1

I've wrote a systemd file for systemctl to start unicorn:

[Unit]
Description=Unicorn server

[Service]
SyslogIdentifier=my-app-unicorn
User=deployer
PIDFile=/tmp/unicorn.my-app.pid
WorkingDirectory=/opt/www/my-app.com

ExecStart=/home/deployer/.rvm/gems/ruby-2.2.1@my-app/bin/bundle exec "unicorn_rails -D -c /opt/www/my-app.com/config/unicorn.rb -E production"
#ExecReload=/bin/kill -s HUP $MAINPID
ExecReload=/bin/kill -s USR2 $MAINPID
ExecStop=/bin/kill -s QUIT $MAINPID

[Install]
WantedBy=multi-user.target

Here are the commands I used to start the service

$ sudo systemctl daemon-reload
$ sudo systemctl start my-app.service

Here I check the status:

$ sudo systemctl status my-app
● my-app.service - My app unicorn server
   Loaded: loaded (/lib/systemd/system/my-app.service; disabled; vendor preset: enabled)
   Active: failed (Result: exit-code) since Tue 2016-03-15 14:56:31 UTC; 4s ago
  Process: 22165 ExecStop=/bin/kill -s QUIT $MAINPID (code=exited, status=200/CHDIR)
  Process: 22162 ExecStart=/home/deployer/.rvm/gems/ruby-2.2.1@my-app/bin/bundle exec unicorn_rails -D -c /opt/www/my-app.com/config/unicorn.rb -E production (code=exited, status=200/CHDIR)
 Main PID: 22162 (code=exited, status=200/CHDIR)

Mar 15 14:56:31 fat-man systemd[1]: Started My-App unicorn server.
Mar 15 14:56:31 fat-man systemd[22162]: my-app.service: Failed at step CHDIR spawning /home/deployer/.rvm/gems/ruby-2.2.1@my-app/bin/bundle: No such file or directory
Mar 15 14:56:31 fat-man systemd[1]: my-app.service: Main process exited, code=exited, status=200/CHDIR
Mar 15 14:56:31 fat-man systemd[1]: my-app.service: Control process exited, code=exited status=200
Mar 15 14:56:31 fat-man systemd[1]: my-app.service: Unit entered failed state.
Mar 15 14:56:31 fat-man systemd[1]: my-app.service: Failed with result 'exit-code'.

What could be the problem?

LE: The access grants of user deployer to the /opt/www/my-app directory:

$ ls -alh
total 84K
drwxr-xr-x 12 deployer admin 4.0K Mar 12 20:10 .
drwxr-xr-x  3 root     root  4.0K Mar 11 20:43 ..
drwxr-xr-x  9 deployer admin 4.0K Mar 11 20:52 app
drwxr-xr-x  2 deployer admin 4.0K Mar 11 20:52 bin
drwxr-xr-x  5 deployer admin 4.0K Mar 13 14:02 config
-rw-r--r--  1 deployer admin  153 Mar 11 20:52 config.ru
drwxr-xr-x  3 deployer admin 4.0K Mar 11 20:52 db
-rw-r--r--  1 deployer admin 1.9K Mar 11 20:52 Gemfile
-rw-r--r--  1 deployer admin 5.1K Mar 11 20:52 Gemfile.lock
-rw-r--r--  1 deployer admin  616 Mar 11 20:52 .gitignore
drwxr-xr-x  5 deployer admin 4.0K Mar 11 20:52 lib
drwxr-xr-x  2 deployer admin 4.0K Mar 12 20:53 log
drwxr-xr-x  3 deployer admin 4.0K Mar 12 19:12 public
-rw-r--r--  1 deployer admin  249 Mar 11 20:52 Rakefile
-rw-r--r--  1 deployer admin  478 Mar 11 20:52 README.rdoc
-rw-r--r--  1 deployer admin    9 Mar 11 20:52 .ruby-gemset
-rw-r--r--  1 deployer admin   11 Mar 11 20:52 .ruby-version
drwxr-xr-x  8 deployer admin 4.0K Mar 11 20:52 test
drwxr-xr-x  6 deployer admin 4.0K Mar 12 20:53 tmp
drwxr-xr-x  3 deployer admin 4.0K Mar 11 20:52 vendor
  • Look at the error in the log. It says Failed at step CHDIR spawning /home/deployer/.rvm/gems/ruby-2.2.1@my-app/bin/bundle: No such file or directory. That's the problem here. – Munir Mar 15 '16 at 22:05
  • I saw that error. The file exists. If I type in the linux terminal /home/deployer/.rvm/gems/ruby-2.2.1@my-app/bin/bundle exec "unicorn_rails -D -c /opt/www/my-app.com/config/unicorn.rb -E production" the unicorn will start but apparently it doesn't work in the .service file – Razvan El Mar 15 '16 at 22:38
  • Multi-posted to stackoverflow.com/questions/36014993 . help center – JdeBP Mar 22 '16 at 10:44
3

Error message Failed at step CHDIR means that systemd is unable to cd into directory specified in WorkingDirectory. It is specified as /opt/www/my-app.com but later you list it as /opt/www/my-app.

Another issue is that you are starting unicorn_rails with option -D which tells unicorn_rails process to daemonize (started process forks child and immediately exits) but your [Service] section doesn't specify Type so it defaults to simple and systemd expects process to stay alive. You need to remove -D option or specify Type=forking.

See example unit file for unicorn

0

It's propably because of letter case instead of eg /opt/www/my-app.com try /opt/www/My-app.com

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.