Systemd

uWSGI is a new-style daemon for systemd.

It can notify status change and readyness.

When uWSGI detects it is running under systemd, the notification system is enabled.

Adding the Emperor to systemd

One approach to integrate uWSGI apps with your init system is using the Emperor.

Your init system will talk only with the Emperor that will rule all of the apps itself.

Create a systemd service file (you can save it as /etc/systemd/system/emperor.uwsgi.service)

Note

Be careful with some systemd versions (e.g. 215 in Debian Jessie), since SIGQUIT signal will trash the systemd services. Use KillSignal=SIGTERM + “die-on-term” UWSGI option there.

  1. [Unit]
  2. Description=uWSGI Emperor
  3. After=syslog.target
  4.  
  5. [Service]
  6. ExecStart=/root/uwsgi/uwsgi --ini /etc/uwsgi/emperor.ini
  7. # Requires systemd version 211 or newer
  8. RuntimeDirectory=uwsgi
  9. Restart=always
  10. KillSignal=SIGQUIT
  11. Type=notify
  12. StandardError=syslog
  13. NotifyAccess=all
  14.  
  15. [Install]
  16. WantedBy=multi-user.target

Then run it

  1. systemctl start emperor.uwsgi.service

And check its status.

  1. systemctl status emperor.uwsgi.service

You will see the Emperor reporting the number of governed vassals to systemd (and to you).

  1. emperor.uwsgi.service - uWSGI Emperor
  2.  Loaded: loaded (/etc/systemd/system/emperor.uwsgi.service)
  3.        Active: active (running) since Tue, 17 May 2011 08:51:31 +0200; 5s ago
  4. Main PID: 30567 (uwsgi)
  5.        Status: "The Emperor is governing 1 vassals"
  6.        CGroup: name=systemd:/system/emperor.uwsgi.service
  7.                ├ 30567 /root/uwsgi/uwsgi --ini /etc/uwsgi/emperor.ini
  8.                ├ 30568 /root/uwsgi/uwsgi --ini werkzeug.ini
  9.                └ 30569 /root/uwsgi/uwsgi --ini werkzeug.ini

You can stop the Emperor (and all the apps it governs) with

  1. systemctl stop emperor.uwsgi.service

A simple emperor.ini could look like this (www-data is just an anonymous user)

NOTE: DO NOT daemonize the Emperor (or the master) unless you know what you are doing!!!

  1. [uwsgi]
  2. emperor = /etc/uwsgi/vassals
  3. uid = www-data
  4. gid = www-data

If you want to allow each vassal to run under different privileges, remove the uid and gid options from the emperor configuration (and please read the Emperor docs!)

Logging

Using the previous service file all of the Emperor messages go to the syslog. You can avoid it by removing the StandardError=syslog directive.

If you do that, be sure to set a —logto option in your Emperor configuration, otherwise all of your logs will be lost!

Putting sockets in /run/

On a modern system, /run/ is mounted as a tmpfs and is the right place to put sockets and pidfiles into. To have systemd automatically create a /run/uwsgi/ subdirectory with the correct user/group ownership, as well as cleaning up the directory when the daemon is stopped, add

  1. RuntimeDirectory=uwsgi

to the [Service] section of your systemd uwsgi unit file. This RuntimeDirectory parameter requires systemd version 211 or newer. For older versions of systemd, create a systemd-tmpfiles configuration file (you can save it as /etc/tmpfiles.d/emperor.uwsgi.conf):

  1. d /run/uwsgi 0755 www-data www-data -

Socket activation

Starting from uWSGI 0.9.8.3 socket activation is available. You can setup systemd to spawn uWSGI instances only after the first socket connection.

Create the required emperor.uwsgi.socket (in /etc/systemd/system/emperor.uwsgi.socket). Note that the *.socket file name must match the *.service file name.

  1. [Unit]
  2. Description=Socket for uWSGI Emperor
  3.  
  4. [Socket]
  5. # Change this to your uwsgi application port or unix socket location
  6. ListenStream=/tmp/uwsgid.sock
  7.  
  8. [Install]
  9. WantedBy=sockets.target

Then disable the service and enable the socket unit.

  1. # systemctl disable emperor.uwsgi.service
  2. # systemctl enable emperor.uwsgi.socket

When using Systemd socket activation, you do not need to specify any socket in your uWSGI configuration;the instance will inherit the socket from Systemd.

To have uWSGI serve HTTP (instead of the binary uwsgi protocol) under Systemd socket activation,set protocol to http; for instance, in an INI, do this:

  1. [uwsgi]
  2. protocol = http
  3. wsgi = ...
  4. ...

One service per app in systemd

Another approach is to let systemd handle starting individual apps while takingadvantage of systemd template unit files, and of course socket activation. Eachapp will run under its own user.

/etc/systemd/system/uwsgi-app@.socket:

  1. [Unit]
  2. Description=Socket for uWSGI app %i
  3.  
  4. [Socket]
  5. ListenStream=/var/run/uwsgi/%i.socket
  6. SocketUser=www-%i
  7. SocketGroup=www-data
  8. SocketMode=0660
  9.  
  10. [Install]
  11. WantedBy=sockets.target

/etc/systemd/system/uwsgi-app@.service:

  1. [Unit]
  2. Description=%i uWSGI app
  3. After=syslog.target
  4.  
  5. [Service]
  6. ExecStart=/usr/bin/uwsgi \
  7.         --ini /etc/uwsgi/apps-available/%i.ini \
  8.         --socket /var/run/uwsgi/%i.socket
  9. User=www-%i
  10. Group=www-data
  11. Restart=on-failure
  12. KillSignal=SIGQUIT
  13. Type=notify
  14. StandardError=syslog
  15. NotifyAccess=all

Now, adding a new app to your system is a matter of creating the appropriateuser and enabling the socket and the service. For instance, if one were toconfigure cgit:

  1. adduser www-cgit --disabled-login --disabled-password \
  2.   --ingroup www-data --home /var/lib/www/cgit --shell /bin/false
  3. systemctl enable uwsgi-app@cgit.socket
  4. systemctl enable uwsgi-app@cgit.service
  5. systemctl start uwsgi-app@cgit.socket

Then configure the ini file /etc/uwsgi/apps-available/cgit.ini:

  1. [uwsgi]
  2. master = True
  3. cheap = True
  4. idle = 600
  5. die-on-idle = True # If app is not used often, it will exit and be launched
  6.                    # again by systemd requested by users.
  7.  
  8. manage-script-name = True
  9.  
  10. plugins = 0:cgi
  11. cgi = /usr/lib/cgit/cgit.cgi

And last, if applicable, configure your HTTP server the usual way.