15:05:04 GMT <sembiance> badboy_: Scaled down the DB from 47GB to just 6GB. `MEMORY DOCTOR` still results in a crash. tried with both 4.0-rc2 and latest unstable branch and lastest 4.0 branch from github. Submitting an issue now with crash log and redis.conf
15:05:14 GMT <sembiance> badboy_: issue submitted.
15:06:06 GMT <badboy_> interesting
15:06:34 GMT <sembiance> if a dump would be really useful, I could spend some time pairing it down and removing any sensitive info so it could be shared.
15:07:05 GMT <badboy_> I can crash it with a nearly empty instance
15:07:12 GMT <badboy_> so the dump is not really needed
15:07:13 GMT <sembiance> ahh , well then :)
15:08:54 GMT <badboy_> investigating
15:21:20 GMT <badboy_> uuuuh found the problem
15:22:13 GMT <badboy_> sembiance: you don't have a slave attached right?
15:24:59 GMT <badboy_> sembiance: fixed it: https://github.com/antirez/redis/pull/3784
15:46:53 GMT <badboy_> and merged
17:58:14 GMT <sembiance> badboy_: yup. fixed. thanks :)
17:58:21 GMT <sembiance> the doctor didn't find anything wrong :)
18:37:45 GMT <sembiance> thank goodness for the new `MEMORY USAGE key` command. so useful :)
18:38:50 GMT <minus> TIL that exists
18:39:04 GMT <minus> i did it using rdbtools and a dump
19:23:01 GMT <sembiance> well, in a few months I think I'm gonna have to get me a server that has more than 128GB of memory as I'm using about 47GB of RAM already for redis and since you need twice that during saving. :)
19:27:39 GMT <minus> it shouldn't need twice the memory unless you have overcommit turned off or have a massive turnover
20:11:39 GMT <Habbie> minus, unless you manage to churn ALL your data while save is ongoing, right
20:11:45 GMT <Habbie> minus, which means you probably maybe should not be saving
20:12:18 GMT <minus> seems about right
20:12:32 GMT <minus> source: hearsay