Hello. Why sentinel remove is not considered valid configuration statement?
geeze, I was looking forward nice ansible role, but this.
dosmanak: because it is a command
the configuration file isn't really the place to do dynamic reconfiguration
So why redis stores its dynamic configuration into this file.
Design Decision ^^
Althought I appended include of my setting at the end of the file, then new lines are appended at the end, so my local config is not true anymore.
Sentinel is not particular friendly to be fully managed by configuration management systems
So I cannot prepare static config file, which will get my redis architecture into known state.
Ok, thanks for discussion, I will try different aproach :-)
You can use an initial config file, but Redis will rewrite it, yes
One more thing.
Oh, not really, bad idea :-)
But it seems feasible to maintaing basic configuration in systemd service.
But still sentinel monitor is command and config statement, but sentinel remove is only command, oh man...
is it necessary to use a redis instance (c = redis.Redis()) in case of multithreaded applications ? (I mean for each threads)
Check your client library's documentation on whether it is thread-safe
long gone, badboy_, long gone.
Migrating redis data always seemed like a daunting task, but I just discovered that I can set one instance as a slave of another, and all the data replicates so quick. I set up an SSH tunnel to a remote server hosting 11GB of redis data, and within about five minutes, it was all replicated on the slave. I set it back to a master and the job was done. I just wanted to share this. Fantastic experience.
if you need to clone all data, yeah
hello everyone, i was curious if anyone has seen an issue where after increasing 'hz', memory exponentially increases. I'm seeing this after i bumped hz from 100 to 500. despite reverting the change, memory is still slowly getting consumed and never releasing.