Commit 9b065652 authored by Luca Toscano's avatar Luca Toscano
Browse files

Clarify some mod_proxy concepts in the docs 

I have recently got some basic mod-proxy concepts
and these simple info would have helped me a lot:

- connection pooling == reuse. My brain did not pick
  up this simple nomenclature at first read for some
  reason, probably because we have parameters called
  "*reuse*" and not "*pooling*". I find the term
  "reuse" clearer than "pooling", but this is
  debatable :)
- the concept of "Worker" needs to be read before
  the ProxyPass section (that people probably read
  directly after checking a search engine) to fully
  understand its power.
- the fact that a worker is related to a single process
  (and not shared as a Balancer) is a really important
  concept that I tried to highlight when talking
  about maxconns (question popped up a lot of times
  from users).

Hope that my understanding is correct, feedback is welcome.



git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@1803171 13f79535-47bb-0310-9956-ffa450edef68
parent df4fbcf4

docs/manual/mod/mod_proxy.html.en

+18 −12
Original line number Diff line number Diff line
@@ -244,7 +244,7 @@ ProxyVia On 


      <p>The two default workers have a fixed configuration

      and will be used if no other worker matches the request.

      They do not use HTTP Keep-Alive or connection pooling.

      They do not use HTTP Keep-Alive or connection reuse.

      The TCP connections to the origin server will instead be

      opened and closed for each request.</p>
@@ -345,8 +345,8 @@ ProxyPass "/examples" "http://backend.example.com/examples" timeout=10</pre> 
      <div class="note"><h3>DNS resolution for origin domains</h3>

      <p>DNS resolution happens when the socket to

        the origin domain is created for the first time.

        When connection pooling is used, each backend domain is resolved 

        only once per child process, and reused for all further connections 

        When connection reuse is enabled, each backend domain is resolved 

        only once per child process, and cached for all further connections 

        until the child is recycled. This information should to be considered 

        while planning DNS maintenance tasks involving backend domains. 

        Please also check <code class="directive"><a href="#proxypass">ProxyPass</a></code>
@@ -1020,7 +1020,11 @@ through</td></tr> 
    a local virtual path; <var>url</var> is a partial URL for the

    remote server and cannot include a query string.</p>



    <div class="note"><strong>Note: </strong>This directive is not supported within

    <div class="note">It is strongly suggested to review the concept of a

    <a href="#workers">Worker</a> before proceeding any further

    with this section.</div>



    <div class="note">This directive is not supported within

    <code class="directive"><a href="../mod/core.html#directory">&lt;Directory&gt;</a></code> and

    <code class="directive"><a href="../mod/core.html#files">&lt;Files&gt;</a></code> containers.</div>
@@ -1133,24 +1137,26 @@ ProxyPass "/mirror/foo" "http://backend.example.com"</pre> 
    using  <code>key=value</code> parameters, described in the tables

    below.</p>



    <div class="warning"><h3>Maximum connections to the backend</h3>

    <p>By default, mod_proxy will allow and retain the maximum number of

    connections that could be used simultaneously by that web server child

    process.  Use the <code>max</code> parameter to reduce the number from

    the default.  Use the <code>ttl</code> parameter to set an optional

    the default. The pool of connections is maintained per web server child

    process, and <code>max</code> and other settings are not coordinated

    among all child processes, except when only one child process is allowed

    by configuration or MPM design.</p>

    </div>



    <p>Use the <code>ttl</code> parameter to set an optional

    time to live; connections which have been unused for at least

    <code>ttl</code> seconds will be closed.  <code>ttl</code> can be used

    to avoid using a connection which is subject to closing because of the

    backend server's keep-alive timeout.</p>



    <p>The pool of connections is maintained per web server child

    process, and <code>max</code> and other settings are not coordinated

    among all child processes, except when only one child process is allowed

    by configuration or MPM design.</p>



    <div class="example"><h3>Example</h3><pre class="prettyprint lang-config">ProxyPass "/example" "http://backend.example.com" max=20 ttl=120 retry=300</pre>

</div>



    <table class="bordered"><tr><th>BalancerMember parameters</th></tr></table>

    <table class="bordered"><tr><th>Worker|BalancerMember parameters</th></tr></table>

    <table>

    <tr><th>Parameter</th>

        <th>Default</th>
@@ -1201,7 +1207,7 @@ ProxyPass "/mirror/foo" "http://backend.example.com"</pre> 
    httpd and

    the backend server (regardless of protocol) tends to silently

    drop connections or when backends themselves may be under round-

    robin DNS. To disable connection pooling reuse,

    robin DNS. To disable connection reuse,

    set this property value to <code>On</code>.

    </td></tr>

    <tr><td>enablereuse</td>

docs/manual/mod/mod_proxy.xml

+18 −12
Original line number Diff line number Diff line
@@ -201,7 +201,7 @@ ProxyVia On 


      <p>The two default workers have a fixed configuration

      and will be used if no other worker matches the request.

      They do not use HTTP Keep-Alive or connection pooling.

      They do not use HTTP Keep-Alive or connection reuse.

      The TCP connections to the origin server will instead be

      opened and closed for each request.</p>
@@ -307,8 +307,8 @@ ProxyPass "/examples" "http://backend.example.com/examples" timeout=10 
      <note><title>DNS resolution for origin domains</title>

      <p>DNS resolution happens when the socket to

        the origin domain is created for the first time.

