3 In general, signals need only be sent to the master process. However,
4 the signals unicorn uses internally to communicate with the worker
5 processes are documented here as well.
9 * HUP - reload config file and gracefully restart all workers
10 If "preload_app" is false (the default), the application code
11 will be reloaded when workers are restarted as well.
13 * INT/TERM - quick shutdown, kills all workers immediately
15 * QUIT - graceful shutdown, waits for workers to finish their
16 current request before finishing.
18 * USR1 - reopen all logs owned by the master and all workers
19 See Unicorn::Util.reopen_logs for what is considered a log.
21 * USR2 - reexecute the running binary. A separate QUIT
22 should be sent to the original process once the child is verified to
25 * WINCH - gracefully stops workers but keep the master running.
26 This will only work for daemonized processes.
28 * TTIN - increment the number of worker processes by one
30 * TTOU - decrement the number of worker processes by one
34 Sending signals directly to the worker processes should not normally be
35 needed. If the master process is running, any exited worker will be
36 automatically respawned.
38 * INT/TERM - Quick shutdown, immediately exit.
39 Unless WINCH has been sent to the master (or the master is killed),
40 the master process will respawn a worker to replace this one.
42 * QUIT - Gracefully exit after finishing the current request.
43 Unless WINCH has been sent to the master (or the master is killed),
44 the master process will respawn a worker to replace this one.
46 * USR1 - Reopen all logs owned by the worker process.
47 See Unicorn::Util.reopen_logs for what is considered a log.
48 Log files are not reopened until it is done processing
49 the current request, so multiple log lines for one request
50 (as done by Rails) will not be split across multiple logs.
52 === Procedure to replace a running unicorn executable
54 You may replace a running instance of unicorn with a new one without
55 losing any incoming connections. Doing so will reload all of your
56 application code, Unicorn config, Ruby executable, and all libraries.
57 The only things that will not change (due to OS limitations) are:
59 1. The listener backlog size of already-bound sockets
61 2. The path to the unicorn executable script. If you want to change to
62 a different installation of Ruby, you can modify the shebang
63 line to point to your alternative interpreter.
65 The procedure is exactly like that of nginx:
67 1. Send USR2 to the master process
69 2. Check your process manager or pid files to see if a new master spawned
70 successfully. If you're using a pid file, the old process will have
71 ".oldbin" appended to its path. You should have two master instances
72 of unicorn running now, both of which will have workers servicing
73 requests. Your process tree should look something like this:
86 3. You can now send WINCH to the old master process so only the new workers
87 serve requests. If your unicorn process is bound to an interactive
88 terminal, you can skip this step. Step 5 will be more difficult but
89 you can also skip it if your process is not daemonized.
91 4. You should now ensure that everything is running correctly with the
92 new workers as the old workers die off.
94 5. If everything seems ok, then send QUIT to the old master. You're done!
96 If something is broken, then send HUP to the old master to reload
97 the config and restart its workers. Then send QUIT to the new master