httpd.conf 組態檔

紀錄Apache HTTP Server的組態檔(configuration file)，這是預設的原始檔案內容，版本是Apache 2.2.16版。注意下方粗體字是常用的組態指令與注意事項。

#
# This is the main Apache HTTP server configuration file.  It contains the
# configuration directives that give the server its instructions.
# See for detailed information.
# In particular, see
#
# for a discussion of each configuration directive.
#
# Do NOT simply read the instructions in here without understanding
# what they do.  They're here only as hints or reminders.  If you are unsure
# consult the online docs. You have been warned.
#
# Configuration and logfile names: If the filenames you specify for many
# of the server's control files begin with "/" (or "drive:/" for Win32), the
# server will use that explicit path.  If the filenames do *not* begin
# with "/", the value of ServerRoot is prepended -- so "logs/foo.log"
# with ServerRoot set to "C:/Program Files/Apache Software Foundation/Apache2.2" will be interpreted by the
# server as "C:/Program Files/Apache Software Foundation/Apache2.2/logs/foo.log".
#
# NOTE: Where filenames are specified, you must use forward slashes
# instead of backslashes (e.g., "c:/apache" instead of "c:\apache").
# If a drive letter is omitted, the drive on which httpd.exe is located
# will be used by default.  It is recommended that you always supply
# an explicit drive letter in absolute paths to avoid confusion.

#
# ServerRoot: The top of the directory tree under which the server's
# configuration, error, and log files are kept.
#
# Do not add a slash at the end of the directory path.  If you point
# ServerRoot at a non-local disk, be sure to point the LockFile directive
# at a local disk.  If you wish to share the same ServerRoot for multiple
# httpd daemons, you will need to change at least LockFile and PidFile.
#
ServerRoot "C:/Program Files/Apache Software Foundation/Apache2.2"

#
# Listen: Allows you to bind Apache to specific IP addresses and/or
# ports, instead of the default. See also the
# directive.
#
# Change this to Listen on specific IP addresses as shown below to
# prevent Apache from glomming onto all bound IP addresses.
#
#Listen 12.34.56.78:80
Listen 80

#
# Dynamic Shared Object (DSO) Support
#
# To be able to use the functionality of a module which was built as a DSO you
# have to place corresponding `LoadModule' lines at this location so the
# directives contained in it are actually available _before_ they are used.
# Statically compiled modules (those listed by `httpd -l') do not need
# to be loaded here.
#
# Example:
# LoadModule foo_module modules/mod_foo.so
#
LoadModule actions_module modules/mod_actions.so
LoadModule alias_module modules/mod_alias.so
LoadModule asis_module modules/mod_asis.so
LoadModule auth_basic_module modules/mod_auth_basic.so
#LoadModule auth_digest_module modules/mod_auth_digest.so
#LoadModule authn_alias_module modules/mod_authn_alias.so
#LoadModule authn_anon_module modules/mod_authn_anon.so
#LoadModule authn_dbd_module modules/mod_authn_dbd.so
#LoadModule authn_dbm_module modules/mod_authn_dbm.so
LoadModule authn_default_module modules/mod_authn_default.so
LoadModule authn_file_module modules/mod_authn_file.so
#LoadModule authnz_ldap_module modules/mod_authnz_ldap.so
#LoadModule authz_dbm_module modules/mod_authz_dbm.so
LoadModule authz_default_module modules/mod_authz_default.so
LoadModule authz_groupfile_module modules/mod_authz_groupfile.so
LoadModule authz_host_module modules/mod_authz_host.so
#LoadModule authz_owner_module modules/mod_authz_owner.so
LoadModule authz_user_module modules/mod_authz_user.so
LoadModule autoindex_module modules/mod_autoindex.so
#LoadModule cache_module modules/mod_cache.so
#LoadModule cern_meta_module modules/mod_cern_meta.so
LoadModule cgi_module modules/mod_cgi.so
#LoadModule charset_lite_module modules/mod_charset_lite.so
#LoadModule dav_module modules/mod_dav.so
#LoadModule dav_fs_module modules/mod_dav_fs.so
#LoadModule dav_lock_module modules/mod_dav_lock.so
#LoadModule dbd_module modules/mod_dbd.so
#LoadModule deflate_module modules/mod_deflate.so
LoadModule dir_module modules/mod_dir.so
#LoadModule disk_cache_module modules/mod_disk_cache.so
#LoadModule dumpio_module modules/mod_dumpio.so
LoadModule env_module modules/mod_env.so
#LoadModule expires_module modules/mod_expires.so
#LoadModule ext_filter_module modules/mod_ext_filter.so
#LoadModule file_cache_module modules/mod_file_cache.so
#LoadModule filter_module modules/mod_filter.so
#LoadModule headers_module modules/mod_headers.so
#LoadModule ident_module modules/mod_ident.so
#LoadModule imagemap_module modules/mod_imagemap.so
LoadModule include_module modules/mod_include.so
#LoadModule info_module modules/mod_info.so
LoadModule isapi_module modules/mod_isapi.so
#LoadModule ldap_module modules/mod_ldap.so
#LoadModule logio_module modules/mod_logio.so
LoadModule log_config_module modules/mod_log_config.so
#LoadModule log_forensic_module modules/mod_log_forensic.so
#LoadModule mem_cache_module modules/mod_mem_cache.so
LoadModule mime_module modules/mod_mime.so
#LoadModule mime_magic_module modules/mod_mime_magic.so
LoadModule negotiation_module modules/mod_negotiation.so
#LoadModule proxy_module modules/mod_proxy.so
#LoadModule proxy_ajp_module modules/mod_proxy_ajp.so
#LoadModule proxy_balancer_module modules/mod_proxy_balancer.so
#LoadModule proxy_connect_module modules/mod_proxy_connect.so
#LoadModule proxy_ftp_module modules/mod_proxy_ftp.so
#LoadModule proxy_http_module modules/mod_proxy_http.so
#LoadModule reqtimeout_module modules/mod_reqtimeout.so
#LoadModule rewrite_module modules/mod_rewrite.so
LoadModule setenvif_module modules/mod_setenvif.so
#LoadModule speling_module modules/mod_speling.so
#LoadModule ssl_module modules/mod_ssl.so
#LoadModule status_module modules/mod_status.so
#LoadModule substitute_module modules/mod_substitute.so
#LoadModule unique_id_module modules/mod_unique_id.so
#LoadModule userdir_module modules/mod_userdir.so
#LoadModule usertrack_module modules/mod_usertrack.so
#LoadModule version_module modules/mod_version.so
#LoadModule vhost_alias_module modules/mod_vhost_alias.so



#
# If you wish httpd to run as a different user or group, you must run
# httpd as root initially and it will switch.
#
# User/Group: The name (or #number) of the user/group to run httpd as.
# It is usually good practice to create a dedicated user and group for
# running httpd, as with most system services.
#
User daemon
Group daemon




# 'Main' server configuration
#
# The directives in this section set up the values used by the 'main'
# server, which responds to any requests that aren't handled by a
# definition.  These values also provide defaults for
# any containers you may define later in the file.
#
# All of these directives may appear inside containers,
# in which case these default settings will be overridden for the
# virtual host being defined.
#

#
# ServerAdmin: Your address, where problems with the server should be
# e-mailed.  This address appears on some server-generated pages, such
# as error documents.  e.g. admin@your-domain.com
#
ServerAdmin admin@localhost

#
# ServerName gives the name and port that the server uses to identify itself.
# This can often be determined automatically, but we recommend you specify
# it explicitly to prevent problems during startup.
#
# If your host doesn't have a registered DNS name, enter its IP address here.
#
#ServerName localhost:80

#
# DocumentRoot: The directory out of which you will serve your
# documents. By default, all requests are taken from this directory, but
# symbolic links and aliases may be used to point to other locations.
#
DocumentRoot "C:/Program Files/Apache Software Foundation/Apache2.2/htdocs"

#
# Each directory to which Apache has access can be configured with respect
# to which services and features are allowed and/or disabled in that
# directory (and its subdirectories).
#
# First, we configure the "default" to be a very restrictive set of
# features.
#

    Options FollowSymLinks
    AllowOverride None
    Order deny,allow
    Deny from all


#
# Note that from this point forward you must specifically allow
# particular features to be enabled - so if something's not working as
# you might expect, make sure that you have specifically enabled it
# below.
#

#
# This should be changed to whatever you set DocumentRoot to.
#

    #
    # Possible values for the Options directive are "None", "All",
    # or any combination of:
    #   Indexes Includes FollowSymLinks SymLinksifOwnerMatch ExecCGI MultiViews
    #
    # Note that "MultiViews" must be named *explicitly* --- "Options All"
    # doesn't give it to you.
    #
    # The Options directive is both complicated and important.  Please see
    # http://httpd.apache.org/docs/2.2/mod/core.html#options
    # for more information.
    #
    Options Indexes FollowSymLinks

