1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
|
<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: scan directories</title>
<meta name="Description" content="s6: scan directory" />
<meta name="Keywords" content="s6 scandir supervision svscan s6-svscan scan directory servicedir" />
<!-- <link rel="stylesheet" type="text/css" href="//skarnet.org/default.css" /> -->
</head>
<body>
<p>
<a href="index.html">s6</a><br />
<a href="//skarnet.org/software/">Software</a><br />
<a href="//skarnet.org/">skarnet.org</a>
</p>
<h1> Scan directories </h1>
<p>
A <em>scan directory</em> is a directory containing a list of
<a href="servicedir.html">service directories</a>, or symbolic links
pointing to service directories.
</p>
<p>
A scan directory represents a list of services that are supposed to
be supervised. Running <a href="s6-svscan.html">s6-svscan</a> on this
scan directory launches a <em>supervision tree</em>: every service
listed in the scan directory will be supervised.
</p>
<p>
There is normally only one scan directory per system, although nothing
prevents a system administrator from having more.
<a href="https://cr.yp.to/daemontools.html">daemontools</a> traditionally
uses <tt>/service</tt>, and <a href="http://smarden.org/runit/">runit</a>
traditionally uses <tt>/etc/service</tt>. s6 does not care where your
scan directory is, but I would advise <tt>/service</tt> for compatibility
with daemontools. Depending on your installation, <tt>/service</tt> could
be a symbolic link and point to a place either in a RAM filesystem or in
<tt>/var</tt>.
</p>
<a name="where">
<h2> Where and how to build a scan directory </h2>
</a>
<p>
Opinions and practices differ.
</p>
<p>
It is generally accepted that the place where you store all your
service directories (your "service repository") should <em>not</em> be
used as a scan directory - for a simple reason: you might want to have
service directories for more services than what you want to start at
any given time. In other words, your scan directory will be a <em>subset</em>
of your service repository, so you cannot just run s6-svscan on every
service you have a service directory for. So, the first thing is to
separate your <em>service repository</em>, which is just a storage place
for all the services you might want to manage someday, and your <em>scan
directory</em>, which is a directory representing all the services that
you are <em>currently</em> managing.
</p>
<h3> Service repository </h3>
<p>
Where to store your service repository is purely a matter of personal
preference. You just have to be aware that <a href="s6-supervise.html">
s6-supervise</a> needs writable <tt>supervise</tt> and <tt>event</tt>
subdirectories in a service directory it monitors.
</p>
<h3> Scan directory </h3>
<p>
Where and how to build your scan directory depends heavily on your boot
system - and on your personal preference too.
</p>
<p>
Standard <a href="https://cr.yp.to/daemontools.html">daemontools</a> and
<a href="http://smarden.org/runit/">runit</a> installations like to have
a fixed scan directory containing symlinks to service directories located
in the service repository. In other words, the service repository contains
the real <em>working copies</em> of the service directories. This works,
as long as:
</p>
<ul>
<li> It is possible to create writable <tt>supervise</tt> and <tt>event</tt>
subdirectories in every managed service directory. This can be achieved for
instance via symlinks, or by having the service repository stored on a writable
filesystem. </li>
<li> The scan program (<a href="s6-svscan.html">s6-svscan</a>,
<a href="https://cr.yp.to/daemontools/svscan.html">svscan</a>,
<a href="http://smarden.org/runit/runsvdir.8.html">runsvdir</a>...) is
started late enough for all the necessary filesystems to be mounted. </li>
</ul>
<p>
My own recommendation would be to have working copies of the service
directories <em>entirely separate</em> from the service repository. The
service repository can be safely stored on the root filesystem, and the
needed directories copied to a RAM filesystem at boot time. The scan
directory can be either the place where the working copies are written,
or another directory containing symlinks to those working copies. (The
latter is useful if you are using the <tt>-t</tt> option to
<a href="s6-svscan.html">s6-svscan</a> with a nonzero argument: copying a
directory is not atomic, but making a symlink is, so there is no risk
of your scanner finding a directory while it is being copied - which
could result in <a href="s6-supervise.html">s6-supervise</a> getting
the wrong information and not managing the service properly.)
</p>
<p>
An example:
</p>
<ul>
<li> Have your service repository in <tt>/img/services</tt>, i.e. have
service directories in <tt>/img/services/ftpd</tt>, <tt>/img/services/httpd</tt>,
<tt>/img/services/sshd</tt>, etc. </li>
<li> When booting, make <tt>/tmp</tt> a RAM filesystem, and create the
directories <tt>/tmp/services</tt> and <tt>/tmp/service</tt>. </li>
<li> Have s6-svscan run on <tt>/tmp/service</tt>, as early as possible in your
boot sequence. This is possible whether you want to run s6-svscan
<a href="s6-svscan-1.html">as process 1</a> or <a href="s6-svscan-not-1.html">not</a>. </li>
<li> During the boot sequence, populate <tt>/tmp/services</tt> with copies of the
service directories you need: for instance,
<ul>
<li> <tt>cp -a /img/services/sshd /tmp/services/sshd</tt> </li>
<li> <tt>cp -a /img/services/ftpd /tmp/services/ftpd</tt> </li>
<li> etc. </li>
</ul> </li>
<li> When you are ready to start a service, make a symlink in the
<tt>/tmp/service</tt> <em>scan directory</em> pointing to the working copy of
the service directory you need in <tt>/tmp/services</tt>, then notify s6-svscan.
For instance, to start ftpd and httpd together:
<pre> ln -s ../services/ftpd /tmp/service
ln -s ../services/httpd /tmp/service
s6-svscanctl -a /tmp/service</pre> </li>
</ul>
</body>
</html>
|
