- Add timeout-finish support and "down-readiness"

- LOTS of refactoring to make this work - Remove s6-notifywhenup - s6-supervise now rocks the casbah - rc for 2.2.0.0
author: Laurent Bercot <ska-skaware@skarnet.org> 2015-07-20 20:20:54 +0000
committer: Laurent Bercot <ska-skaware@skarnet.org> 2015-07-20 20:20:54 +0000
commit: a3cdeecf0033919e3b5a79c17c19b5ac98719256 (patch)
tree: 92a930930f18a4f8ae897b1a69c39358137ce1e2 /doc
parent: bd34de9054cec794d96b0fde1eee9100e1d34215 (diff)
download: s6-a3cdeecf0033919e3b5a79c17c19b5ac98719256.tar.xz
12 files changed, 92 insertions, 150 deletions
diff --git a/doc/index.html b/doc/index.html
index 87fe66b..0c23ff0 100644
--- a/doc/index.html
+++ b/doc/index.html
@@ -84,7 +84,7 @@ with s6</a> </li>
  <li> GNU make, version 4.0 or later. Please be aware that s6 will not build
 with an earlier version. </li>
  <li> <a href="http://skarnet.org/software/skalibs/">skalibs</a> version
-2.3.5.1 or later. It's a build-time requirement. It's also a run-time
+2.3.6.0 or later. It's a build-time requirement. It's also a run-time
 requirement if you link against the shared version of the skalibs
 library. </li>
  <li> <a href="http://skarnet.org/software/execline/">execline</a> version
@@ -101,7 +101,7 @@ library. </li>
 <h3> Download </h3>
 
 <ul>
- <li> The current released version of s6 is <a href="s6-2.1.6.0.tar.gz">2.1.6.0</a>. </li>
+ <li> The current released version of s6 is <a href="s6-2.2.0.0.tar.gz">2.2.0.0</a>. </li>
  <li> Alternatively, you can checkout a copy of the s6 git repository:
 <pre> git clone git://git.skarnet.org/s6 </pre> </li>
  <li> There's also a
@@ -153,8 +153,6 @@ a user interface to control those processes and monitor service states.
 <li><a href="s6-svwait.html">The <tt>s6-svwait</tt> program</a></li>
 <li><a href="s6-svlisten1.html">The <tt>s6-svlisten1</tt> program</a></li>
 <li><a href="s6-svlisten.html">The <tt>s6-svlisten</tt> program</a></li>
-<li><a href="s6-notifywhenup.html">The <tt>s6-notifywhenup</tt> program</a>
-<strong>(deprecated)</strong></li>
 </ul>
 
 <h4> Daemontools-like utilities </h4>
diff --git a/doc/notifywhenup.html b/doc/notifywhenup.html
index 9a288ac..6bc8b99 100644
--- a/doc/notifywhenup.html
+++ b/doc/notifywhenup.html
@@ -67,16 +67,16 @@ a generic mechanism that some daemons already implement.
 <a href="servicedir.html">service directory</a> for the daemon contains
 a valid <tt>notification-fd</tt> file, the daemon's supervisor, i.e. the
 <a href="s6-supervise.html">s6-supervise</a> program, will properly catch
-the daemon's message, update a state file (<tt>supervise/ready</tt>), then
+the daemon's message, update the status file (<tt>supervise/status</tt>),
 then notify all the subscribers
-with a 'U' event, meaning that the service is now up and ready.
+with a <tt>'U'</tt> event, meaning that the service is now up and ready.
 </p>
 
 <p>
  This method should really be implemented in every long-running
 program providing a service. When it is not the case, it's impossible
 to provide reliable startup notifications, and subscribers should then
-be content with the unreliable 'u' events provided by s6-supervise.
+be content with the unreliable <tt>'u'</tt> events provided by s6-supervise.
 </p>
 
