all. is there a way to use a specific config file on linux (at boot) that updates won't complain about or clobber? be nice if there was an /etc/redis/redis.conf.d built in
19:51:29 GMT <rhizome> and, naturally, this means uncommenting/adding the include line(s) in the default redis.conf will result in upgrade pain
19:52:06 GMT <minus> sure, just create your own config and your own service
20:22:51 GMT <rhizome> grmbl
20:44:30 GMT <minus> you could also chattr +i the config to prevent it from being modified
21:39:27 GMT <dingdong> greetings #redis - what am i doing wrong?
21:39:42 GMT <dingdong> this lua script fails on line 9
21:39:58 GMT <dingdong> on 3.x - fine on < 3.x
21:40:27 GMT <dingdong> #9 is the EXISTS call
21:40:41 GMT <dingdong> local lst = KEYS[1] local result = nil repeat     result = redis.call("LPOP", lst)     if result == nil then         break     end     if (not redis.call("EXISTS", result)) then         result = nil     end until (result ~= nil) return result
21:40:52 GMT <dingdong> oh, great :-(
21:42:25 GMT <dingdong> even though result is NIL after the LPOP the if-clause does not break
21:44:38 GMT <minus> you could've used a pastebin, you know
21:47:16 GMT <dingdong> like this? http://pastebin.com/ijVdZDeL
22:30:58 GMT <aarwine> I'm trying to figure out if the lazy free ever made it out? https://github.com/antirez/redis/issues/1748 http://antirez.com/news/93
22:31:35 GMT <aarwine> I'm seeing 3.2 and 3.4 versions being thrown around, but these posts were a long time ago, so I wanted to confirm
22:32:48 GMT <aarwine> oh shoot, 3.2 released this month - apologies