    #
    # AllowOverride controls what directives may be placed in .htaccess files.
    # It can be "All", "None", or any combination of the keywords:
    #   Options FileInfo AuthConfig Limit
    #
    AllowOverride None

    #
    # Controls who can get stuff from this server.
    #
    Order allow,deny
    Allow from all



#
# DirectoryIndex: sets the file that Apache will serve if a directory
# is requested.
#

    DirectoryIndex index.html


#
# The following lines prevent .htaccess and .htpasswd files from being
# viewed by Web clients.
#

    Order allow,deny
    Deny from all
    Satisfy All


#
# ErrorLog: The location of the error log file.
# If you do not specify an ErrorLog directive within a
# container, error messages relating to that virtual host will be
# logged here.  If you *do* define an error logfile for a
# container, that host's errors will be logged there and not here.
#
ErrorLog "logs/error.log"

#
# LogLevel: Control the number of messages logged to the error_log.
# Possible values include: debug, info, notice, warn, error, crit,
# alert, emerg.
#
LogLevel warn


    #
    # The following directives define some format nicknames for use with
    # a CustomLog directive (see below).
    #
    LogFormat "%h %l %u %t \"%r\" %>s %b \"%{Referer}i\" \"%{User-Agent}i\"" combined
    LogFormat "%h %l %u %t \"%r\" %>s %b" common

    
      # You need to enable mod_logio.c to use %I and %O
      LogFormat "%h %l %u %t \"%r\" %>s %b \"%{Referer}i\" \"%{User-Agent}i\" %I %O" combinedio
  

    #
    # The location and format of the access logfile (Common Logfile Format).
    # If you do not define any access logfiles within a
    # container, they will be logged here.  Contrariwise, if you *do*
    # define per- access logfiles, transactions will be
    # logged therein and *not* in this file.
    #
    CustomLog "logs/access.log" common

    #
    # If you prefer a logfile with access, agent, and referer information
    # (Combined Logfile Format) you can use the following directive.
    #
    #CustomLog "logs/access.log" combined



    #
    # Redirect: Allows you to tell clients about documents that used to
    # exist in your server's namespace, but do not anymore. The client
    # will make a new request for the document at its new location.
    # Example:
    # Redirect permanent /foo http://localhost/bar

    #
    # Alias: Maps web paths into filesystem paths and is used to
    # access content that does not live under the DocumentRoot.
    # Example:
    # Alias /webpath /full/filesystem/path
    #
    # If you include a trailing / on /webpath then the server will
    # require it to be present in the URL.  You will also likely
    # need to provide a section to allow access to
    # the filesystem path.

    #
    # ScriptAlias: This controls which directories contain server scripts.
    # ScriptAliases are essentially the same as Aliases, except that
    # documents in the target directory are treated as applications and
    # run by the server when requested rather than as documents sent to the
    # client.  The same rules about trailing "/" apply to ScriptAlias
    # directives as to Alias.
    #
    ScriptAlias /cgi-bin/ "C:/Program Files/Apache Software Foundation/Apache2.2/cgi-bin/"




    #
    # ScriptSock: On threaded servers, designate the path to the UNIX
    # socket used to communicate with the CGI daemon of mod_cgid.
    #
    #Scriptsock logs/cgisock


#
# "C:/Program Files/Apache Software Foundation/Apache2.2/cgi-bin" should be changed to whatever your ScriptAliased
# CGI directory exists, if you have that configured.
#

    AllowOverride None
    Options None
    Order allow,deny
    Allow from all


#
# DefaultType: the default MIME type the server will use for a document
# if it cannot otherwise determine one, such as from filename extensions.
# If your server contains mostly text or HTML documents, "text/plain" is
# a good value.  If most of your content is binary, such as applications
# or images, you may want to use "application/octet-stream" instead to
# keep browsers from trying to display binary files as though they are
# text.
#
DefaultType text/plain


    #
    # TypesConfig points to the file containing the list of mappings from
    # filename extension to MIME-type.
    #
    TypesConfig conf/mime.types

    #
    # AddType allows you to add to or override the MIME configuration
    # file specified in TypesConfig for specific file types.
    #
    #AddType application/x-gzip .tgz
    #
    # AddEncoding allows you to have certain browsers uncompress
    # information on the fly. Note: Not all browsers support this.
    #
    #AddEncoding x-compress .Z
    #AddEncoding x-gzip .gz .tgz
    #
    # If the AddEncoding directives above are commented-out, then you
    # probably should define those extensions to indicate media types:
    #
    AddType application/x-compress .Z
    AddType application/x-gzip .gz .tgz

    #
    # AddHandler allows you to map certain file extensions to "handlers":
    # actions unrelated to filetype. These can be either built into the server
    # or added with the Action directive (see below)
    #
    # To use CGI scripts outside of ScriptAliased directories:
    # (You will also need to add "ExecCGI" to the "Options" directive.)
    #
    #AddHandler cgi-script .cgi

    # For type maps (negotiated resources):
    #AddHandler type-map var

    #
    # Filters allow you to process content before it is sent to the client.
    #
    # To parse .shtml files for server-side includes (SSI):
    # (You will also need to add "Includes" to the "Options" directive.)
    #
    #AddType text/html .shtml
    #AddOutputFilter INCLUDES .shtml


#
# The mod_mime_magic module allows the server to use various hints from the
# contents of the file itself to determine its type.  The MIMEMagicFile
# directive tells the module where the hint definitions are located.
#
#MIMEMagicFile conf/magic

#
# Customizable error responses come in three flavors:
# 1) plain text 2) local redirects 3) external redirects
#
# Some examples:
#ErrorDocument 500 "The server made a boo boo."
#ErrorDocument 404 /missing.html
#ErrorDocument 404 "/cgi-bin/missing_handler.pl"
#ErrorDocument 402 http://localhost/subscription_info.html
#

#
# EnableMMAP and EnableSendfile: On systems that support it,
# memory-mapping or the sendfile syscall is used to deliver
# files.  This usually improves server performance, but must
# be turned off when serving from networked-mounted
# filesystems or if support for these functions is otherwise
# broken on your system.
#
#EnableMMAP off
#EnableSendfile off

# Supplemental configuration
#
# The configuration files in the conf/extra/ directory can be
# included to add extra features or to modify the default configuration of
# the server, or you may simply copy their contents here and change as
# necessary.

# Server-pool management (MPM specific)
#Include conf/extra/httpd-mpm.conf

# Multi-language error messages
#Include conf/extra/httpd-multilang-errordoc.conf

# Fancy directory listings
#Include conf/extra/httpd-autoindex.conf

# Language settings
#Include conf/extra/httpd-languages.conf

# User home directories
#Include conf/extra/httpd-userdir.conf

# Real-time info on requests and configuration
#Include conf/extra/httpd-info.conf

# Virtual hosts
#Include conf/extra/httpd-vhosts.conf

# Local access to the Apache HTTP Server Manual
#Include conf/extra/httpd-manual.conf

# Distributed authoring and versioning (WebDAV)
#Include conf/extra/httpd-dav.conf

# Various default settings
#Include conf/extra/httpd-default.conf

# Secure (SSL/TLS) connections
#Include conf/extra/httpd-ssl.conf
#
# Note: The following must must be present to support
#       starting without SSL on platforms with no /dev/random equivalent
#       but a statically compiled-in mod_ssl.
#

SSLRandomSeed startup builtin
SSLRandomSeed connect builtin

Apache HTTP Server 主目錄

今天看看Apache HTTP Server安裝於Win32目錄之下的情形，一般來說，預設值的安裝路徑是C:\Program Files\Apache Software Foundation\Apache2.2之中，裡面的資料夾與檔案如下：

• 檔案資料夾
• bin，
  主程式(httpd.exe)的存放位置，還有一些工具像是ApacheMonitor.exe的程式。
• cgi-bin，
  儲存CGI script的地方。
• conf，
  組態檔(Configuration File)的存放位置，最重要的是httpd.conf檔案
• error，
  錯誤訊息頁面。
• htdocs，
  網站根目錄的位置。
• icons，
  存放圖示檔。
• logs，
  存放日誌檔案，像是access.log和error.log。
• manual，
  使用手冊。
• modules，
  模組的存放位置。
• 檔案
• ABOUT_APACHE.txt，說明Apache HTTP Server是什麼、做什麼、誰做的。
• CHANGES.txt，修改記錄
• INSTALL.txt，安裝說明文件
• LICENSE.txt，授權條款(許可說明文件)
• NOTICE.txt，注意事項
• README.txt，請先讀我檔案
• README-win32.txt，有關Windows平台的README檔案

###

Apache The Definitive Guide 技術手冊

經過了一年多一點，這段期間累積對於Windows應用程式開發的知識與經驗，現在回頭看看Apache，我的視野已經不再是相同的高度。再次重拾書籍複習，這本書是「Apache技術手冊(第三版)」，大概是中文書裡唯一有關Apache的經典翻譯書籍！前一年，沒有時間細細品味Apache的美，現在經歷多了，有能力可以仔細咀嚼那其中的奧妙之處。

這只是開端。This is an inception.