 </body>
diff --git a/doc/s6-notifywhenup.html b/doc/s6-notifywhenup.html
deleted file mode 100644
index cdfa693..0000000
--- a/doc/s6-notifywhenup.html
+++ /dev/null
@@ -1,111 +0,0 @@
-<html>
-  <head>
-    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
-    <meta http-equiv="Content-Language" content="en" />
-    <title>s6: the s6-notifywhenup program</title>
-    <meta name="Description" content="s6: the s6-notifywhenup program" />
-    <meta name="Keywords" content="s6 command s6-notifywhenup fifodir notification event notifier send service daemon ready" />
-    <!-- <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> The s6-notifywhenup program </h1>
-
-<em>
-<p>
-Starting with s6-2.1.4.0, the s6-notifywhenup program has been deprecated,
-because there was still a case (albeit extremely rare) where a race
-condition would occur and readiness would be improperly advertised.
-Readiness notification for a service can now be achieved via a
-<tt>notification-fd</tt> file in the
-<a href="servicedir.html">service directory</a>, containing the
-number of the descriptor the service will write its readiness
-notification newline to. The notification will directly be picked by
-<a href="s6-supervise.html">s6-supervise</a>. <br />
-</p>
-
-<p>
- Quick upgrade recipe: for every service using s6-notifywhenup,
-replace the s6-notifywhenup invocation in your run script with
-<tt>fdmove 1 3</tt>, then perform <tt>echo 3 &gt; notification-fd</tt>.
-Done!
-</p>
-</em>
-
-<p>
-s6-notifywhenup launches a daemon while listening to a file descriptor,
-and sends a 'U' event to a <a href="fifodir.html">fifodir</a> when it
-receives something on that file descriptor.
-</p>
-
-<p>
-<a href="notifywhenup.html">This page</a> explains why this program is
-needed.
-</p>
-
-<h2> Interface </h2>
-
-<pre>
-     s6-notifywhenup [ -d <em>fd</em> ] [ -e <em>fifodir</em> ] [ -f ] [ -X ] [ -t <em>timeout</em> ] <em>prog...</em>
-</pre>
-
-<ul>
- <li> s6-notifywhenup forks and executes <em>prog...</em> as the
-parent, with a pipe from <em>prog...</em>'s stdout to the child. </li>
- <li> The child waits for a newline (<tt>\n</tt>) to be written
-on the pipe. When it gets it, it creates a
-<tt>./supervise/ready</tt> file then sends a 'U' event to the
-<tt>./event</tt> fifodir. </li>
- <li> The child exits 0. </li>
-</ul>
-
-<h2> Options </h2>
-
-<ul>
- <li> <tt>-d&nbsp;<em>fd</em></tt>&nbsp;: listen to
-<em>prog</em>'s output on descriptor <em>fd</em>. The default is 1. </li>
- <li> <tt>-e&nbsp;<em>fifodir</em></tt>&nbsp;: send a 'U' event to fifodir
-<em>fifodir</em>. Default is <tt>./event</tt>. </li>
- <li> <tt>-f</tt>&nbsp;: simple fork. Normally, s6-notifywhenup doubleforks,
-in order to accommodate for a
-<em>prog</em> that does not expect to have a child and avoid a
-pending zombie. This option avoids the doublefork, but it should only be
-set if <em>prog</em> reaps even children it doesn't know it has. </li>
- <li> <tt>-t&nbsp;<em>timeout</em></tt>&nbsp;: if no EOF has been received
-after <em>timeout</em> milliseconds, exit without sending the event.
-Default is 0, meaning infinite. </li>
- <li> <tt>-X</tt>&nbsp;: fake readiness. s6-notifywhenup will actually send
-the newline itself before executing <em>prog</em>. This option should in
-principle never be used. </li>
-</ul>
-
-<h2> Notes </h2>
-
-<ul>
- <li> s6-notifywhenup executes <em>prog...</em> as the parent in order
-for <em>prog</em> to keep the same pid, which is vital for supervised
-processes. </li>
- <li> s6-notifywhenup can be used, for instance, with
-<a href="s6-ipcserver.html">s6-ipcserver</a>
-and its <tt>-1</tt> option, so that reliable startup notification is
-achieved. <tt>s6-notifywhenup -f s6-ipcserver -1 <em>args</em></tt> will
-send a 'U' event to <tt>./event</tt> when s6-ipcserver is actually
-listening to its socket. </li>
- <li> The <a href="s6-svwait.html">s6-svwait</a> program can be used
-to wait for 'U' events, as well as the
-<a href="s6-svlisten1.html">s6-svlisten1</a> and
-<a href="s6-svlisten.html">s6-svlisten</a> programs. </li>
- <li> The <tt>supervise/ready</tt> file, when created, contains at least
-the absolute time when s6-notifywhenup detected service readiness. The
-format and contents of this file are subject to change. </li>
-</ul>
