09:28:15 GMT <abcd> Hey anybody there? I need Urgent help
13:10:44 GMT <nirvanko> What is default location of the Redis conf file?
13:11:16 GMT <nirvanko> And how can I specify it, will that work? redis-server /path/to/redis.conf
13:13:47 GMT <minus> There's no default and yes
13:16:52 GMT <nirvanko> thanks
14:31:42 GMT <theorem> woo
14:36:40 GMT <theorem> if I'm upgrading from 2.8.24 to 3.2.44 are there things that I need to look out for ?  I want to use the clustering capability of 3.2.4
14:40:53 GMT <_Wise_> Hi *,
14:42:57 GMT <_Wise_> I'm using Redis 3.0 and I'm having hard times when a slave reboots and reconnects to a master that syncs a 40GB database
14:43:05 GMT <_Wise_> Should I consider diskless sync ?
14:43:15 GMT <_Wise_> this is experimental in 3.0 according to doc
14:51:02 GMT <minus> _Wise_: maybe increasing the sync backlog can help if your slave is persisting to disk, so that it can resume syncing and doesn't need to do a full sync
14:51:30 GMT <minus> theorem: the standard answer is: should be fine but check the changelog
14:51:44 GMT <theorem> minus: hmm, yes.  that's generally what I was thinking.
14:51:51 GMT <minus> didn't have any problems here, but it's always good to have a backup plan
14:52:16 GMT <theorem> minus: is there a "good practice" way of testing the cache to make sure it all works ?  (test scripts?)
14:52:49 GMT <minus> nah, if it loads it works
14:52:57 GMT <theorem> ok
14:52:58 GMT <theorem> :)
14:54:25 GMT <_Wise_> minus: I tried that. I changed the default 1MB to 100MB, but even if I only add one 'hello world' key to the master while the slave is down, a full sync is triggered
14:54:41 GMT <_Wise_> minus: and yes, both master and slave are aof-persistent
15:00:00 GMT <badboy_> resume syncing won't work after a reboot
15:09:47 GMT <_Wise_> badboy_: I'm not rebooting, I'm just restarting slave service
15:10:00 GMT <badboy_> reboot of the slave
15:10:30 GMT <badboy_> partial sync (=resume syncing) only works after a brief network connection loss, not across different processed
15:10:38 GMT <_Wise_> aaah , ok
15:12:01 GMT <_Wise_> you made my day, I now have a story to tell :)
16:01:49 GMT <minus> why would that not work after a reboot?
17:16:27 GMT <_Wise_> whois badboy_
17:17:51 GMT <badboy_> _Wise_: It's me!
17:17:58 GMT <badboy_> minus: because the ruid of the master is not persisted
17:18:10 GMT <badboy_> and that runid is used to determine if a psync is even possible
17:18:21 GMT <minus> i see
17:18:25 GMT <minus> that sucks a bit, doesn't it?
17:18:58 GMT <badboy_> weeeeell
17:19:02 GMT <badboy_> a bit
17:19:07 GMT <badboy_> but no one implemented it yet
17:55:09 GMT <minus> it's pretty terrible for that one server we have in china. a couple hundred megs at 50KB/s is no fun
23:10:18 GMT <_savage> Hello. Question: when I run a Redis server and multiple people on localhost use it, I assume they won't be able to see the others' data?
23:12:47 GMT <minus> well, you're assuming wrong there
23:13:30 GMT <_savage> minus: So it really is just a flat data store?  If I flush then everybody's data is gone?
23:13:34 GMT <minus> at most you can put a password on it
23:13:45 GMT <minus> yup
23:14:01 GMT <_savage> minus: well I guess that's a good thing to keep in mind :)
23:14:07 GMT <minus> if you need separation, run multiple instances and bind them to a unix socket only
23:14:39 GMT <_savage> minus: that makes sense, thank you.
23:15:17 GMT <_savage> Another question: does Redis back someplace persistent too if I need to shut down? Or is that optional?
23:16:24 GMT <minus> that's optional
23:16:41 GMT <minus> it is enabled in the example config though
23:18:29 GMT <_savage> minus: thank you, I'll take a look.
23:26:27 GMT <_savage> minus: So everybody who connects to Redis through the same socket sees the same "namespace" (or whatever it's called) no matter what user they actually are?
23:27:33 GMT <minus> there are no users in redis and redis doesn't try to find out what unix user you are
23:30:20 GMT <_savage> minus: Ta.