CentOS + Nginx from sources. Why is the service not starting?

Question

linux

Good day dear!
Due to subjective reasons, I decided to tinker with CentOS (before that, I tinkered with Debian for 10+ years). I downloaded CentOS-7, installed it on a virtual machine, updated it, everything is fine. I decided to build Nginx from source, because. this is one of the most frequent procedures in my practice (building software from source) when working with Linux, and Nginx is a program that is quite familiar to me. Well, for a change, I decided to screw ngx_pagespeed into it.
I read the instructions, for example here (there are a lot of very similar ones, I won’t give all the links, they are essentially the same).
I did everything as it should be, downloaded src.rpm, unpacked it, added the module, built Nginx, delivered the dependencies, built the package, installed the package via: yum install local nginx...rpm
I'm writing: service nginx start- error. I look at the error:

янв 21 18:19:27 localhost.localdomain systemd[1]: Starting nginx - high performance web server...
янв 21 18:19:27 localhost.localdomain nginx[2241]: /usr/sbin/nginx: error while loading shared libraries: cannot create cache for search path: Cannot allocate memory
янв 21 18:19:27 localhost.localdomain systemd[1]: nginx.service: control process exited, code=exited status=127
янв 21 18:19:27 localhost.localdomain systemd[1]: Failed to start nginx - high performance web server.

Ok, I try to start nginx with my hands, I write the path as in an error: /usr/sbin/nginx- lo and behold, Nginx has started!
I try again - service start nginx- the error is the same. From the error, I realized that nginx cannot load the "space" libraries, cannot create a folder for the cache at the address "space" and cannot allocate memory ...
Ok, I'm going to look at the "magic" service startup parameters: /usr/lib/systemd/system/nginx.service
I see:

[Service]
Type=forking
PIDFile=/run/nginx.pid
ExecStartPre=/usr/sbin/nginx -t -c /etc/nginx/nginx.conf
ExecStart=/usr/sbin/nginx -c /etc/nginx/nginx.conf
ExecReload=/bin/kill -s HUP $MAINPID
ExecStop=/bin/kill -s QUIT $MAINPID
PrivateTmp=true

I start in order, "hands":
/usr/sbin/nginx -t -c /etc/nginx/nginx.conf
Conclusion:

nginx: the configuration file /etc/nginx/nginx.conf syntax is ok
nginx: configuration file /etc/nginx/nginx.conf test is successful

I start further: /usr/sbin/nginx -c /etc/nginx/nginx.conf
And... nginx is started! It is in the processes and netstat shows that the process is listening on the port such and such, etc. I mean, everything is great.
I try to start the service again, after unloading nginx from memory, of course.:

nginx -s stop
service nginx start

- we return to the same as it was at the beginning.
Can you please explain what kind of magic is this?

Answer 1

[nginx] Don't use type=forking in systemd service file
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=...

Answer 2

This is a memory allocation error. It can happen to both muscle and gzip. Either the server eats a lot of memory when it is started from service, or there is little memory in principle. See how much it eats.