Ben Laurie & Peter Laurie原著，郭文生編譯，Apache技術手冊(第三版)，台北：歐萊禮，2003。

發行者：美商歐萊禮股份有限公司台灣分公司(O'Reilly Media, Inc. Taiwan Branch)

目錄

第一章 Apache 入門

• Web伺服器的功能
• Apache的運作方式
• Apache與網路
• HTTP客戶端的運作方式
• 伺服端的運作方式
• Apache之安裝規劃
• Windows的問題
• Apache的版本
• 安裝Apache
• Unix下建製Apache 1.3.X
• Apache 2.0的新功能
• 在Unix下製成與安裝Apache 2.0
• Windows下的Apache

第二章 設定 Apache：起手式

• Apache 網站的背後
• site.toddle
• 架設 Unix 伺服器
• 指令

第三章 建立真實網站

• 較佳的網站：site.simple
• Butterthlies 公司上線了
• 區塊指令
• 其它的指令
• HTTP 的回應標頭
• 再啟動
• .htaccess
• CERN 中介檔
• 逾期

第四章 虛擬主機

• 兩個網站與 Apache
• 虛擬主機
• 兩個 Apache 的分身
• 動態設定的虛擬主機

第五章 驗證

• 驗證協定
• 驗證指令
• Unix 下的密碼
• Win32 下的密碼
• Web 上的密碼
• 從用戶端的觀點出發
• CGI Script
• 主題的變化
• Order、Allow、與 Deny
• Unix 上的 DBM 檔
• 摘要驗證
• 匿名存取
• 實驗
• 自動的使用者資訊
• 使用 .htaccess 檔
• 指令的改寫

第六章 網頁內容描述與修正

• MIME 類型
• 網頁內容協商
• 語言協商
• 類型對應表
• 瀏覽器與 HTTP 1.1
• 過濾器

第七章 製作索引

• 製作較佳的索引
• 製作自己的索引
• 影像地圖（Imagemap）
• 影像地圖的指令

第八章 重導向

• Alias
• Rewrite
• 重寫的範例
• 拼字

第九章 代理伺服器

• 安全防護
• 代理伺服器的指令
• 外表上的錯誤
• 效能
• 設定

第十章 日誌紀錄

• 用 Script 及資料庫建立日誌
• Apache 的日誌機制
• 組態的日誌紀錄
• AddModuleInfo
• 狀態

第十一章 安全防護

• 內部與外部使用者
• 數位簽章、虛擬現金
• 憑證
• 防火牆
• 法律上的問題
• Secure Sockets Layer（SSL）
• Apache 的安全防護措施
• SSL 的指令
• 密碼套件
• 實際狀況下的安全防護
• 未來的方向

第十二章 執行大型網站

• 主機架設
• 伺服器的安全防護
• 管理大型網站
• 支援軟體
• 擴充性
• 負載平衡

第十三章 建製應用程式

• 以網站當作應用程式
• 提供應用程式的邏輯
• XML、XSLT，與 Web 應用程式

第十四章 伺服端擴充

• 檔案大小
• 檔案修改時間
• 引入
• 執行 CGI
• Echo
• Apache v2：SSI 過濾器

第十五章 PHP

• 安裝 PHP
• site.php

第十六章 CGI 與 Perl

• CGI 的世界
• 告訴 Apache 關於 script 的資訊『318』
• 設定環境變數
• Cookies
• Script 指令
• Unix 上的 suEXEC
• 處理器（Handler）
• Actions
• 瀏覽器

第十七章 mod_perl

• mod_perl 的運作方式
• mod_perl 的文件
• 安裝 mod_perl — 簡易法
• 修正你的 script 以便在 mod_perl 下執行
• 全域變數
• Strict 前置條件
• 載入變動
• 開啟與關閉檔案
• 設定 Apache 以使用 mod_perl

第十八章 mod_jserv 與 Tomcat

• mod_jserv
• Tomcat
• 連接 Tomcat 與 Apache

第十九章 XML 與 Cocoon

• XML
• XML 與 Perl
• Cocoon
• Cocoon 1.8 與 JServ
• Cocoon 2.0.3 與 Tomcat
• 測試 Cocoon

第二十章 Apache 的 API

• 文件
• APR
• 集散區（Pool）
• 以伺服器為基準的組態
• 以目錄為基準的組態
• 以請求為基準（per-request）的資訊
• 存取組態與請求資訊
• 掛鉤、選擇性的掛鉤，以及選擇性的函式
• 過濾器、儲存區，及儲存區串列
• 模組

第廿一章 編寫 Apache 的模組

• 概要說明
• 狀態碼
• 模組結構
• 完整的範例
• 一般的提示
• 移植至 Apache 2.0

附錄 Apache 1.x 的 API

索引

Apache Filter Module過濾器模組

從應用模組程式開發的角度來看，Apache 2.x的最重要改進與創新就是過濾器(filter)的架構，過濾器使得處理不同的資料可以隨心所欲的運作。

從Apache 2的架構來看，可以分為兩個正交軸，一為處理軸(Processing Axis)，另一為資料軸(Data Axis)。每一次的請求皆由處理軸來接受，其中會經過許許多多的掛勾(hook)，於是進行一些處理(handler)的呼叫。

過濾器依據過濾資料的處理階段有這幾種不同類型：

• Content Filter：在接受請求中，用來處理文件的資料。一般來說我們都式設計這類的過濾器居多，例如過濾HTML中特定的文字。
• Protocol Filter：用來處理通訊協定(protocol)的資料。在Apache中，Protocol Filter用在將HTTP通訊協定轉換request_rec的資料結構。
• Connection Filter：處理TCP連線的資料。用來處理conn_rec的資料結構，跟HTTP和request_recr皆無關係，你可以看成是操作在傳輸層(Transport Layer)的部分，而Content Filter和Protocol Filter是操作在應用層(Application Layer)。

過濾器依據放置的位置則有這幾種不同類型：

• Input Filter：處理來自client請求(request)的資料，故稱為輸入過濾器(Input Filter)。
• Output Filter：處理將回應(response)給client的資料，故稱為輸出過濾器(Output Filter)。

此外，過濾器也有點像掛勾的樣子，有著另一種資料型態來處理過濾器之間的傳遞，是要依靠ap_filter_t的資料結構

/**
 * @brief The representation of a filter chain.  
 *
 * Each request has a list
 * of these structures which are called in turn to filter the data.  Sub
 * requests get an exact copy of the main requests filter chain.
 */
struct ap_filter_t {
    /** The internal representation of this filter.  This includes
     *  the filter's name, type, and the actual function pointer.
     */
    ap_filter_rec_t *frec;

    /** A place to store any data associated with the current filter */
    void *ctx;

    /** The next filter in the chain */
    ap_filter_t *next;

    /** The request_rec associated with the current filter.  If a sub-request
     *  adds filters, then the sub-request is the request associated with the
     *  filter.
     */
    request_rec *r;

    /** The conn_rec associated with the current filter.  This is analogous
     *  to the request_rec, except that it is used for input filtering.
     */
    conn_rec *c;
};

如果需要在多個過濾器之間傳送資料，你就需要用ctx這個資料欄位。

Apache WinNT MPM多執行模組的運作方式

WinNT的MPM模組當中，使用一種稱為「I/O Completion Ports」的技術來處理各執行緒的請求分配，「I/O Completion Ports」中文稱為「I/O完全埠」、「I/O完成通訊埠」或「完成端口」等，也有簡稱IOCP。

主要資料可以參考MSDN的兩個網址：

http://msdn.microsoft.com/en-us/library/aa365198(VS.85).aspx

http://www.microsoft.com/taiwan/technet/sysinternals/information/iocompletionports.mspx

Apache MultiThread多執行緒寫法

在WIN32的Apache中，MPM模組(多進程並發處理模塊)使用多執行緒來處理Request，其中避免單一資源被多個執行緒搶奪，必須使用一些技巧，可以參考MSDN中的Synchronization Objects主題文章(http://msdn.microsoft.com/en-us/library/ms686364(VS.85).aspx)，常用的Synchronization Object有下列幾種：Event、Mutex、Semaphore和Waitable Timer。利用Synchronization Objects可以被指定給其中之一的Wait Function，讓這些Thread可以等等，如此避掉單一資源被多個執行緒搶奪。

