Update docs on MaxClients under worker, and clean up a few things.

This has been wrong for a LONG time, and it could really use a review
by some of the worker MPM developers.

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/trunk@96458 13f79535-47bb-0310-9956-ffa450edef68

diff --git a/docs/manual/mod/mpm_common.xml b/docs/manual/mod/mpm_common.xml
index e391bc018da72baa5f8ad44b572dc4be9ff42be9..9cd6d3ee533539de06c7658e72a4aad159fd6a05 100644 (file)
--- a/docs/manual/mod/mpm_common.xml
+++ b/docs/manual/mod/mpm_common.xml
@@ -220,34 +220,33 @@ listens to</description>
 <description>Maximum number of child processes that will be created
 to serve requests</description>
 <syntax>MaxClients <em>number</em></syntax>
-<default>MaxClients
-    8 (with threads) MaxClients 256</default>
 <contextlist><context>server config</context></contextlist>
 <modulelist><module>worker</module><module>prefork</module>
 </modulelist>
 
 <usage>
     <p>The <directive>MaxClients</directive> directive sets the limit
-    on the number of child processes that will be created to serve
-    requests. When the server is built without threading, no more than
-    this number of clients can be served simultaneously. To configure
-    more than 256 clients with the prefork MPM, you must use the
-    <directive module="mpm_common">ServerLimit</directive> directive.
-    To configure more than 1024 clients with the worker MPM, you must
-    use the <directive module="mpm_common">ServerLimit</directive> and
-    <directive module="mpm_common">ThreadLimit</directive> directives.</p>
-
-    <p>Any connection attempts over the
-    <directive>MaxClients</directive> limit will normally be queued,
-    up to a number based on the <directive module="mpm_common"
-    >ListenBacklog</directive> directive. Once a child
-    process is freed at the end of a different request, the connection
-    will then be serviced.</p>
-
-    <p>When the server is compiled with threading, then the maximum
-    number of simultaneous requests that can be served is obtained
-    from the value of this directive multiplied by
-    <directive module="mpm_common">ThreadsPerChild</directive>.</p>
+    on the number of simultaneous requests that will be served.  Any
+    connection attempts over the <directive>MaxClients</directive>
+    limit will normally be queued, up to a number based on the
+    <directive module="mpm_common">ListenBacklog</directive>
+    directive. Once a child process is freed at the end of a different
+    request, the connection will then be serviced.</p>
+
+    <p>For non-threaded servers (i.e., <module>prefork</module>),
+    <directive>MaxClients</directive> translates into the maximum
+    number of child processes that will be launched to serve requests.
+    The default value is 256; to increase it, you must also raise
+    <directive module="mpm_common">ServerLimit</directive>.</p>
+
+    <p>For threaded servers (i.e., <module>worker</module>),
+    <directive>MaxClients</directive> restricts the total
+    number of threads that will be available to serve clients.
+    The default value is 16 multiplied by the value of
+    <directive module="mpm_common">ThreadsPerChild</directive>.
+    To increase <directive>MaxClients</directive> to a value
+    that requires more than 16 processes, you must also
+    raise <directive module="mpm_common">ServerLimit</directive>.</p>
 </usage>
 </directivesynopsis>
 

diff --git a/docs/manual/mod/worker.xml b/docs/manual/mod/worker.xml
index af18c1f6333b8276efcc6b02c976b13c2d443587..38a94eb3230d740e33733c14b38e0e8fe07edf99 100644 (file)
--- a/docs/manual/mod/worker.xml
+++ b/docs/manual/mod/worker.xml
@@ -43,18 +43,35 @@
     idle server threads, which stand ready to serve incoming
     requests. In this way, clients do not need to wait for a new
     threads or processes to be created before their requests can be
-    served. Apache assesses the total number of idle threads in all
-    processes, and forks or kills processes to keep this number within
-    the boundaries specified by <directive
+    served. The number of processes that will initially launched is
+    set by the <directive module="mpm_common">StartServers</directive>
+    directive. Then during operation, Apache assesses the total number
+    of idle threads in all processes, and forks or kills processes to
+    keep this number within the boundaries specified by <directive
     module="mpm_common">MinSpareThreads</directive> and <directive
     module="mpm_common">MaxSpareThreads</directive>. Since this
     process is very self-regulating, it is rarely necessary to modify
     these directives from their default values. The maximum number of
-    clients that may be served simultaneously is determined by
-    multiplying the maximum number of server processes that will be
-    created (<directive module="mpm_common">MaxClients</directive>) by
-    the number of threads created in each process (<directive
-    module="mpm_common">ThreadsPerChild</directive>).</p>
+    clients that may be served simultaneously (i.e., the maximum total
+    number of threads in all processes) is determined by the
+    <directive module="mpm_common">MaxClients</directive> directive,
+    while the maximum number of processes that can be launched is set
+    by the <directive module="mpm_common">ServerLimit</directive>
+    directive.  <directive module="mpm_common">ServerLimit</directive>
+    multiplied by <directive
+    module="mpm_common">ThreadsPerChild</directive> must be greater
+    than or equal to <directive
+    module="mpm_common">MaxClients</directive></p>
+
+    <p>A typical configuration of the process-thread controls in
+    the <module>worker</module> MPM could look as follows:</p>
+
+<example><pre>StartServers         2
+MaxClients         150
+MinSpareThreads     25
+MaxSpareThreads     75
+ThreadsPerChild     25
+ServerLimit         16</pre></example>
 
     <p>While the parent process is usually started as root under Unix
     in order to bind to port 80, the child processes and threads are