aboutsummaryrefslogtreecommitdiffstats
path: root/doc/s6-dnsip4-filter.html
diff options
context:
space:
mode:
Diffstat (limited to 'doc/s6-dnsip4-filter.html')
-rw-r--r--doc/s6-dnsip4-filter.html111
1 files changed, 111 insertions, 0 deletions
diff --git a/doc/s6-dnsip4-filter.html b/doc/s6-dnsip4-filter.html
new file mode 100644
index 0000000..5433438
--- /dev/null
+++ b/doc/s6-dnsip4-filter.html
@@ -0,0 +1,111 @@
+<html>
+ <head>
+ <meta http-equiv="Content-Type" content="text/html; charset=UTF-8" />
+ <meta http-equiv="Content-Language" content="en" />
+ <title>s6-dns: the s6-dnsip4-filter program</title>
+ <meta name="Description" content="s6-dns: the s6-dnsip4-filter program" />
+ <meta name="Keywords" content="s6-dns client s6-dnsip4-filter dnsip domain name ip address ipv4 filter" />
+ <!-- <link rel="stylesheet" type="text/css" href="http://skarnet.org/default.css" /> -->
+ </head>
+<body>
+
+<p>
+<a href="index.html">s6-dns</a><br />
+<a href="http://skarnet.org/software/">Software</a><br />
+<a href="http://skarnet.org/">skarnet.org</a>
+</p>
+
+<h1> The s6-dnsip4-filter program </h1>
+
+<p>
+ s6-dnsip4-filter reads domain names on its standard input and
+prints the corresponding IPv4 addresses on its standard output.
+</p>
+
+<h2> Interface </h2>
+
+<pre>
+ s6-dnsip4-filter [ -l <em>maxlines</em> ] [ -c <em>maxconn</em> ] [ -t <em>timeout</em> ] [ -f <em>normalfmt</em> ] [ -e <em>errorfmt</em> ]
+</pre>
+
+<ul>
+ <li> s6-dnsip4-filter reads lines from its stdin, processes them, and prints
+the processed lines to its stdout, in the same order. It exits 0 when it
+reads EOF on stdin. </li>
+ <li> Every line starting with a domain name is processed. Lines not
+starting with a domain name are not, and the formatting rules treat them
+as a pure "remainder". </li>
+ <li> For every processed line, an A query is made, and the result is
+printed according to the formatting rules. The domain is not qualified
+before being resolved. </li>
+ <li> By default, s6-dnsip4-filter looks for DNS cache addresses in the
+<tt>/etc/resolv.conf</tt> file. If the DNSCACHEIP environment variable is set
+and contains a list of IP (v4 or v6) addresses, separated by commas,
+semicolons, spaces, tabs, newlines or carriage returns, then this list
+is used instead. </li>
+</ul>
+
+<h2> Options </h2>
+
+<ul>
+ <li> <tt>-l</tt>&nbsp;<em>maxlines</em>&nbsp;: maximum lines. s6-dnsip4-filter
+will keep at most <em>maxlines</em> lines in memory at the same time. If line
+<em>n</em> is still waiting for resolution and cannot be printed, then
+s6-dnsip4-filter will stop reading from stdin before line <em>n+maxlines</em>,
+until the result arrives for line <em>n</em>. By default, <em>maxlines</em>
+is 256. </li>
+ <li> <tt>-c</tt>&nbsp;<em>maxconn</em>&nbsp;: maximum concurrent resolutions.
+s6-dnsip4-filter will perform at most <em>maxconn</em> resolutions in
+parallel at the same time. <em>maxlines</em> cannot be lesser than <em>maxconn</em>.
+By default, <em>maxconn</em> is 128. </li>
+ <li> <tt>-t</tt>&nbsp;<em>timeout</em>&nbsp;: if any resolution takes more
+than <em>timeout</em> milliseconds, then it is aborted and printed as a
+timeout error. By default, <em>timeout</em> is 0, which means no timeout. </li>
+ <li> <tt>-f</tt>&nbsp;<em>normalfmt</em>&nbsp;: print the positive results according to
+the <em>normalfmt</em> format string. By default, <em>normalfmt</em> is
+<tt>%s=%d%w%r</tt>, which means: print the original domain name, then an equal
+sign, then the corresponding IPv4 address, then the remainder of the line. </li>
+ <li> <tt>-e</tt>&nbsp;<em>errorfmt</em>&nbsp;: print the negative results according to
+the <em>errorfmt</em> format string. By default, <em>errorfmt</em> is
+<tt>%s=&lt;%e%&gt;%w%r</tt>, which means: print the original domain name, then an equal
+sign, then the error message between angle brackets, then the remainder of the line. </li>
+</ul>
+
+<h2> Formatting rules </h2>
+
+<p>
+ <em>normalfmt</em> and <em>errorfmt</em> are format strings, i.e. they tell the
+program how a line must be printed. The following sequences are recognized:
+</p>
+
+<ul>
+ <li> <tt>%%</tt> prints a single <tt>%</tt> character </li>
+ <li> <tt>%s</tt> prints the original domain name </li>
+ <li> <tt>%d</tt> prints the IP address, if any </li>
+ <li> <tt>%e</tt> prints the error message, if any </li>
+ <li> <tt>%w%r</tt> prints the remainder of the input line </li>
+</ul>
+
+<h2> Example </h2>
+
+<pre>
+ s6-dnsns google.com | s6-dnsip4-filter -f "%d"
+</pre>
+
+<p>
+ prints all the nameserver addresses for the <tt>google.com</tt>
+domain. This is useful, for instance, to give the result as an
+argument to <a href="s6-dnsq.html">s6-dnsq</a>.
+</p>
+
+<h2> Notes </h2>
+
+<p>
+ s6-dnsip4-filter does not perform DNS resolutions itself. Instead, if forks
+a <a href="skadns/skadnsd.html">skadnsd</a> child and sends it queries, getting
+the results asynchronously. The s6-dns filter programs have actually been
+written as example uses of the <a href="skadns/">skadns library</a>.
+</p>
+
+</body>
+</html>