這些Object的特點如下：

• Event：用來通知一個或多個執行緒已經發生一件事情，詳見http://msdn.microsoft.com/en-us/library/ms682655(VS.85).aspx
• Mutex：每一時間點只能被一個執行緒擁有此物件(Mutex Object)，主要是用在多執行緒對共用資源的存取，可以達到執行緒互相唯一(Mutually Exclusive, Mut+Ex=Mutex)的操作方式。詳見http://msdn.microsoft.com/en-us/library/ms684266(VS.85).aspx
• Semaphore：維持記數(count)在0和某一最大數值之間，用來限制同時存取共用資源的執行緒數量。詳見http://msdn.microsoft.com/en-us/library/ms685129(VS.85).aspx
• Waitable Timer：設定一個或多個已經定時的執行緒，當時間到達時則通知這些執行緒。詳見http://msdn.microsoft.com/en-us/library/ms687012(VS.85).aspx

除了這四個物件之外，其他可以用來同步執行緒的物件還有：Change notification、Console input、Job、Memory resource notification、Process、Thread等。

這裡還沒有提到一點，為了讓執行緒可以停下來等同步，我們必須使用Wait Function這個功能，Wait Function會在執行的時候block停止住，直到特定的條件(criteria)符合達成才會繼續執行下去，當然你也可以設定一段時間(time-out interval)，當條件不符合時，進入等待的狀態，若時間超過則會繼續執行而不等待。

Apache Platform平台運作說明

當我在拆解Apache的同時，發現到一個不能忽略的問題，我們必須將視野提高到較高層次來觀看原始碼(source code)，才不會見樹不見林，也較能對Apache的運作有深刻的體悟！樹、林是兩難。

Apache HTTP Server(簡稱Apache)通常在作業系統啟動的時候，就會開始在背後執行，成為 Linux中的Deamon，或是Windows中的Service，有一件事：Apache在啟動的時候，是比較耗費系統資源，所以通常是OS啟動後就Apache提供服務，不太像一般程式，你要用到才去呼叫。

Apache的程式當中分成兩個階段執行，一個是啟動階段(start-up phase)，另一個是運行階段(operational pahse)。

啟動階段(start-up phase)的作用有：讀取組態檔(httpd.conf或是.htaccess)；載入各個模組(module)；初始化系統資源，包含log檔、共享記憶體、資料連結等。

啟動階段完成之後，進入運行階段(operational pahse)，運行階段主要由MPM(Multi-Processing Module)模組接管控制，MPM會產生一個或多個的執行緒池(Thread Pool)或是行程(process)，像Windows中只會有兩個process(即httpd.exe)，其中之一是啟動階段的Master Process，另一個是運行階段的Child Process，由Child Process產生Thread Pool，此後由Thread Pool服務所有的Request請求。主要是因為Windows是支援Multi-Thread的作業系統。

Request Processing in Apache 2.0請求的處理過程

Apache中HTTP Request的程序主要是request.c中的函數來處理，這個函數是ap_process_request_internal，所有的請求都由它處理，包含子請求(subrequest)和轉向(redirect)。

其程式碼如下所示：

/* This is the master logic for processing requests.  Do NOT duplicate
* this logic elsewhere, or the security model will be broken by future
* API changes.  Each phase must be individually optimized to pick up
* redundant/duplicate calls by subrequests, and redirects.
*/
AP_DECLARE(int) ap_process_request_internal(request_rec *r)
{
   int file_req = (r->main && r->filename);
   int access_status;

   /* Ignore embedded %2F's in path for proxy requests */
   if (!r->proxyreq && r->parsed_uri.path) {
       core_dir_config *d;
       d = ap_get_module_config(r->per_dir_config, &core_module);
       if (d->allow_encoded_slashes) {
           access_status = ap_unescape_url_keep2f(r->parsed_uri.path);
       }
       else {
           access_status = ap_unescape_url(r->parsed_uri.path);
       }
       if (access_status) {
           if (access_status == HTTP_NOT_FOUND) {
               if (! d->allow_encoded_slashes) {
                   ap_log_rerror(APLOG_MARK, APLOG_INFO, 0, r,
                                 "found %%2f (encoded '/') in URI "
                                 "(decoded='%s'), returning 404",
                                 r->parsed_uri.path);
               }
           }
           return access_status;
       }
   }

   ap_getparents(r->uri);     /* OK --- shrinking transformations... */

   /* All file subrequests are a huge pain... they cannot bubble through the
    * next several steps.  Only file subrequests are allowed an empty uri,
    * otherwise let translate_name kill the request.
    */
   if (!file_req) {
       if ((access_status = ap_location_walk(r))) {
           return access_status;
       }

       if ((access_status = ap_run_translate_name(r))) {
           return decl_die(access_status, "translate", r);
       }
   }

   /* Reset to the server default config prior to running map_to_storage
    */
   r->per_dir_config = r->server->lookup_defaults;

   if ((access_status = ap_run_map_to_storage(r))) {
       /* This request wasn't in storage (e.g. TRACE) */
       return access_status;
   }

   /* Rerun the location walk, which overrides any map_to_storage config.
    */
   if ((access_status = ap_location_walk(r))) {
       return access_status;
   }

   /* Only on the main request! */
   if (r->main == NULL) {
       if ((access_status = ap_run_header_parser(r))) {
           return access_status;
       }
   }

   /* Skip authn/authz if the parent or prior request passed the authn/authz,
    * and that configuration didn't change (this requires optimized _walk()
    * functions in map_to_storage that use the same merge results given
    * identical input.)  If the config changes, we must re-auth.
    */
   if (r->main && (r->main->per_dir_config == r->per_dir_config)) {
       r->user = r->main->user;
       r->ap_auth_type = r->main->ap_auth_type;
   }
   else if (r->prev && (r->prev->per_dir_config == r->per_dir_config)) {
       r->user = r->prev->user;
       r->ap_auth_type = r->prev->ap_auth_type;
   }
   else {
       switch (ap_satisfies(r)) {
       case SATISFY_ALL:
       case SATISFY_NOSPEC:
           if ((access_status = ap_run_access_checker(r)) != 0) {
               return decl_die(access_status, "check access", r);
           }

           if (ap_some_auth_required(r)) {
               if (((access_status = ap_run_check_user_id(r)) != 0)
                   || !ap_auth_type(r)) {
                   return decl_die(access_status, ap_auth_type(r)
                                 ? "check user.  No user file?"
                                 : "perform authentication. AuthType not set!",
                                 r);
               }

               if (((access_status = ap_run_auth_checker(r)) != 0)
                   || !ap_auth_type(r)) {
                   return decl_die(access_status, ap_auth_type(r)
                                 ? "check access.  No groups file?"
                                 : "perform authentication. AuthType not set!",
                                  r);
               }
           }
           break;

       case SATISFY_ANY:
           if (((access_status = ap_run_access_checker(r)) != 0)) {
               if (!ap_some_auth_required(r)) {
                   return decl_die(access_status, "check access", r);
               }

               if (((access_status = ap_run_check_user_id(r)) != 0)
                   || !ap_auth_type(r)) {
                   return decl_die(access_status, ap_auth_type(r)
                                 ? "check user.  No user file?"
                                 : "perform authentication. AuthType not set!",
                                 r);
               }

               if (((access_status = ap_run_auth_checker(r)) != 0)
                   || !ap_auth_type(r)) {
                   return decl_die(access_status, ap_auth_type(r)
                                 ? "check access.  No groups file?"
                                 : "perform authentication. AuthType not set!",
                                 r);
               }
           }
           break;
       }
   }
   /* XXX Must make certain the ap_run_type_checker short circuits mime
    * in mod-proxy for r->proxyreq && r->parsed_uri.scheme
    *                              && !strcmp(r->parsed_uri.scheme, "http")
    */
   if ((access_status = ap_run_type_checker(r)) != 0) {
       return decl_die(access_status, "find types", r);
   }

   if ((access_status = ap_run_fixups(r)) != 0) {
       return access_status;
   }

   return OK;
}

Apache Portable Runtime可移植執行環境

Apache HTTP Server運作時需要依賴另一個程式庫，也就是APR，全名為「Apache Portable Runtime可移植執行環境」，這其實是Apache Software Foundation(ASF)底下另一個專案計畫，請參考http://apr.apache.org/

