03:54:41 GMT <Redis-Opsworks> Any pointers for Redis Opsworks cookbook?
08:46:17 GMT <JEEB> so the redis FAQ/some of the documentation recommends doing kernel poking wrt THP and overcommit. I take it that those are required for specific use cases rather than "whatever the box it is, you should do these kernel changes." as I'm still not sure if redis will share a system with other services or not. if redis was always defined to be on its own "node" then doing kernel configuration entries for it
08:46:23 GMT <JEEB> without too much thinking would be a no-brainer of course
09:00:07 GMT <minus> The thp thing is a performance thing
09:00:59 GMT <minus> So if you don't need full performance it shouldn't matter
10:13:47 GMT <King_DuckZ> minus: hi, I've been looking into escaping html into tawashi as you suggested, but I don't think I need to do that, since pasties either go through the syntax colorizer already or are returned as plain text
10:14:24 GMT <King_DuckZ> maybe it's a bit offtopic here, but I wanted to hear your opinion... maybe we can continue in pvt?
10:15:32 GMT <King_DuckZ> minus: https://github.com/KingDuckZ/tawashi/blob/master/src/pastie_response.cpp#L68 if I escape html I get strange results anyways
10:50:17 GMT <adac> Hi there! Was wondeing if there  are any news about using redis with docker swarm? Is there now a good way to do that?
11:47:09 GMT <minus> King_DuckZ: yes, if you transmit it as text/plain or run it through a html syntax highlighter it'll be fine
11:54:00 GMT <King_DuckZ> minus: I'm a bit confused tbh... if you paste html and then choose to highlight syntax in a language that is not html and that leaves <tags> untouched, I'm confused as to what would happen there
11:54:49 GMT <minus> it never leaves <> untouched because the display medium doesn't allow it to do so
11:58:34 GMT <King_DuckZ> good point, it should be ok then
11:59:08 GMT <King_DuckZ> I'm writing code to sanitize utf8 input, then I'm confident enough to expose it online, so it can be demo'ed :)
14:57:13 GMT <efphe> ?
15:00:50 GMT <minus> !
20:58:11 GMT <dannysantos> I have many node servers on my pc and I have made 1 redis-server for each node server. My goal is to have 1 separate database for each node app/server. I can't seem to acomplish this. I have different redis process id but one modification on one redis database replicates to the other database. How can I make this not happen? Thank you
21:00:37 GMT <minus> one redis per application is good if they are not meant to interact
21:03:02 GMT <dannysantos> yes, I have one redis per application. But although I have 2 database files, when I change one the other also changes. I don't want that
21:03:39 GMT <dannysantos> how can I make each database separate?
21:07:43 GMT <dannysantos> I can't seem to find someone that struggles with this, but for me it doesn't work
21:08:05 GMT <dannysantos> what information do you need in order to see what am I doing wrong?
21:20:51 GMT <dannysantos> in node app I was setting up the redis client port
23:22:16 GMT <dfin> Hi all, looking for a little clarification on Sentinel. In the docs it says that a majority is needed to preform a failover. However in the three box setup (1 master, 2 replicas, each with sentinel on them) it says that when a network partition removes he master, Sentinel will elect one of the replicas to a master. I thought that couldn't happen without a majority vote. Shouldn't that result in a deadlock?
23:22:46 GMT <dfin> Link: https://redis.io/topics/sentinel#example-2-basic-setup-with-three-boxes