-
-</body>
-</html>
diff --git a/doc/s6-sudod.html b/doc/s6-sudod.html
index 2c613da..54e9574 100644
--- a/doc/s6-sudod.html
+++ b/doc/s6-sudod.html
@@ -91,8 +91,8 @@ will set up a privileged program:
 <pre>
 #!/command/execlineb -P
 fdmove -c 2 1
+fdmove 1 3
 s6-envuidgid serveruser
-s6-notifywhenup -f
 s6-ipcserver -U -1 -- serversocket
 s6-ipcserver-access -v2 -l0 -i rules --
 exec -c
@@ -106,12 +106,15 @@ sargv
 executes the script. </li>
  <li> <a href="http://skarnet.org/software/execline/fdmove.html">fdmove</a> makes
 sure the script's error messages are sent to the service's logger. </li>
+ <li> <a href="http://skarnet.org/software/execline/fdmove.html">fdmove</a>
+redirects the script's stdout to file descriptor 3. This is useful if
+the service directory contains a <tt>notification-fd</tt> file containing
+<tt>3</tt>, so the daemon can perform
+<a href="notifywhenup.html">readiness notification</a> by writing a
+newline to its stdout. (The
+<tt>-1</tt> option to s6-ipcserver tells it to do this.) </li>
  <li> <a href="s6-envuidgid.html">s6-envuidgid</a>
 sets the UID, GID and GIDLIST environment variables for s6-ipcserver to interpret. </li>
- <li> <a href="s6-notifywhenup.html">s6-notifywhenup</a> primes the
-service for readiness notification (and the
-<tt>-1</tt> option to s6-ipcserver tells the daemon to actually
-notify when it's ready). </li>
  <li> <a href="s6-ipcserver.html">s6-ipcserver</a> binds to <em>serversocket</em>,
 drops its privileges to those of <em>serveruser</em>, and announces its
 readiness. Then, for every client connecting to <em>serversocket</em>:
diff --git a/doc/s6-supervise.html b/doc/s6-supervise.html
index a16e0f5..081aeae 100644
--- a/doc/s6-supervise.html
+++ b/doc/s6-supervise.html
@@ -44,15 +44,18 @@ If it already exists, it uses it as is, without modifying the subscription right
  <li> If the default service state is up, s6-supervise spawns <tt>./run</tt>. </li>
  <li> s6-supervise sends a <tt>'u'</tt> event to <tt>./event</tt> whenever it
 successfully spawns <tt>./run</tt>. </li>
- <li> When <tt>./run</tt> dies, s6-supervise sends a <tt>'d'</tt> event to <tt>./event</tt>. </li>
- <li> When <tt>./run</tt> dies, s6-supervise spawns <tt>./finish</tt> if it exists.
+ <li> When <tt>./run</tt> dies, s6-supervise sends a <tt>'d'</tt> event to <tt>./event</tt>.
+It then spawns <tt>./finish</tt> if it exists.
 <tt>./finish</tt> will have <tt>./run</tt>'s exit code as first argument, or 256 if
 <tt>./run</tt> was signaled; it will have the number of the signal that killed <tt>./run</tt>
 as second argument, or an undefined number if <tt>./run</tt> was not signaled. </li>
- <li> <tt>./finish</tt> must exit in less than 5 seconds. If it takes more than that,
-s6-supervise kills it with a SIGKILL. </li>
- <li> When <tt>./finish</tt> dies, s6-supervise restarts <tt>./run</tt> unless it has been
-told not to. </li>
+ <li> By default, <tt>./finish</tt> must exit in less than 5 seconds. If it takes more than that,
+s6-supervise kills it with a SIGKILL. This can be configured via the
+<tt>./timeout-finish</tt> file, see the description in the
+<a href="servicedir.html">service directory page</a>. </li>
+ <li> When <tt>./finish</tt> dies (or is killed),
+s6-supervise sends a <tt>'D'</tt> event to <tt>./event</tt>. Then
+it restarts <tt>./run</tt> unless it has been told not to. </li>
  <li> There is a minimum 1-second delay between two <tt>./run</tt> spawns, to avoid busylooping
 if <tt>./run</tt> exits too quickly. </li>
  <li> When killed or asked to exit, it waits for the service to go down one last time, then
@@ -78,10 +81,9 @@ what they do, are listed on the
 <tt>notification-fd</tt> file when the service is started, or restarted,
 s6-supervise creates and listens to an additional pipe from the service
 for <a href="notifywhenup.html">readiness notification</a>. When the
-notification occurs, s6-supervise creates a <tt>./supervise/ready</tt>
-file containing the absolute time when readiness occurred, then sends
-a <tt>'U'</tt> event to <tt>./event</tt>. The <tt>./supervise/ready</tt>
-file is deleted on service death.
+notification occurs, s6-supervise updates the <tt>./supervise/status</tt>
+file accordingly, then sends
+a <tt>'U'</tt> event to <tt>./event</tt>.
 </p>
 