APR中有一項重要的程式庫模組，這個模組是APR Pools(apr_pools.h)，這是APR和Apchee中管理資源的重要模組！其資料型態是apr_poot_t。在APR當中則是Memory Pool Functions這個模組裡面，原始碼是memory\apr_pools.c，他主要功能其實也就是記憶體管理！

其他定義如下(ver 1.3)

Defines

• #define APR_POOL_DECLARE_ACCESSOR(type)
• #define APR_POOL_IMPLEMENT_ACCESSOR(type)
• #define APR_POOL_DEBUG 0
• #define APR_POOL__FILE_LINE__ __FILE__ ":" APR_STRINGIFY(__LINE__)

Typedefs

• typedef struct apr_pool_t apr_pool_t
• typedef int(* apr_abortfunc_t )(int retcode)

Functions

• apr_status_t apr_pool_initialize (void)
• void apr_pool_terminate (void)
• apr_status_t apr_pool_create_ex (apr_pool_t **newpool, apr_pool_t *parent, apr_abortfunc_t abort_fn, apr_allocator_t *allocator)
• apr_status_t apr_pool_create_core_ex (apr_pool_t **newpool, apr_abortfunc_t abort_fn, apr_allocator_t *allocator)
• apr_status_t apr_pool_create_unmanaged_ex (apr_pool_t **newpool, apr_abortfunc_t abort_fn, apr_allocator_t *allocator)
• apr_status_t apr_pool_create_ex_debug (apr_pool_t **newpool, apr_pool_t *parent, apr_abortfunc_t abort_fn, apr_allocator_t *allocator, const char *file_line)
• apr_status_t apr_pool_create_core_ex_debug (apr_pool_t **newpool, apr_abortfunc_t abort_fn, apr_allocator_t *allocator, const char *file_line)
• apr_status_t apr_pool_create_unmanaged_ex_debug (apr_pool_t **newpool, apr_abortfunc_t abort_fn, apr_allocator_t *allocator, const char *file_line)
• apr_status_t apr_pool_create (apr_pool_t **newpool, apr_pool_t *parent)
• apr_status_t apr_pool_create_core (apr_pool_t **newpool)
• apr_status_t apr_pool_create_unmanaged (apr_pool_t **newpool)
• apr_allocator_t * apr_pool_allocator_get (apr_pool_t *pool)
• void apr_pool_clear (apr_pool_t *p)
• void apr_pool_clear_debug (apr_pool_t *p, const char *file_line)
• void apr_pool_destroy (apr_pool_t *p)
• void apr_pool_destroy_debug (apr_pool_t *p, const char *file_line)
• void * apr_palloc (apr_pool_t *p, apr_size_t size)
• void * apr_palloc_debug (apr_pool_t *p, apr_size_t size, const char *file_line)
• void * apr_pcalloc (apr_pool_t *p, apr_size_t size)
• void * apr_pcalloc_debug (apr_pool_t *p, apr_size_t size, const char *file_line)
• void apr_pool_abort_set (apr_abortfunc_t abortfunc, apr_pool_t *pool)
• apr_abortfunc_t apr_pool_abort_get (apr_pool_t *pool)
• apr_pool_t * apr_pool_parent_get (apr_pool_t *pool)
• int apr_pool_is_ancestor (apr_pool_t *a, apr_pool_t *b)
• void apr_pool_tag (apr_pool_t *pool, const char *tag)
• apr_status_t apr_pool_userdata_set (const void *data, const char *key, apr_status_t(*cleanup)(void *), apr_pool_t *pool)
• apr_status_t apr_pool_userdata_setn (const void *data, const char *key, apr_status_t(*cleanup)(void *), apr_pool_t *pool)
• apr_status_t apr_pool_userdata_get (void **data, const char *key, apr_pool_t *pool)

Request in Apache請求處理的流程

我們必須先知道一件事，使用者的瀏覽器(browser)向Apache伺服器請求資料時，必須先建立TPC連線，依照HTTP通訊協定的規範向Apache伺服器請求資料。也就是說，Aapche處理請求的流程可以分為：建立TPC連線之前、建立TPC連線之後(開始用HTTP)，這兩個部分。

請求處理的流程

1. 接收請求(Accept Request)
   
2. 中繼資料處理(Metadata Processing)，request會傳遞很多個掛勾(hook)，呼叫所屬階段的程序(function)。
3. 產生回應內容(Content Generator)
4. 記錄(Logging)

正常來說，大致上是依據這個流程，如果是錯誤(error)或是產生新請求，則有可能使得request導向其他程序，另外，每個程序也可以讓其他程序呼叫，簡言之，有點複雜！(應該要畫圖解釋...)

處理的流程當中，Apache提供掛勾(hook)這項功能，可讓我們開發應用模組(module)的時候，利用回呼(callback)的方式值型模組中的功能。掛勾的總類依照請求的先後順序為：

1. ap_hook_post_read_request
   
2. ap_hook_translate_name
   
3. ap_hook_map_to_storage
   
4. ap_hook_header_parser
   
5. ap_hook_access_checker
   
6. ap_hook_check_user_id
   
7. ap_hook_auth_checker
   
8. ap_hook_type_checker
   
9. ap_hook_fixups
   
10. ap_hook_handler
    
11. ap_hook_log_transaction
    

最後，Apache預設處理請求的程式是掛在ap_hook_handler之中，其函數原形為static int default_handler(request_rec *r)，位於server\core.c檔案當中。

Apache Core Objects主要物件

Apache程式碼定義了一些結構可以使用，其中有四個結構相當重要，會在開發應用程式模組中使用到，主要是定義在httpd.h這個標頭檔中，其結構名稱有：

• request_rec每個請求的相關資訊
  
• server_rec每個主機的相關資訊
  
• conn_rec每次連線的相關資訊
  
• process_rec每個程序的相關資訊
  

詳細程式碼如下：

struct request_rec {

    /** The pool associated with the request */

    apr_pool_t *pool;

    /** The connection to the client */

    conn_rec *connection;

    /** The virtual host for this request */

    server_rec *server;

    /** Pointer to the redirected request if this is an external redirect */

    request_rec *next;

    /** Pointer to the previous request if this is an internal redirect */

    request_rec *prev;

    /** Pointer to the main request if this is a sub-request

     * (see http_request.h) */

    request_rec *main;

    /* Info about the request itself... we begin with stuff that only

     * protocol.c should ever touch...

     */

    /** First line of request */

    char *the_request;

    /** HTTP/0.9, "simple" request (e.g. GET /foo\n w/no headers) */

    int assbackwards;

    /** A proxy request (calculated during post_read_request/translate_name)

     *  possible values PROXYREQ_NONE, PROXYREQ_PROXY, PROXYREQ_REVERSE,

     *                  PROXYREQ_RESPONSE

     */

    int proxyreq;

    /** HEAD request, as opposed to GET */

    int header_only;

    /** Protocol string, as given to us, or HTTP/0.9 */

    char *protocol;

    /** Protocol version number of protocol; 1.1 = 1001 */

    int proto_num;

    /** Host, as set by full URI or Host: */

    const char *hostname;

    /** Time when the request started */

    apr_time_t request_time;

    /** Status line, if set by script */

    const char *status_line;

    /** Status line */

    int status;

    /* Request method, two ways; also, protocol, etc..  Outside of protocol.c,

     * look, but don't touch.

     */

    /** Request method (eg. GET, HEAD, POST, etc.) */

    const char *method;

    /** M_GET, M_POST, etc. */

    int method_number;

    /**

     *  'allowed' is a bitvector of the allowed methods.

     *

     *  A handler must ensure that the request method is one that

     *  it is capable of handling.  Generally modules should DECLINE

     *  any request methods they do not handle.  Prior to aborting the

     *  handler like this the handler should set r->allowed to the list

     *  of methods that it is willing to handle.  This bitvector is used

     *  to construct the "Allow:" header required for OPTIONS requests,

     *  and HTTP_METHOD_NOT_ALLOWED and HTTP_NOT_IMPLEMENTED status codes.

     *

     *  Since the default_handler deals with OPTIONS, all modules can

     *  usually decline to deal with OPTIONS.  TRACE is always allowed,

     *  modules don't need to set it explicitly.

     *

     *  Since the default_handler will always handle a GET, a

     *  module which does *not* implement GET should probably return

     *  HTTP_METHOD_NOT_ALLOWED.  Unfortunately this means that a Script GET

     *  handler can't be installed by mod_actions.

