The CentOS7 version is fine, but RockyLinux 8 has a high overall memory footprint.
In addition, the nginx - reload command does not return memory normally and causes a leak.
RPM and compile installation are all the same symptoms
I’ve also tried changing the nginx version, but I couldn’t solve it.
I would like to ask if there is a solution regarding this.
Absolutely seeing the same behavior. We’ve switched from Centos 7 to RL8 and nginx is constantly allocating memory, which turn out to be inactive anon pages. Upon reload, if the available memory is low, you can get oom errors when new workers are fork()-ed.