 <p>
diff --git a/doc/s6-svc.html b/doc/s6-svc.html
index 2de8ba4..f21847a 100644
--- a/doc/s6-svc.html
+++ b/doc/s6-svc.html
@@ -28,7 +28,7 @@ knowing their PIDs, and without using horrible hacks such as .pid files.
 <h2> Interface </h2>
 
 <pre>
-     s6-svc [ -D | -U ] [ -T <em>timeout</em> ] [ -abqhkti12pcoduxO ] <em>servicedir</em>
+     s6-svc [ -wu | -wU | -wd | -wD ] [ -T <em>timeout</em> ] [ -abqhkti12pcoduxO ] <em>servicedir</em>
 </pre>
 
 <p>
@@ -75,13 +75,20 @@ it. </li>
 (in milliseconds) after which s6-svc will exit 1 with an error message if
 the service still hasn't reached the desired state. By default, the
 timeout is 0, which means that s6-svc will block indefinitely. </li>
- <li> <tt>-D</tt>&nbsp;: s6-svc will not exit until the service is down. </li>
- <li> <tt>-U</tt>&nbsp;: s6-svc will not exit until the service is up and
+ <li> <tt>-wd</tt>&nbsp;: s6-svc will not exit until the service is down,
+i.e. until the <tt>run</tt> process has died. </li>
+ <li> <tt>-wD</tt>&nbsp;: s6-svc will not exit until the service is down
+<em>and</em> ready to be brought up, i.e. a possible <tt>finish</tt> script has
+exited. </li>
+ <li> <tt>-wu</tt>&nbsp;: s6-svc will not exit until the service is up,
+i.e. there is a process running the <tt>run</tt> executable. </li>
+ <li> <tt>-wU</tt>&nbsp;: s6-svc will not exit until the service is up <em>and</em>
 <a href="notifywhenup.html">ready</a> as notified by the daemon itself.
 If the <a href="servicedir.html">service directory</a> does not contain
 a <tt>notification-fd</tt> file to tell
 <a href="s6-supervise.html">s6-supervise</a> to accept readiness
-notification, s6-svc will print a warning and ignore the command. </li>
+notification, s6-svc will print a warning and act as if the <tt>-wu</tt>
+option had been given instead. </li>
 </ul>
 
