08:45:35 GMT <xqo> I'm making a modeling agency site, which lets you browse models and their portfolio-photos. each model has some data like hair-color, eye-color, and the agency must be able to add/remove models, add/remove photos, and edit the data. Is redis suitable for this project, or would I be better off with something like postgres?
08:56:23 GMT <badboy_> xqo: that sounds like easy to model using relational data, which means you are likely better off using an sql database
08:56:33 GMT <badboy_> of course you can use Redis in addition as a cache or similar
09:27:37 GMT <xqo> thank you
10:47:10 GMT <redis_noob> How to get key: value pair in redis hash? I mean hmget require keys to return the value. What method can i use which will return all the keys and corresponding values in hash stored in redis. I need it because I don't know which keys will be avilable at a given time in my redis.
10:51:18 GMT <redis_noob> Anyone?
19:50:17 GMT <mainiak> Hi everyone. I am writing (not yet released) open source project based on Redis and I found myself in little trouble with my Lua scripts. They work most of the time but not all the time / as expected. Any advice would be highly appreciated.
19:51:54 GMT <mainiak> For example https://gist.github.com/mainiak/5d17acfc4e2b580f212e91f6955853c6 ... this doesn't work, but when I return all hostnames in Lua table, it works.
19:54:32 GMT <minus> mainiak: redis-cli has a debugger for scripts, give that a try
19:55:00 GMT <minus> i'd guess you might have to return something that's not a plain string
19:55:09 GMT <minus> a table with an ok field or so
19:55:15 GMT <minus> should be in the docs
19:58:03 GMT <mainiak> minus: redis-cli is only way to debug Lua scripts in Redis - meaning run script trough cli and debug it. Or I can attach debugger to existing Lua code executed on (testing) server?
20:03:03 GMT <mainiak> Also I don't think that Lua script is require to return complex structure - because I have working code returning integers just fine.
20:04:35 GMT <minus> dunno, i said everything i know. i haven't ever had to debug lua scripts.
20:07:07 GMT <mainiak> minus: Thank you for all your help!
20:07:57 GMT <mainiak> If authors and contributors are here on this channel - I have to say - I really love Redis. Great work guys.
20:11:15 GMT <tomaz_br> hi, this is my first time here. I am checking if Redis could be a solution to my problem. Long story short. I am getting JSON data from internet (Connected Car platform). If car is having a 30 minutes trip I collect this JSON files every minute. Due to further data enrichment and other modification on it I would have to keep this data in-memory for the time during the trip. After that data is persistent in another DB. I was checking if
20:11:16 GMT <tomaz_br>  Redis could be used for gathering JSONs for the duration of car trip (in memory) ... so somehow appending JSON files... under 1 unique ID (Device ID) where I could have even > 10.000 device ID active. Is this possible with Redis? Would it be hard to even somehow “persist” this for the case if Redis (Or better AWS EC2) crashes?
20:12:58 GMT <tomaz_br> Those separate Device ID "queues" would be various length in size and time... but eventually all would be "marked" as finished and then data would proceed to data enrichment and would be deleted from Redis
20:15:30 GMT <minus> should be okay: use lists to store your per-device data (LPUSH car:<id> <data>) or so. redis does save to disk regularly (but not immediately) with the example configuration (can be further tuned)
20:47:07 GMT <tomaz_br> minux: so with lists I would eventualy push into a list N number of complete JSONs? After all JSONs are there ... can I grab everything at once and manipulate it?
20:47:17 GMT <tomaz_br> minus: so with lists I would eventualy push into a list N number of complete JSONs? After all JSONs are there ... can I grab everything at once and manipulate it?
20:56:26 GMT <tomaz_br> can Redis be distributed?
21:00:17 GMT <tomaz_br> Ok... I can see Redis Cluster, master - slave. So are this nodes all writable?
21:00:46 GMT <tomaz_br> or are like read replicas? or standby nodes... for the case when master (writable) fails?
21:05:24 GMT <minus> yes, in the end you can just grab everything
21:05:46 GMT <minus> cluster partitions data (and can have slaves additionally for failover)