     */

    apr_int64_t allowed;

    /** Array of extension methods */

    apr_array_header_t *allowed_xmethods; 

    /** List of allowed methods */

    ap_method_list_t *allowed_methods; 

    /** byte count in stream is for body */

    apr_off_t sent_bodyct;

    /** body byte count, for easy access */

    apr_off_t bytes_sent;

    /** Last modified time of the requested resource */

    apr_time_t mtime;

    /* HTTP/1.1 connection-level features */

    /** sending chunked transfer-coding */

    int chunked;

    /** The Range: header */

    const char *range;

    /** The "real" content length */

    apr_off_t clength;

    /** Remaining bytes left to read from the request body */

    apr_off_t remaining;

    /** Number of bytes that have been read  from the request body */

    apr_off_t read_length;

    /** Method for reading the request body

     * (eg. REQUEST_CHUNKED_ERROR, REQUEST_NO_BODY,

     *  REQUEST_CHUNKED_DECHUNK, etc...) */

    int read_body;

    /** reading chunked transfer-coding */

    int read_chunked;

    /** is client waiting for a 100 response? */

    unsigned expecting_100;

    /* MIME header environments, in and out.  Also, an array containing

     * environment variables to be passed to subprocesses, so people can

     * write modules to add to that environment.

     *

     * The difference between headers_out and err_headers_out is that the

     * latter are printed even on error, and persist across internal redirects

     * (so the headers printed for ErrorDocument handlers will have them).

     *

     * The 'notes' apr_table_t is for notes from one module to another, with no

     * other set purpose in mind...

     */

    /** MIME header environment from the request */

    apr_table_t *headers_in;

    /** MIME header environment for the response */

    apr_table_t *headers_out;

    /** MIME header environment for the response, printed even on errors and

     * persist across internal redirects */

    apr_table_t *err_headers_out;

    /** Array of environment variables to be used for sub processes */

    apr_table_t *subprocess_env;

    /** Notes from one module to another */

    apr_table_t *notes;

    /* content_type, handler, content_encoding, and all content_languages 

     * MUST be lowercased strings.  They may be pointers to static strings;

     * they should not be modified in place.

     */

    /** The content-type for the current request */

    const char *content_type; /* Break these out --- we dispatch on 'em */

    /** The handler string that we use to call a handler function */

    const char *handler; /* What we *really* dispatch on */

    /** How to encode the data */

    const char *content_encoding;

    /** Array of strings representing the content languages */

    apr_array_header_t *content_languages;

    /** variant list validator (if negotiated) */

    char *vlist_validator;

    

    /** If an authentication check was made, this gets set to the user name. */

    char *user;

    /** If an authentication check was made, this gets set to the auth type. */

    char *ap_auth_type;

    /** This response can not be cached */

    int no_cache;

    /** There is no local copy of this response */

    int no_local_copy;

    /* What object is being requested (either directly, or via include

     * or content-negotiation mapping).

     */

    /** The URI without any parsing performed */

    char *unparsed_uri;

    /** The path portion of the URI */

    char *uri;

    /** The filename on disk corresponding to this response */

    char *filename;

    /* XXX: What does this mean? Please define "canonicalize" -aaron */

    /** The true filename, we canonicalize r->filename if these don't match */

    char *canonical_filename;

    /** The PATH_INFO extracted from this request */

    char *path_info;

    /** The QUERY_ARGS extracted from this request */

    char *args;

    /**  finfo.protection (st_mode) set to zero if no such file */

    apr_finfo_t finfo;

    /** A struct containing the components of URI */

    apr_uri_t parsed_uri;

    /**

     * Flag for the handler to accept or reject path_info on 

     * the current request.  All modules should respect the

     * AP_REQ_ACCEPT_PATH_INFO and AP_REQ_REJECT_PATH_INFO 

     * values, while AP_REQ_DEFAULT_PATH_INFO indicates they

     * may follow existing conventions.  This is set to the

     * user's preference upon HOOK_VERY_FIRST of the fixups.

     */

    int used_path_info;

    /* Various other config info which may change with .htaccess files

     * These are config vectors, with one void* pointer for each module

     * (the thing pointed to being the module's business).

     */

    /** Options set in config files, etc. */

    struct ap_conf_vector_t *per_dir_config;

    /** Notes on *this* request */

    struct ap_conf_vector_t *request_config;

    /**

     * A linked list of the .htaccess configuration directives

     * accessed by this request.

     * N.B. always add to the head of the list, _never_ to the end.

     * that way, a sub request's list can (temporarily) point to a parent's list

     */

    const struct htaccess_result *htaccess;

    /** A list of output filters to be used for this request */

    struct ap_filter_t *output_filters;

    /** A list of input filters to be used for this request */

    struct ap_filter_t *input_filters;

    /** A list of protocol level output filters to be used for this

     *  request */

    struct ap_filter_t *proto_output_filters;

    /** A list of protocol level input filters to be used for this

     *  request */

    struct ap_filter_t *proto_input_filters;

    /** A flag to determine if the eos bucket has been sent yet */

    int eos_sent;

/* Things placed at the end of the record to avoid breaking binary

 * compatibility.  It would be nice to remember to reorder the entire

 * record to improve 64bit alignment the next time we need to break

 * binary compatibility for some other reason.

 */

};

struct server_rec {

    /** The process this server is running in */這個伺服器正在執行的行程，也許是虛擬伺服器

    process_rec *process;

    /** The next server in the list */位於清單中，下一個伺服器組態

    server_rec *next;

    /** The name of the server */這個伺服器的名子

    const char *defn_name;

    /** The line of the config file that the server was defined on */組態檔httpd.conf中，定義伺服器的行號

    unsigned defn_line_number;

    /* Contact information */

    /** The admin's contact information */管理員的聯絡資訊(電子信箱)

    char *server_admin;

    /** The server hostname */這個伺服器的主機名稱

    char *server_hostname;

    /** for redirects, etc. */用於重新導向，埠號

    apr_port_t port;

    /* Log files --- note that transfer log is now in the modules... */

    /** The name of the error log */錯誤記錄檔的名稱

    char *error_fname;

    /** A file descriptor that references the error log */指標，指向錯誤記錄檔

    apr_file_t *error_log;

    /** The log level for this server */記錄檔記錄的層次

    int loglevel;

    /* Module-specific configuration for server, and defaults... */

    /** true if this is the virtual server */是否是虛擬伺服器

    int is_virtual;

    /** Config vector containing pointers to modules' per-server config 

     *  structures. */指向伺服器組態的向量

    struct ap_conf_vector_t *module_config; 

    /** MIME type info, etc., before we start checking per-directory info */MIME型態資訊

    struct ap_conf_vector_t *lookup_defaults;

    /* Transaction handling */

    /** I haven't got a clue */還不知道這是什麼！

    server_addr_rec *addrs;

    /** Timeout, as an apr interval, before we give up */逾時時間

    apr_interval_time_t timeout;

    /** The apr interval we will wait for another request */持續連線的逾時時間

    apr_interval_time_t keep_alive_timeout;

    /** Maximum requests per connection */每個連線的最大請求數目

    int keep_alive_max;

    /** Use persistent connections? */是否使用http的持續連線

    int keep_alive;

    /** Pathname for ServerPath */伺服器路徑的名稱

    const char *path;

    /** Length of path */路徑長度

    int pathlen;

    /** Normal names for ServerAlias servers */別名伺服器的正常名稱

    apr_array_header_t *names;

    /** Wildcarded names for ServerAlias servers */別名伺服器的萬用名稱

    apr_array_header_t *wild_names;

    /** limit on size of the HTTP request line    */請求的行數限制

    int limit_req_line;

    /** limit on size of any request header field */請求標頭欄位大小的限制

    int limit_req_fieldsize;

    /** limit on number of request header fields  */請求標頭欄位數目的限制

    int limit_req_fields; 

    /** The server request scheme for redirect responses */重新導向的請求方法

    const char *server_scheme;

};

struct conn_rec {

    /** Pool associated with this connection */

    apr_pool_t *pool;

    /** Physical vhost this conn came in on */

    server_rec *base_server;

    /** used by http_vhost.c */

    void *vhost_lookup_data;

    /* Information about the connection itself */

    /** local address */

    apr_sockaddr_t *local_addr;

    /** remote address */

    apr_sockaddr_t *remote_addr;

    /** Client's IP address */

    char *remote_ip;

    /** Client's DNS name, if known.  NULL if DNS hasn't been checked,

     *  "" if it has and no address was found.  N.B. Only access this though

     * get_remote_host() */

    char *remote_host;

    /** Only ever set if doing rfc1413 lookups.  N.B. Only access this through

     *  get_remote_logname() */

    char *remote_logname;

    /** Are we still talking? */

    unsigned aborted:1;

    /** Are we going to keep the connection alive for another request?