        When connection pooling is used, each backend domain is resolved 

        only once per child process, and reused for all further connections 

        When connection reuse is enabled, each backend domain is resolved 

        only once per child process, and cached for all further connections 

        until the child is recycled. This information should to be considered 

        while planning DNS maintenance tasks involving backend domains. 

        Please also check <directive module="mod_proxy">ProxyPass</directive>
@@ -881,7 +881,11 @@ expressions</description> 
    a local virtual path; <var>url</var> is a partial URL for the

    remote server and cannot include a query string.</p>



    <note><strong>Note: </strong>This directive is not supported within

    <note>It is strongly suggested to review the concept of a

    <a href="#workers">Worker</a> before proceeding any further

    with this section.</note>



    <note>This directive is not supported within

    <directive type="section" module="core">Directory</directive> and

    <directive type="section" module="core">Files</directive> containers.</note>
@@ -1002,27 +1006,29 @@ ProxyPass "/mirror/foo" "http://backend.example.com" 
    using  <code>key=value</code> parameters, described in the tables

    below.</p>



    <note type="warning"><title>Maximum connections to the backend</title>

    <p>By default, mod_proxy will allow and retain the maximum number of

    connections that could be used simultaneously by that web server child

    process.  Use the <code>max</code> parameter to reduce the number from

    the default.  Use the <code>ttl</code> parameter to set an optional

    the default. The pool of connections is maintained per web server child

    process, and <code>max</code> and other settings are not coordinated

    among all child processes, except when only one child process is allowed

    by configuration or MPM design.</p>

    </note>



    <p>Use the <code>ttl</code> parameter to set an optional

    time to live; connections which have been unused for at least

    <code>ttl</code> seconds will be closed.  <code>ttl</code> can be used

    to avoid using a connection which is subject to closing because of the

    backend server's keep-alive timeout.</p>



    <p>The pool of connections is maintained per web server child

    process, and <code>max</code> and other settings are not coordinated

    among all child processes, except when only one child process is allowed

    by configuration or MPM design.</p>



    <example><title>Example</title>

        <highlight language="config">

ProxyPass "/example" "http://backend.example.com" max=20 ttl=120 retry=300

        </highlight>

    </example>



    <table border="2"><tr><th>BalancerMember parameters</th></tr></table>

    <table border="2"><tr><th>Worker|BalancerMember parameters</th></tr></table>

    <table>

    <tr><th>Parameter</th>

        <th>Default</th>
@@ -1073,7 +1079,7 @@ ProxyPass "/example" "http://backend.example.com" max=20 ttl=120 retry=300 
    httpd and

    the backend server (regardless of protocol) tends to silently

    drop connections or when backends themselves may be under round-

    robin DNS. To disable connection pooling reuse,

    robin DNS. To disable connection reuse,

    set this property value to <code>On</code>.

    </td></tr>

    <tr><td>enablereuse</td>

docs/manual/mod/mod_proxy_fcgi.html.en

+1 −1
Original line number Diff line number Diff line
@@ -95,7 +95,7 @@ 
    <p> The following example passes the request URI as a filesystem

    path for the PHP-FPM daemon to run. The request URL is implicitly added

    to the 2nd parameter. The hostname and port following fcgi:// are where

    PHP-FPM is listening.  Connection pooling is enabled.</p>

    PHP-FPM is listening.  Connection pooling/reuse is enabled.</p>

    <div class="example"><h3>PHP-FPM</h3><pre class="prettyprint lang-config">ProxyPassMatch "^/myapp/.*\.php(/.*)?$" "fcgi://localhost:9000/var/www/" enablereuse=on</pre>

</div>

docs/manual/mod/mod_proxy_fcgi.xml

+1 −1
Original line number Diff line number Diff line
@@ -85,7 +85,7 @@ ProxyPass "/myapp/" "fcgi://localhost:4000/" enablereuse=on 
    <p> The following example passes the request URI as a filesystem

    path for the PHP-FPM daemon to run. The request URL is implicitly added

    to the 2nd parameter. The hostname and port following fcgi:// are where

    PHP-FPM is listening.  Connection pooling is enabled.</p>

    PHP-FPM is listening.  Connection pooling/reuse is enabled.</p>

    <example><title>PHP-FPM</title>

    <highlight language="config">

ProxyPassMatch "^/myapp/.*\.php(/.*)?$" "fcgi://localhost:9000/var/www/" enablereuse=on

docs/manual/rewrite/flags.html.en

+1 −1
Original line number Diff line number Diff line
@@ -536,7 +536,7 @@ client undue influence.</p> 
<p>Using this flag triggers the use of <code class="module"><a href="../mod/mod_proxy.html">mod_proxy</a></code>, without handling of persistent connections. This

means the performance of your proxy will be better if you set it up with <code class="directive"><a href="../mod/mod_proxy.html#proxypass">ProxyPass</a></code> or

<code class="directive"><a href="../mod/mod_proxy.html#proxypassmatch">ProxyPassMatch</a></code></p>

<p>This is because this flag triggers the use of the default worker, which does not handle connection pooling.</p>

<p>This is because this flag triggers the use of the default worker, which does not handle connection pooling/reuse.</p>

<p>Avoid using this flag and prefer those directives, whenever you can.</p>

</div>