diff options
| author | Laurent Bercot <ska-skaware@skarnet.org> | 2014-12-05 22:26:11 +0000 |
|---|---|---|
| committer | Laurent Bercot <ska-skaware@skarnet.org> | 2014-12-05 22:26:11 +0000 |
| commit | 90b12bd71bb9fc79a4640b9112c13ef529d0196a (patch) | |
| tree | 523b3f4ee2969e7a729bab2ba749c4b924ae62af /doc/notifywhenup.html | |
| download | s6-90b12bd71bb9fc79a4640b9112c13ef529d0196a.tar.xz | |
Initial commit
Diffstat (limited to 'doc/notifywhenup.html')
| -rw-r--r-- | doc/notifywhenup.html | 75 |
1 files changed, 75 insertions, 0 deletions
diff --git a/doc/notifywhenup.html b/doc/notifywhenup.html new file mode 100644 index 0000000..40b0593 --- /dev/null +++ b/doc/notifywhenup.html @@ -0,0 +1,75 @@ +<html> + <head> + <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" /> + <meta http-equiv="Content-Language" content="en" /> + <title>s6: service startup notifications</title> + <meta name="Description" content="s6: service startup notifications" /> + <meta name="Keywords" content="s6 ftrig notification notifier writer libftrigw ftrigw startup U up svwait s6-svwait" /> + <!-- <link rel="stylesheet" type="text/css" href="http://skarnet.org/default.css" /> --> + </head> +<body> + +<p> +<a href="index.html">s6</a><br /> +<a href="http://skarnet.org/software/">Software</a><br /> +<a href="http://skarnet.org/">skarnet.org</a> +</p> + +<h1> Service startup notifications </h1> + +<p> + It is easy for a process supervision suite to know when a service that was <em>up</em> +is now <em>down</em>: the long-lived process implementing the service is dead. The +supervisor, running as the daemon's parent, is instantly notified via a SIGCHLD. +When it happens, <a href="s6-supervise.html">s6-supervise</a> sends a 'd' event +to its <tt>./event</tt> <a href="fifodir.html">fifodir</a>, so every subscriber +knows that the service is down. All is well. +</p> + +<p> + It is much trickier for a process supervision suite to know when a service +that was <em>down</em> is now <em>up</em>. The supervisor forks and execs the +daemon, and knows when the exec has succeeded; but after that point, it's all +up to the daemon itself. Some daemons do a lot of initialization work before +they're actually ready to serve, and it is impossible for the supervisor to +know exactly <em>when</em> the service is really ready. +<a href="s6-supervise.html">s6-supervise</a> sends a 'u' event to its +<tt>./event</tt> <a href="fifodir.html">fifodir</a> when it successfully +spawns the daemon, but any subscriber +reacting to 'u' is subject to a race condition - the service provided by the +daemon may not be ready yet. +</p> + +<p> + Reliable startup notifications need support from the daemons themselves. +Daemons should notify the outside world when the service they are providing +is reliably up - because only they know when it is the case. +</p> + +<p> + s6 provides two ways for daemons to perform startup notification. +</p> + +<ol> + <li> Daemons can use the <tt>ftrigw_notify()</tt> function, provided in +<a href="libftrigw.html">the ftrigw library</a>. This is extremely +simple and efficient, but requires specific s6 support in the daemon. </li> + <li> Daemons can write something to a file descriptor of their choice, +then close that file descriptor, when they're ready to serve. This is +a generic mechanism that some daemons already implement, and does not +require anything specific in the daemon's code. The administrator can +then run the daemon under <a href="s6-notifywhenup.html">s6-notifywhenup</a>, +which will properly catch the daemon's message and notify all the subscribers +with a 'U' event, meaning that the service is now up with no possible race +condition. </li> +</ol> + +<p> + The second method should really be implemented in every long-running +program providing a service. When it is not the case, it's impossible +to provide race-free startup notifications, and subscribers should be +content with the unreliable 'u' events provided by s6-supervise. +</p> + +</body> +</html> |