     * @see ap_conn_keepalive_e */

    ap_conn_keepalive_e keepalive;

    /** have we done double-reverse DNS? -1 yes/failure, 0 not yet, 

     *  1 yes/success */

    signed int double_reverse:2;

    /** How many times have we used it? */

    int keepalives;

    /** server IP address */

    char *local_ip;

    /** used for ap_get_server_name when UseCanonicalName is set to DNS

     *  (ignores setting of HostnameLookups) */

    char *local_host;

    /** ID of this connection; unique at any point in time */

    long id; 

    /** Config vector containing pointers to connections per-server

     *  config structures. */

    struct ap_conf_vector_t *conn_config;

    /** Notes on *this* connection: send note from one module to

     *  another. must remain valid for all requests on this conn */

    apr_table_t *notes;

    /** A list of input filters to be used for this connection */

    struct ap_filter_t *input_filters;

    /** A list of output filters to be used for this connection */

    struct ap_filter_t *output_filters;

    /** handle to scoreboard information for this connection */

    void *sbh;

    /** The bucket allocator to use for all bucket/brigade creations */

    struct apr_bucket_alloc_t *bucket_alloc;

    /** The current state of this connection */

    conn_state_t *cs;

    /** Is there data pending in the input filters? */ 

    int data_in_input_filters;

    

    /** Are there any filters that clogg/buffer the input stream, breaking

     *  the event mpm.

     */

    int clogging_input_filters;

};

struct process_rec {

    /** Global pool. Cleared upon normal exit */

    apr_pool_t *pool;

    /** Configuration pool. Cleared upon restart */

    apr_pool_t *pconf;

    /** Number of command line arguments passed to the program */

    int argc;

    /** The command line arguments */

    const char * const *argv;

    /** The program name used to execute the program */

    const char *short_name;

};

Apache Multi-Processing Modules多行程模組

MPM是Apache中的一個模組，全名為Multi-Processing Module多行程模組，這個模組是介於APR和底層的作業系統(Operation System, OS)的一個模組，其目的是用來最佳化，使得Apache在各式的OS上能有傑出的效能表現。

MPM本身是系統層級(system-level)的模組，意思就是說，開發上和其他模組會不一樣，哪裡不一樣？目前還在研究當中。不過，我們大多數的應用只會按照一般模組的樣子來開發，絕大多數都是這種開發方式(即非系統層級)。

早期Apache發展在UNIX之類的作業系統上，這類系統多為多處理器的伺服器，而作業系統會依照請求(request)分岔(fork)產生其他行程(process)來處理，一般來說，Apache啟動的時候，會產生一個集散區(pool，也有稱為"池")來管理這些行程。之後2.0版本，為了能夠跨平台(cross-platform)的高效率、最佳化運作，所以才設計出MPM這個模組。

實務應用上來說，只有UNIX之類的作業系統有較多的設定可供選擇使用。我目前是以windows開發，真的選項就只有兩個可以設定：MaxRequestsPerChild和ThreadsPerChild兩個指令。其他指令請參考http://httpd.apache.org/docs/2.2/mod/mpm_common.html

在MPM這個部分當中，因為跟底層OS有密切關係，其實MPM有好幾個類別，有些是給UNIX用的，有些是給Windows(WIN32)用的，有些則給其他OS用的，所以再看這些指令的時候，必須注意它所依賴的模組，否則社定會失敗。另一個問題是，我們可以選擇使用哪個MPM嗎？應該是不行，除非你自己重新complie一次，MPM的設定在編譯Apache時就已經決定了！

以UNIX這群的MPM來說，有Prefork MPM、Worker MPM以及正在試驗的Event MPM。

• Prefork MPM是非執行緒的架構，和以往1版的類似。
  
• Worker MPM是多執行緒的架構，其執行緒和每個請求連線有關。
  
• Event MPM是正在試驗當中，將執行緒與請求連線的關係解耦合，可以避免請求連線超出執行緒的問題。
  

結論：開發Apache的模組、應用程式來說，不需要理會MPM，請忽略MPM這個模組！

我目前以windows開發，所以主要是專注於mpm_winnt，其指令Directives有

• CoreDumpDirectory
  
• Listen
  
• ListenBacklog
  
• MaxMemFree
  
• MaxRequestsPerChild
  
• PidFile
  
• ReceiveBufferSize
  
• ScoreBoardFile
  
• SendBufferSize
  
• ThreadLimit
  
• ThreadsPerChild
  
• ThreadStackSize
  
• Win32DisableAcceptEx
  

Deconstruction of Apache Code解構阿帕契程式碼

太久，真的是一個月沒有更新部落格。忙著，也跟著茫了。

目前正在使用Apache Http Server的程式撰寫。我想之後會寫有關於Apache的程式碼內容，希望有這個能力和空閒時間。(忙著活下去，還有時間嗎！好問題~)

先講參考書籍，討論Apache的書籍很少(手指頭的數目之內)，中文可以參考下面這本：

Ben Laurie & Peter Laurie原著，郭文生編譯，Apache技術手冊(第三版)，台北：歐萊禮，2003。

怎麼開始呢？Deconstruction，解構。應該不稱之為解構主義，一方面我現在不了解何謂解構主義，只是想將Apache的Code猜(拆)解開來看看，這套軟體的運作方式？我們有沒有能力做出這樣的軟體呢？需要哪些能力和人力？如何設計更是大問題！(我的疑問)

Apache的進入點是main.c，在httpd這個專案(Project)裡面，程式碼架構如下：

用到的標頭檔，Apache底層是Apache Portability Runtime(APR)，因此會引入許多apr_的header file，而基於APR建置的程式庫則有APR-util，簡稱APU，以及實現iconv()的APR-iconv。另外，http_是關於http protocol的一些處理，也需要因引入。

#include "apr.h" //平台定義

#include "apr_strings.h" //字串處理程式庫

#include "apr_getopt.h" //命令參數

#include "apr_general.h" //混雜的程式庫，未歸類的

#include "apr_lib.h" //通用程式庫

#include "apr_md5.h" //md5的程式庫，屬於APU

#include "apr_time.h" //時間程式庫

#include "apr_version.h" //APR的版本介面

#include "apu_version.h" //APU的版本介面

#define APR_WANT_STDIO

#define APR_WANT_STRFUNC

#include "apr_want.h" //APR對於標準標頭檔支援

#define CORE_PRIVATE

#include "ap_config.h" //定義掛勾hook的標頭檔

#include "httpd.h" //主要基本的定義

#include "http_main.h" //main程式的定義

#include "http_log.h" //記錄檔的定義

#include "http_config.h" //一些設定的定義

#include "http_core.h" //核心模組的定義

#include "http_vhost.h" //虛擬主機的定義

#include "apr_uri.h" //uri的程式庫，屬於APU

#include "util_ebcdic.h" //定義Extended Binary Coded Decimal Interchange Code(EBCDIC)的轉換功能

#include "ap_mpm.h" //定義Multi-Processing Module(MPM)，多處理模組

#include "mpm_common.h" //MPM用的程式庫

定義到的函數(function)：

static void show_mpm_settings(void)

static void show_compile_settings(void)

static void destroy_and_exit_process(process_rec *process, int process_exit_value)

static process_rec *init_process(int *argc, const char * const * *argv)

static void usage(process_rec *process)

主程式碼：

int main(int argc, const char * const argv[])
{
    char c;
    int configtestonly = 0;
    const char *confname = SERVER_CONFIG_FILE;
    const char *def_server_root = HTTPD_ROOT;
    const char *temp_error_log = NULL;
    const char *error;
    process_rec *process;
    server_rec *server_conf;
    apr_pool_t *pglobal;
    apr_pool_t *pconf;
    apr_pool_t *plog; /* Pool of log streams, reset _after_ each read of conf */
    apr_pool_t *ptemp; /* Pool for temporary config stuff, reset often */
    apr_pool_t *pcommands; /* Pool for -D, -C and -c switches */
    apr_getopt_t *opt;
    apr_status_t rv;
    module **mod;
    const char *optarg;
    APR_OPTIONAL_FN_TYPE(ap_signal_server) *signal_server;

    AP_MONCONTROL(0); /* turn off profiling of startup */

    process = init_process(&argc, &argv);
    pglobal = process->pool;
    pconf = process->pconf;
    ap_server_argv0 = process->short_name;

#if APR_CHARSET_EBCDIC
    if (ap_init_ebcdic(pglobal) != APR_SUCCESS) {
        destroy_and_exit_process(process, 1);
    }
#endif

    apr_pool_create(&pcommands, pglobal);
    apr_pool_tag(pcommands, "pcommands");
    ap_server_pre_read_config  = apr_array_make(pcommands, 1, sizeof(char *));
    ap_server_post_read_config = apr_array_make(pcommands, 1, sizeof(char *));
    ap_server_config_defines   = apr_array_make(pcommands, 1, sizeof(char *));

    error = ap_setup_prelinked_modules(process);
    if (error) {
        ap_log_error(APLOG_MARK, APLOG_STARTUP|APLOG_EMERG, 0, NULL, "%s: %s",
                     ap_server_argv0, error);
        destroy_and_exit_process(process, 1);
    }

    ap_run_rewrite_args(process);