 <h2> Usage examples </h2>
@@ -100,12 +107,13 @@ the process represented by the <tt>/service/sshd</tt> service directory -
 typically the sshd server.
 </p>
 
-<pre> s6-svc -Dd /service/ftpd </pre>
+<pre> s6-svc -wD -d /service/ftpd </pre>
 <p>
- Take down the ftpd server and block until the process is really down.
+ Take down the ftpd server and block until the process is down and
+the finish script has completed.
 </p>
 
-<pre> s6-svc -Uu -T 5000 /service/ftpd </pre>
+<pre> s6-svc -wU -T 5000 -u /service/ftpd </pre>
 <p>
  Bring up the ftpd server and block until it has sent notification that it
 is ready. Exit 1 if it is still not ready after 5 seconds.
@@ -123,10 +131,10 @@ process is <a href="s6-log.html">s6-log</a>, this triggers a log rotation.
  <li> s6-svc writes control commands into the <tt><em>servicedir</em>/supervise/control</tt>
 FIFO. A s6-supervise process running on <em>servicedir</em> will be listening to this FIFO,
 and will read and interpret those commands. </li>
- <li> When invoked with the <tt>-D</tt> or <tt>-U</tt> option, s6-svc executes into
+ <li> When invoked with one of the <tt>-w</tt> options, s6-svc executes into
 <a href="s6-svlisten1.html">s6-svlisten1</a>, which will listen to service state
-changes and spawn another s6-svc instance (without the <tt>-D</tt> or <tt>-U</tt>
-option) that will send the commands to the service. Any error message written during
+changes and spawn another s6-svc instance (without the <tt>-w</tt> option)
+that will send the commands to the service. Any error message written during
 the waiting period will mention it is being written by s6-svlisten1; this is normal. </li>
 </ul>
 
diff --git a/doc/s6-svlisten.html b/doc/s6-svlisten.html
index 2316a2b..1db24c0 100644
--- a/doc/s6-svlisten.html
+++ b/doc/s6-svlisten.html
@@ -34,7 +34,7 @@ a collection of supervised services, and blocks until they all go up, or down.
 </p>
 
 <pre>
-     s6-svlisten [ -U | -u | -d ] [ -a | -o ] [ -t <em>timeout</em> ] { <em>servicedir</em> <em>servicedir...</em> } <em>prog...</em>
+     s6-svlisten [ -U | -u | -D | -d ] [ -a | -o ] [ -t <em>timeout</em> ] { <em>servicedir</em> <em>servicedir...</em> } <em>prog...</em>
 </pre>
 
 <p>
@@ -42,7 +42,7 @@ a collection of supervised services, and blocks until they all go up, or down.
 </p>
 
 <pre>
-     s6-svlisten [ -U | -u | -d ] [ -a | -o ] [ -t <em>timeout</em> ] <em>servicedir</em> <em>servicedir...</em> "" <em>prog...</em>
+     s6-svlisten [ -U | -u | -D | -d ] [ -a | -o ] [ -t <em>timeout</em> ] <em>servicedir</em> <em>servicedir...</em> "" <em>prog...</em>
 </pre>
 
