1Apache::DBI::Cache(3) User Contributed Perl DocumentationApache::DBI::Cache(3)
2
6 Apache::DBI::Cache - a DBI connection cache
7
9 use Apache::DBI::Cache debug=>3, bdb_env=>'/tmp/tmp',
10 plugin=>'Apache::DBI::Cache::mysql',
11 plugin=>['DBM', sub {...}, sub {...}],
12 ...;
13
15 This module is an alternative to the famous Apache::DBI module. As
16 Apache::DBI it provides persistent DBI connections.
17 It can be used with mod_perl1, mod_perl2 and even standalone.
19 WHY ANOTHER MODULE FOR THE SAME?
21 Apache::DBI has a number of limitations. Firstly, it is not possible to
22 get multiple connections with the same parameters. A common scenario
23 for example is to use one connection to perform transactions and
24 another to perform simple lookups in the same database. With
25 Apache::DBI it is very likely to get the same connection if you mean to
26 use different.
27 With Apache::DBI all connections are reset at end of a request.
29 Apache::DBI does not regard database specific functions to cache
31 handles more aggressively. For example a mysql DSN can look like
32 dbi:mysql:test:localhost:3306
34 or
36 dbi:mysql:host=localhost;db=test
38 Both point to the same database but for Apache::DBI they are different.
40 Apache::DBI::Cache recognizes these two by means of a mysql plugin.
41 The plugin even recognizes connections to different databases on the
43 same mysql server as the same connection and issues a "USE database"
44 command before returning the actual handle to the user. Hence, with
45 Apache::DBI::Cache many the overall number of connections to a DB
46 server can be dramatically reduced.
47 HOW DOES IT WORK?
49 To decide whether to use Apache::DBI::Cache or not it is essential to
50 know how it works. As with Apache::DBI Apache::DBI::Cache uses a hook
51 provided by the DBI module to intercept "DBI->connect()" calls. Also
52 do Apache::DBI::Cache maintain a cache of active handles.
53 When a new connection is requested and the cache is empty a new
55 connection is established and returned to the user. At this point it is
56 not cached at all. Here Apache::DBI::Cache differs from Apache::DBI.
57 Later either "disconnect" is called on the handle or it simply goes out
59 of scope and the garbage collector calls a "DESTROY" method if
60 provided. Both events are intercepted by Apache::DBI::Cache. Only then
61 the handle is put in the cache.
62 This means a handle is never really disconnected. "$dbh->{Active}" will
64 always return true no matter how often "disconnect" is called.
65 Further, you can prevent a handle from getting reused by simply not
66 forgetting it.
67 USAGE
69 Different to Apache::DBI Apache::DBI::Cache must be "use"ed not
70 "require"ed. That means it's "import" function must be called.
71 When used with mod_perl (versions 1.x or 2.x) this is best done in a
73 "startup.pl" or in a "<Perl>" section in the "httpd.conf". See
74 mod_perl documentation for more information.
75 Thereafter, "DBI->connect" is called as usual. No special treatment is
77 needed.
78 When Apache::Status or Apache2::Status is used Apache::DBI::Cache
80 provides an extra menu item to show statistics on handles. The loading
81 order of the Apache::Status and Apache::DBI::Cache is irrelevant.
82 FUNCTIONS
84 import - "use" parameter
85 Parameters to the "use" statement are given in a "key => value"
86 fashion.
87 use Apache::DBI::Cache debug=>3, logger=>sub {...},
89 plugin=>['driver', sub {}, sub {}],
90 plugin=>'Apache::DBI::Cache::mysql',
91 use_bdb=>1,
92 bdb_env=>'/tmp/mybdbenv',
93 bdb_memcache=>20*1024,
94 ...;
95 · plugin
97 loads a plugin, see also "PLUGINS" below. The plugin can be
99 specified as a 3-element array or by name. In the second case the
100 "import" simply "use"s the module. This option can be given
101 multiple times.
102 · use_bdb, bdb_env and bdb_memcache
104 Apache::DBI::Cache can use BerleleyDB as a shared memory
106 implementation to maintain statistics for a group of processes
107 instead of a single process.
108 "use_bdb" specify whether to use BerkeleyDB or not. If ommitted
110 Apache::DBI::Cache will try to load and use BerkeleyDB. If that
111 fails it silently provides per process statistics. If "use_bdb"
112 is true Apache::DBI::Cache dies if it cannot use BerkeleyDB. If
113 "use_bdb" is false per process statistics are maintained and
114 BerkeleyDB is not used.
115 "bdb_env" specifies a path to a directory where BerkeleyDB can
117 put it's temporary files. If omitted "/tmp/Apache::DBI::Cache" is
118 used. The parent directory of this directory must exists and be
119 writeable.