    /* Maintain AP_SERVER_BASEARGS list in http_main.h to allow the MPM
     * to safely pass on our args from its rewrite_args() handler.
     */
    apr_getopt_init(&opt, pcommands, process->argc, process->argv);

    while ((rv = apr_getopt(opt, AP_SERVER_BASEARGS, &c, &optarg))
            == APR_SUCCESS) {
        char **new;

        switch (c) {
        case 'c':
            new = (char **)apr_array_push(ap_server_post_read_config);
            *new = apr_pstrdup(pcommands, optarg);
            break;

        case 'C':
            new = (char **)apr_array_push(ap_server_pre_read_config);
            *new = apr_pstrdup(pcommands, optarg);
            break;

        case 'd':
            def_server_root = optarg;
            break;

        case 'D':
            new = (char **)apr_array_push(ap_server_config_defines);
            *new = apr_pstrdup(pcommands, optarg);
            /* Setting -D DUMP_VHOSTS is equivalent to setting -S */
            if (strcmp(optarg, "DUMP_VHOSTS") == 0)
                configtestonly = 1;
            /* Setting -D DUMP_MODULES is equivalent to setting -M */
            if (strcmp(optarg, "DUMP_MODULES") == 0)
                configtestonly = 1;
            break;

        case 'e':
            if (strcasecmp(optarg, "emerg") == 0) {
                ap_default_loglevel = APLOG_EMERG;
            }
            else if (strcasecmp(optarg, "alert") == 0) {
                ap_default_loglevel = APLOG_ALERT;
            }
            else if (strcasecmp(optarg, "crit") == 0) {
                ap_default_loglevel = APLOG_CRIT;
            }
            else if (strncasecmp(optarg, "err", 3) == 0) {
                ap_default_loglevel = APLOG_ERR;
            }
            else if (strncasecmp(optarg, "warn", 4) == 0) {
                ap_default_loglevel = APLOG_WARNING;
            }
            else if (strcasecmp(optarg, "notice") == 0) {
                ap_default_loglevel = APLOG_NOTICE;
            }
            else if (strcasecmp(optarg, "info") == 0) {
                ap_default_loglevel = APLOG_INFO;
            }
            else if (strcasecmp(optarg, "debug") == 0) {
                ap_default_loglevel = APLOG_DEBUG;
            }
            else {
                usage(process);
            }
            break;

        case 'E':
            temp_error_log = apr_pstrdup(process->pool, optarg);
            break;

        case 'X':
            new = (char **)apr_array_push(ap_server_config_defines);
            *new = "DEBUG";
            break;

        case 'f':
            confname = optarg;
            break;

        case 'v':
            printf("Server version: %s\n", ap_get_server_description());
            printf("Server built:   %s\n", ap_get_server_built());
            destroy_and_exit_process(process, 0);

        case 'V':
            show_compile_settings();
            destroy_and_exit_process(process, 0);

        case 'l':
            ap_show_modules();
            destroy_and_exit_process(process, 0);

        case 'L':
            ap_show_directives();
            destroy_and_exit_process(process, 0);

        case 't':
            configtestonly = 1;
            break;

        case 'S':
            configtestonly = 1;
            new = (char **)apr_array_push(ap_server_config_defines);
            *new = "DUMP_VHOSTS";
            break;

        case 'M':
            configtestonly = 1;
            new = (char **)apr_array_push(ap_server_config_defines);
            *new = "DUMP_MODULES";
            break;

        case 'h':
        case '?':
            usage(process);
        }
    }

    /* bad cmdline option?  then we die */
    if (rv != APR_EOF || opt->ind < opt->argc) {
        usage(process);
    }

    apr_pool_create(&plog, pglobal);
    apr_pool_tag(plog, "plog");
    apr_pool_create(&ptemp, pconf);
    apr_pool_tag(ptemp, "ptemp");

    /* Note that we preflight the config file once
     * before reading it _again_ in the main loop.
     * This allows things, log files configuration
     * for example, to settle down.
     */

    ap_server_root = def_server_root;
    if (temp_error_log) {
        ap_replace_stderr_log(process->pool, temp_error_log);
    }
    server_conf = ap_read_config(process, ptemp, confname, &ap_conftree);
    if (!server_conf) {
        destroy_and_exit_process(process, 1);
    }

    if (ap_run_pre_config(pconf, plog, ptemp) != OK) {
        ap_log_error(APLOG_MARK, APLOG_STARTUP |APLOG_ERR, 0,
                     NULL, "Pre-configuration failed");
        destroy_and_exit_process(process, 1);
    }

    rv = ap_process_config_tree(server_conf, ap_conftree,
                                process->pconf, ptemp);
    if (rv == OK) {
        ap_fixup_virtual_hosts(pconf, server_conf);
        ap_fini_vhost_config(pconf, server_conf);
        apr_hook_sort_all();

        if (configtestonly) {
            ap_run_test_config(pconf, server_conf);
            ap_log_error(APLOG_MARK, APLOG_STARTUP, 0, NULL, "Syntax OK");
            destroy_and_exit_process(process, 0);
        }
    }

    signal_server = APR_RETRIEVE_OPTIONAL_FN(ap_signal_server);
    if (signal_server) {
        int exit_status;

        if (signal_server(&exit_status, pconf) != 0) {
            destroy_and_exit_process(process, exit_status);
        }
    }

    /* If our config failed, deal with that here. */
    if (rv != OK) {
        destroy_and_exit_process(process, 1);
    }

    apr_pool_clear(plog);

    if ( ap_run_open_logs(pconf, plog, ptemp, server_conf) != OK) {
        ap_log_error(APLOG_MARK, APLOG_STARTUP |APLOG_ERR,
                     0, NULL, "Unable to open logs");
        destroy_and_exit_process(process, 1);
    }

    if ( ap_run_post_config(pconf, plog, ptemp, server_conf) != OK) {
        ap_log_error(APLOG_MARK, APLOG_STARTUP |APLOG_ERR, 0,
                     NULL, "Configuration Failed");
        destroy_and_exit_process(process, 1);
    }

    apr_pool_destroy(ptemp);

    for (;;) {
        apr_hook_deregister_all();
        apr_pool_clear(pconf);

        for (mod = ap_prelinked_modules; *mod != NULL; mod++) {
            ap_register_hooks(*mod, pconf);
        }

        /* This is a hack until we finish the code so that it only reads
         * the config file once and just operates on the tree already in
         * memory.  rbb
         */
        ap_conftree = NULL;
        apr_pool_create(&ptemp, pconf);
        apr_pool_tag(ptemp, "ptemp");
        ap_server_root = def_server_root;
        server_conf = ap_read_config(process, ptemp, confname, &ap_conftree);
        if (!server_conf) {
            destroy_and_exit_process(process, 1);
        }

        if (ap_run_pre_config(pconf, plog, ptemp) != OK) {
            ap_log_error(APLOG_MARK, APLOG_STARTUP |APLOG_ERR,
                         0, NULL, "Pre-configuration failed");
            destroy_and_exit_process(process, 1);
        }

        if (ap_process_config_tree(server_conf, ap_conftree, process->pconf,
                                   ptemp) != OK) {
            destroy_and_exit_process(process, 1);
        }
        ap_fixup_virtual_hosts(pconf, server_conf);
        ap_fini_vhost_config(pconf, server_conf);
        apr_hook_sort_all();
        apr_pool_clear(plog);
        if (ap_run_open_logs(pconf, plog, ptemp, server_conf) != OK) {
            ap_log_error(APLOG_MARK, APLOG_STARTUP |APLOG_ERR,
                         0, NULL, "Unable to open logs");
            destroy_and_exit_process(process, 1);
        }

        if (ap_run_post_config(pconf, plog, ptemp, server_conf) != OK) {
            ap_log_error(APLOG_MARK, APLOG_STARTUP |APLOG_ERR,
                         0, NULL, "Configuration Failed");
            destroy_and_exit_process(process, 1);
        }

        apr_pool_destroy(ptemp);
        apr_pool_lock(pconf, 1);

        ap_run_optional_fn_retrieve();

        if (ap_mpm_run(pconf, plog, server_conf))
            break;

        apr_pool_lock(pconf, 0);
    }

    apr_pool_lock(pconf, 0);
    destroy_and_exit_process(process, 0);

    return 0; /* Termination 'ok' */
}