 <ul>
@@ -69,6 +69,10 @@ specific support in the service programs, and the use of the
 <a href="servicedir.html">service directory</a>.
 See the explanation on <a href="notifywhenup.html">this page</a>. </li>
  <li> <tt>-d</tt>&nbsp;: down. s6-svlisten will wait until the services are down. </li>
+ <li> <tt>-D</tt>&nbsp;: really down. s6-svlisten will wait until the
+services are down and the cleanup scripts in <tt><em>servicedir</em>/finish</tt>
+for every <em>servicedir</em>
+have finished executing (or have timed out and been killed). </li>
  <li> <tt>-o</tt>&nbsp;: or. s6-svlisten will wait until <em>one</em> of the
 given services comes up or down. </li>
  <li> <tt>-a</tt>&nbsp;: and. s6-svlisten will wait until <em>all</em> of the
diff --git a/doc/s6-svlisten1.html b/doc/s6-svlisten1.html
index a9e147a..c3f5ef4 100644
--- a/doc/s6-svlisten1.html
+++ b/doc/s6-svlisten1.html
@@ -30,7 +30,7 @@ supervised service, and blocks until said service goes up, or down.
 <h2> Interface </h2>
 
 <pre>
-     s6-svlisten [ -U | -u | -d ] [ -t <em>timeout</em> ] <em>servicedir</em> <em>prog...</em>
+     s6-svlisten [ -U | -u | -D | -d ] [ -t <em>timeout</em> ] <em>servicedir</em> <em>prog...</em>
 </pre>
 
 <ul>
@@ -57,6 +57,9 @@ specific support in the service programs, and the use of the
 <a href="servicedir.html">service directory</a>.
 See the explanation on <a href="notifywhenup.html">this page</a>. </li>
  <li> <tt>-d</tt>&nbsp;: down. s6-svlisten1 will wait until the service is down. </li>
+ <li> <tt>-D</tt>&nbsp;: really down. s6-svlisten1 will wait until the
+service is down and the cleanup script in <tt><em>servicedir</em>/finish</tt>
+has finished executing (or has timed out and been killed). </li>
  <li> <tt>-t&nbsp;<em>timeout</em></tt>&nbsp;: if the requested event has not
 happened after <em>timeout</em> milliseconds, s6-svlisten1 will print a message
 to stderr and exit 1. By default, <em>timeout</em> is 0, which means no time
diff --git a/doc/s6-svstat.html b/doc/s6-svstat.html
index e78c13f..5feef0f 100644
--- a/doc/s6-svstat.html
+++ b/doc/s6-svstat.html
@@ -46,7 +46,11 @@ signal, if it is down </li>
 kernel's scheduler picks up s6-supervise </li>
  <li> whether the service is <a href="notifywhenup.html">ready</a>,
 as notified by the daemon itself, and
-if it is, the number of seconds that it has been. </li>
+if it is, the number of seconds that it has been.
+A service reported as down and ready simply means that it is ready
+to be brought up. A service is down and not ready when it is in the
+cleanup phase, i.e. the <tt>./finish</tt> script is still being executed. </li>
+ </li>
 </ul>
 
 <h2> Options </h2>
diff --git a/doc/s6-svwait.html b/doc/s6-svwait.html
index de209ea..c3df0fa 100644
--- a/doc/s6-svwait.html
+++ b/doc/s6-svwait.html
@@ -29,7 +29,7 @@ s6-svwait only waits for notifications; it never polls.
 <h2> Interface </h2>
 
 <pre>
-     s6-svwait [ -U | -u | -d ] [ -a | -o ] [ -t <em>timeout</em> ] <em>servicedir...</em>
+     s6-svwait [ -U | -u | -D | -d ] [ -a | -o ] [ -t <em>timeout</em> ] <em>servicedir...</em>
 </pre>
 