120 "bdb_memcache" specifies the size of the shared memory segment
122 that is allocated by BerkeleyDB. Depending on the number of
123 handles in your configuration a few kilobytes are enough. If
124 omitted 20 kB are used.
125 "bdb_env" and "bdb_memcache" can also be specified by the
127 "APACHE_DBI_CACHE_ENVPATH" and "APACHE_DBI_CACHE_CACHESIZE"
128 environment variables.
129 · debug
131 set a debug level. Under mod_perl this is almost irrelevant, see
133 "logger" below.
134 · logger
136 here a logger function can be specified. It is called with the
138 message verbosity level as the first parameter. The remaining
139 parameters are concatenated to build the actual message.
140 Currently there are 2 verbosity levels used 1 and 2. 0 is
142 reserved for real errors. 1 mentions that the module has been
143 initialized. 2 rattle off normal processing messages.
144 Apache::DBI::Cache provides 2 logger functions. One is controlled
146 by the "debug" level setting (see above). A message is printed to
147 STDERR if it's level is equal or greater the current debug level.
148 The other logger is used when running under mod_perl. It is
150 mainly controlled by the Apache "LogLevel" setting. Messages at
151 level 0 are printed as "$log->error", level 1 as "$log->info" and
152 level 2 as "$log->debug". For level 2 messages additionally the
153 current debug level is checked to be greater or equal 2.
154 · delimiter
156 Here the internal key delimiter can be changed. It defaults to
158 "\1". Changing it is necessary only when your DSN, username or
159 password contain it or to provide more readable debugging
160 messages.
161 statistics
163 returns a reference to the statistics hash. If BerkeleyDB is used
164 it is tied to BerkeleyDB::Btree.
165 statistics_as_html
167 returns a reference to an array of HTML fragments. If mod_perl and
168 Apache::Status or Apache::Status2 is used the output of this
169 function is shown under http://HOST/STATUS/URI?DBI_conn.
170 plugin( 'name', \&mangle, \&setup )
172 installs a new plugin, see "PLUGINS" below. If a plugin for the
173 specified database type was already installed it is returned as a
174 2-element list:
175 ($old_mangle, $old_setup)=
177 plugin( 'name', \&new_mangle, \&new_setup );
178 If called with an name only the current plugin is returned:
180 ($old_mangle, $old_setup)=plugin( 'name' );
182 To delete a plugin call
184 ($old_mangle, $old_setup)=plugin( 'name', undef, undef );
186 connect_on_init
188 call this function multiple times with parameters you would pass to
189 "DBI->connect" before calling "Apache::DBI::Cache::init", i.e. in
190 your "startup.pl". Then "init" will establish all these
191 connections.
192 init
194 This function is called once per child process to initialize
195 Apache::DBI::Cache. If mod_perl is used this is done automatically
196 in a PerlChildInitHandler
197 finish
199 This function must be called before a process is going to
200 terminate. Under mod_perl it is automatically called in a
201 PerlChildExitHandler.
202 As of version 0.08 calling this function is optional.
204 undef_at_request_cleanup( \$dbh1, \$dbh2, ... )
206 When an application uses global variables to store handles they
207 probably won't be reused because a global variable is ..., well
208 global. This can be fixed by explicitly undefining these handles at
209 request cleanup or by using this function. It simply collects all
210 handle references passed in between 2 calls to
211 "Apache::DBI::Cache::request_cleanup". When
212 "Apache::DBI::Cache::request_cleanup" is called all these handles
213 are undefined. The first call to this function during a request
214 cycle installs "Apache::DBI::Cache::request_cleanup" as
215 PerlCleanupHandler.
216 With mod_perl2 this requires the PerlOption "GlobalRequest" to be
218 set:
219 PerlOption +GlobalRequest
221 in your httpd.conf.
223 request_cleanup
225 This is the PerlCleanupHandler. If Apache::DBI::Cache is used
226 standalone the application can call it from time to time.
227 EXPORT
229 Nothing.
230
232 For a module like Apache::DBI::Cache it is complicated to cope with DBI
233 subclasses. There are 2 problems to solve. First, make sure that our
234 "disconnect" and "DESTROY" methods are called instead of the original.
235 Apache::DBI::Cache solves this problem by inserting its own methods
236 into the foreign class.
237 Hence, if a subclass provides "disconnect" and "DESTROY" methods they
239 will never be called. This is ugly but works in most cases.
