nginx can be controlled with signals. The process ID of the master process is written to the file /usr/local/nginx/logs/nginx.pid by default. This name may be changed at configuration time, or in nginx.conf using the pid directive. The master process supports the following signals:

TERM, INT	fast shutdown
QUIT	graceful shutdown
HUP	changing configuration, keeping up with a changed time zone (only for FreeBSD and Linux), starting new worker processes with a new configuration, graceful shutdown of old worker processes
USR1	re-opening log files
USR2	upgrading an executable file
WINCH	graceful shutdown of worker processes

Individual worker processes can be controlled with signals as well, though it is not required. The supported signals are:

TERM, INT	fast shutdown
QUIT	graceful shutdown
USR1	re-opening log files
WINCH	abnormal termination for debugging (requires debug_points to be enabled)

In order for nginx to re-read the configuration file, a HUP signal should be sent to the master process. The master process first checks the syntax validity, then tries to apply new configuration, that is, to open log files and new listen sockets. If this fails, it rolls back changes and continues to work with old configuration. If this succeeds, it starts new worker processes, and sends messages to old worker processes requesting them to shut down gracefully. Old worker processes close listen sockets and continue to service old clients. After all clients are serviced, old worker processes are shut down.

Let's illustrate this by example. Imagine that nginx is run on FreeBSD 4.x and the command

ps ax -o pid,ppid,user,%cpu,vsz,wchan,command | egrep '(nginx|PID)'

produces the following output:

  PID  PPID USER    %CPU   VSZ WCHAN  COMMAND
33126     1 root     0.0  1148 pause  nginx: master process /usr/local/nginx/sb
33127 33126 nobody   0.0  1380 kqread nginx: worker process (nginx)
33128 33126 nobody   0.0  1364 kqread nginx: worker process (nginx)
33129 33126 nobody   0.0  1364 kqread nginx: worker process (nginx)

If HUP is sent to the master process, the output becomes:

  PID  PPID USER    %CPU   VSZ WCHAN  COMMAND
33126     1 root     0.0  1164 pause  nginx: master process /usr/local/nginx/sb
33129 33126 nobody   0.0  1380 kqread nginx: worker process is shutting down (n
33134 33126 nobody   0.0  1368 kqread nginx: worker process (nginx)
33135 33126 nobody   0.0  1368 kqread nginx: worker process (nginx)
33136 33126 nobody   0.0  1368 kqread nginx: worker process (nginx)

One of the old worker processes with PID 33129 still continues to work. After some time it exits:

  PID  PPID USER    %CPU   VSZ WCHAN  COMMAND
33126     1 root     0.0  1164 pause  nginx: master process /usr/local/nginx/sb
33134 33126 nobody   0.0  1368 kqread nginx: worker process (nginx)
33135 33126 nobody   0.0  1368 kqread nginx: worker process (nginx)
33136 33126 nobody   0.0  1368 kqread nginx: worker process (nginx)

In order to rotate log files, they need to be renamed first. After that USR1 signal should be sent to the master process. The master process will then re-open all currently open log files and assign them an unprivileged user under which the worker processes are running, as an owner. After successful re-opening, the master process closes all open files and sends the message to worker process to ask them to re-open files. Worker processes also open new files and close old files right away. As a result, old files are almost immediately available for post processing, such as compression.

In order to upgrade the server executable, the new executable file should be put in place of an old file first. After that USR2 signal should be sent to the master process. The master process first renames its file with the process ID to a new file with the .oldbin suffix, e.g. /usr/local/nginx/logs/nginx.pid.oldbin, then starts a new executable file that in turn starts new worker processes:

  PID  PPID USER    %CPU   VSZ WCHAN  COMMAND
33126     1 root     0.0  1164 pause  nginx: master process /usr/local/nginx/sb
33134 33126 nobody   0.0  1368 kqread nginx: worker process (nginx)
33135 33126 nobody   0.0  1380 kqread nginx: worker process (nginx)
33136 33126 nobody   0.0  1368 kqread nginx: worker process (nginx)
36264 33126 root     0.0  1148 pause  nginx: master process /usr/local/nginx/sb
36265 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)
36266 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)
36267 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)

After that all worker processes (old and new ones) continue to accept requests. If the WINCH signal is sent to the first master process, it will send messages to its worker processes, requesting them to shut down gracefully, and they will start to exit:

  PID  PPID USER    %CPU   VSZ WCHAN  COMMAND
33126     1 root     0.0  1164 pause  nginx: master process /usr/local/nginx/sb
33135 33126 nobody   0.0  1380 kqread nginx: worker process is shutting down (n
36264 33126 root     0.0  1148 pause  nginx: master process /usr/local/nginx/sb
36265 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)
36266 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)
36267 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)

When using the “rtsig” method on Linux, the new processes may not accept connections even after the old master process was sent the WINCH signal. If that is the case, the USR1 signal should be sent to the new master process continuously, until the new processes start to accept connections.

After some time, only the new worker processes will process requests:

  PID  PPID USER    %CPU   VSZ WCHAN  COMMAND
33126     1 root     0.0  1164 pause  nginx: master process /usr/local/nginx/sb
36264 33126 root     0.0  1148 pause  nginx: master process /usr/local/nginx/sb
36265 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)
36266 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)
36267 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)

It should be noted that the old master process does not close its listen sockets, and it can be managed to start its worker processes again if needed. If for some reason the new executable file works unacceptably, the following can be done:

• Send the HUP signal to the old master process. The old process will start new worker processes without re-reading the configuration. After that, new processes can be shut down gracefully, by sending their master process the QUIT signal.

• Send the TERM signal to the new master process, it will then send a message to its worker processes requesting them to exit immediately, and they will all exit almost immediately. When the new master process exits, the old master process will start new worker processes.

• If new processes do not exit, the KILL signal should be sent to them. When the new master process exits, the old master process will start new worker processes.

If the new master process exits then the old master process discards the .oldbin suffix from the file name with the process ID.

If upgrade was successful, then the old master process can be sent the QUIT signal, and only new processes will stay:

  PID  PPID USER    %CPU   VSZ WCHAN  COMMAND
36264     1 root     0.0  1148 pause  nginx: master process /usr/local/nginx/sb
36265 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)
36266 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)
36267 36264 nobody   0.0  1364 kqread nginx: worker process (nginx)