 <p>
@@ -53,6 +53,11 @@ specific support in the service programs, and the use of the
 <a href="servicedir.html">service directory</a>.
 See the explanation on <a href="notifywhenup.html">this page</a>. </li>
  <li> <tt>-d</tt>&nbsp;: down. s6-svwait will wait until the services are down. </li>
+ <li> <tt>-D</tt>&nbsp;: really down. s6-svwait will wait until the
+services are down and the cleanup scripts in
+<tt><em>servicedir</em>/finish</tt>
+for every <em>servicedir</em>
+have finished executing (or have timed out and been killed). </li>
  <li> <tt>-o</tt>&nbsp;: or. s6-svwait will wait until <em>one</em> of the
 given services comes up or down. </li>
  <li> <tt>-a</tt>&nbsp;: and. s6-svwait will wait until <em>all</em> of the
@@ -80,8 +85,7 @@ s6-supervise has not been started yet. </li>
 <p>
 s6-svwait spawns a <a href="libs6/s6-ftrigrd.html">s6-ftrigrd</a> child to
 listen to notifications sent by <a href="s6-supervise.html">s6-supervise</a>.
-It also checks <tt>supervise/status</tt> files, as well as the
-<tt>supervise/ready</tt> files if necessary, to get the current service
+It also checks <tt>supervise/status</tt> files to get the current service
 states, so it is immune to race conditions.
 </p>
 
diff --git a/doc/servicedir.html b/doc/servicedir.html
index 7cebd53..76a0a11 100644
--- a/doc/servicedir.html
+++ b/doc/servicedir.html
@@ -117,6 +117,12 @@ notification from the service and broadcast readiness, i.e. any
 <a href="s6-svlisten1.html">s6-svlisten1 -U</a> or
 <a href="s6-svlisten.html">s6-svlisten -U</a> processes will be
 triggered. </li>
+ <li> An optional regular file named <tt>timeout-finish</tt>. If such a file
+exists, it must only contain an unsigned integer, which is the number of
+milliseconds after which the <tt>./finish</tt> script, if it exists, will
+be killed with a SIGKILL. The default is 5000: finish scripts are killed
+if they're still alive after 5 seconds. A value of 0 allows finish scripts
+to run forever. </li>
  <li> A <a href="fifodir.html">fifodir</a> named <tt>event</tt>. It is automatically
 created by <a href="s6-supervise.html">s6-supervise</a> if it does not exist.
 <em>foo</em><tt>/event</tt>
diff --git a/doc/upgrade.html b/doc/upgrade.html
index b6ae2c9..c77992b 100644
--- a/doc/upgrade.html
+++ b/doc/upgrade.html
@@ -18,6 +18,27 @@
 
 <h1> What has changed in s6 </h1>
 
+<h2> in 2.2.0.0 </h2>
+
+<ul>
+ <li> skalibs dependency bumped to 2.3.6.0. </li>
+ <li> The internals of <a href="s6-supervise.html">s6-supervise</a> have
+changed; the <tt>supervise/status</tt> file ABI has changed and is not
+compatible with the daemontools/runit <tt>supervise/status</tt> files
+anymore. (This should not impact anything.) </li>
+ <li> <a href="s6-supervise.html">s6-supervise</a> features a
+configurable timeout for <tt>./finish</tt> scripts, via the
+<a href="servicedir.html"><tt>./timeout-finish</tt></a> file. </li>
+ <li> <a href="s6-svwait.html">s6-svwait</a>,
+<a href="s6-svlisten1.html">s6-svlisten1</a> and
+<a href="s6-svlisten.html">s6-svlisten</a> can now wait for a <tt>'D'</tt>
+event, which means the <tt>./finish</tt> script has terminated. </li>
+ <li> The deprecated <tt>s6-notifywhenup</tt> program has been
+removed. </li>
+ <li> The syntax for synchronous <a href="s6-svc.html">s6-svc</a>
+waiting has changed. </li>
+</ul>
+
 <h2> in 2.1.6.0 </h2>
 
 <ul>
author	Laurent Bercot <ska-skaware@skarnet.org>	2015-07-20 20:20:54 +0000
committer	Laurent Bercot <ska-skaware@skarnet.org>	2015-07-20 20:20:54 +0000
commit	a3cdeecf0033919e3b5a79c17c19b5ac98719256 (patch)
tree	92a930930f18a4f8ae897b1a69c39358137ce1e2 /doc
parent	bd34de9054cec794d96b0fde1eee9100e1d34215 (diff)
download	s6-a3cdeecf0033919e3b5a79c17c19b5ac98719256.tar.xz