240 To insert our methods into the subclass we need to know its name. This
242 is the second problem. To create a subclassed DBI handle one calls
243 either
244 DBI->connect( $dsn, $user, $passwd, {RootClass=>SUBCLASS} );
246 or
248 SUBCLASS->connect( $dsn, $user, $passwd, {} );
250 The first case is simple since the attribute hash is passed to our
252 connect method. Unfortunately, the second case is not simple since
253 "SUBCLASS" is known only by "DBI::connect". This is solved by searching
254 the current call stack for the DBI::connect call. Then we use its first
255 parameter.
256 That works for me but is even uglier. If you encounter problems don't
258 hesitate to mail me.
259
261 To make "Class::DBI" or "Ima::DBI" work with "Apache::DBI::Cache" see
262 Apache::DBI::Cache::ImaDBI.
263
265 Plugins are used to modify the caching for certain database types. They
266 can change the caching key, issue database commands just before a
267 handle is returned to the user or prevent handle caching entirely for a
268 database type.
269 There can only be one plugin per database type at a time.
271 A plugin registers itself by calling "Apache::DBI::Cache::plugin"
273 passing 3 parameters. The first parameter is simply the name of the
274 database type. It matches the DBI driver name. Thus, a MySQL plugin
275 passes the string "mysql" since the corresponding DBI driver is named
276 "DBD::mysql". Whereas a PostgreSQL plugin passes either "Pg" or "PgPP"
277 depending on the driver actually used.
278 The 2nd and 3rd parameters are CODE references that are called just
280 before a connection is chosen from the cache or newly established and
281 after the connection is made just before it is returned to the caller.
282 The first function can mangle the connection parameters the second
283 perform additional setup steps. Further, I will call them mangle and
284 setup.
285 Thus, a plugin is registered this way:
287 Apache::DBI::Cache::plugin('Name', \&mangle, \&setup);
289 Normally, it is implemented as a separate module according to the
291 following template, see Apache::DBI::Cache::mysql for example:
292 package Apache::DBI::Cache::DRIVER;
294 use strict;
296 BEGIN {
298 die "Please load Apache::DBI::Cache before"
299 unless defined &Apache::DBI::Cache::plugin;
300 ...
302 Apache::DBI::Cache::plugin
304 (
305 'DRIVER',
306 sub {},
307 sub {}
308 );
309 }
310 1;
312 Calling Conventions
314 · mangle
315 ($dsn, $user, $passwd, $attr, $ctx, $nocache)=
317 mangle($dsn, $user, $passwd, $attr);
318 mangle is called with almost the same parameters as the original call
320 to "DBI->connect". The "dbi:DRIVER:" prefix is stripped from the DSN.
321 It is expected to return similar values plus an arbitrary context
322 that is later passed to setup and an optional "nocache" flag.
323 If $nocache is true or mangle returns an empty list a new connection
325 is made and the handle is directly passed to the caller without
326 further processing. Also setup will not be called. Such a handle will
327 not be cached on "disconnect" or "DESTROY".
328 mangle can change all parameters. The MySQL plugin for example
330 deletes the actual database name from the DSN, reformats it according
331 to a standard format and adds the standard port if it is omitted. The
332 database is put in the context.
333 · setup
335 $rc=setup($dbh, $dsn, $user, $passwd, $attr, $ctx);
337 The setup function performs additional setup steps on $dbh. The MySQL
339 plugin for example issues a "USE database" command using the database
340 from the context.
341 A connection is considered dead if setup returns false.
343
345 · periodically ping all cached handles
346 · correct statistics when a process is finishes without calling
348 "finish()"
349 · redirect BerkeleyDB errors to logger
351
353 Apache::DBI::Cache::mysql
354 Apache::DBI
355
357 Torsten Foertsch, <torsten.foertsch@gmx.net>
358 With suggestions from
360 Andreas Nolte < andreas dot nolte at bertelsmann dot de >
362 Dietmar Hanisch < dietmar dot hanisch at bertelsmann dot de > and
364 Ewald Hinrichs < ewald dot hinrichs at bertelsmann dot de >
366
368 Sincere thanks to Arvato Direct Services (http://www.arvato.com/) for
369 sponsoring this module and providing a test platform with several
370 thousand DBI connections.
371
373 Copyright (C) 2005-2006 by Torsten Foertsch
374 This library is free software; you can redistribute it and/or modify it
376 under the same terms as Perl itself.
377perl v5.30.1 2020-01-29 Apache::DBI::Cache(3)