1DBIx::Class::Storage::DUBsIeD:rB:IRCxeo:pn:ltCirlciaabstuset:de::dS:tIPonertrarlgoedD:uo:ccDtuBimIoe:nn:(tR3ae)tpiloincated::Introduction(3)
2
6 DBIx::Class::Storage::DBI::Replicated::Introduction - Minimum Need to
7 Know
8
10 This is an introductory document for
11 DBIx::Class::Storage::DBI::Replicated.
12 This document is not an overview of what replication is or why you
14 should be using it. It is not a document explaining how to setup MySQL
15 native replication either. Copious external resources are available for
16 both. This document presumes you have the basics down.
17
19 DBIx::Class supports a framework for using database replication. This
20 system is integrated completely, which means once it's setup you should
21 be able to automatically just start using a replication cluster without
22 additional work or changes to your code. Some caveats apply, primarily
23 related to the proper use of transactions (you are wrapping all your
24 database modifying statements inside a transaction, right ;) ) however
25 in our experience properly written DBIC will work transparently with
26 Replicated storage.
27 Currently we have support for MySQL native replication, which is
29 relatively easy to install and configure. We also currently support
30 single master to one or more replicants (also called 'slaves' in some
31 documentation). However the framework is not specifically tied to the
32 MySQL framework and supporting other replication systems or
33 topographies should be possible. Please bring your patches and ideas
34 to the #dbix-class IRC channel or the mailing list.
35 For an easy way to start playing with MySQL native replication, see:
37 MySQL::Sandbox.
38 If you are using this with a Catalyst based application, you may also
40 want to see more recent updates to Catalyst::Model::DBIC::Schema, which
41 has support for replication configuration options as well.
42
44 By default, when you start DBIx::Class, your Schema
45 (DBIx::Class::Schema) is assigned a storage_type, which when fully
46 connected will reflect your underlying storage engine as defined by
47 your chosen database driver. For example, if you connect to a MySQL
48 database, your storage_type will be DBIx::Class::Storage::DBI::mysql
49 Your storage type class will contain database specific code to help
50 smooth over the differences between databases and let DBIx::Class do
51 its thing.
52 If you want to use replication, you will override this setting so that
54 the replicated storage engine will 'wrap' your underlying storages and
55 present a unified interface to the end programmer. This wrapper
56 storage class will delegate method calls to either a master database or
57 one or more replicated databases based on if they are read only (by
58 default sent to the replicants) or write (reserved for the master).
59 Additionally, the Replicated storage will monitor the health of your
60 replicants and automatically drop them should one exceed configurable
61 parameters. Later, it can automatically restore a replicant when its
62 health is restored.
63 This gives you a very robust system, since you can add or drop
65 replicants and DBIC will automatically adjust itself accordingly.
66 Additionally, if you need high data integrity, such as when you are
68 executing a transaction, replicated storage will automatically delegate
69 all database traffic to the master storage. There are several ways to
70 enable this high integrity mode, but wrapping your statements inside a
71 transaction is the easy and canonical option.
72
74 A replicated storage contains several parts. First, there is the
75 replicated storage itself (DBIx::Class::Storage::DBI::Replicated). A
76 replicated storage takes a pool of replicants
77 (DBIx::Class::Storage::DBI::Replicated::Pool) and a software balancer
78 (DBIx::Class::Storage::DBI::Replicated::Balancer). The balancer does
79 the job of splitting up all the read traffic amongst the replicants in
80 the Pool. Currently there are two types of balancers, a Random one
81 which chooses a Replicant in the Pool using a naive randomizer
82 algorithm, and a First replicant, which just uses the first one in the
83 Pool (and obviously is only of value when you have a single replicant).
84
86 All the parts of replication can be altered dynamically at runtime,
87 which makes it possibly to create a system that automatically scales
88 under load by creating more replicants as needed, perhaps using a cloud
89 system such as Amazon EC2. However, for common use you can setup your
90 replicated storage to be enabled at the time you connect the databases.
91 The following is a breakdown of how you may wish to do this. Again, if
92 you are using Catalyst, I strongly recommend you use (or upgrade to)
93 the latest Catalyst::Model::DBIC::Schema, which makes this job even
94 easier.
95 First, you need to get a $schema object and set the storage_type:
97 my $schema = MyApp::Schema->clone;
99 $schema->storage_type([
100 '::DBI::Replicated' => {
101 balancer_type => '::Random',
102 balancer_args => {
103 auto_validate_every => 5,
104 master_read_weight => 1
105 },
106 pool_args => {
107 maximum_lag =>2,
108 },
109 }
110 ]);
111 Then, you need to connect your DBIx::Class::Schema.
113 $schema->connection($dsn, $user, $pass);
115 Let's break down the settings. The method "storage_type" in
117 DBIx::Class::Schema takes one mandatory parameter, a scalar value, and
118 an option second value which is a Hash Reference of configuration
119 options for that storage. In this case, we are setting the Replicated
120 storage type using '::DBI::Replicated' as the first value. You will
121 only use a different value if you are subclassing the replicated
122 storage, so for now just copy that first parameter.
123 The second parameter contains a hash reference of stuff that gets
125 passed to the replicated storage. "balancer_type" in
126 DBIx::Class::Storage::DBI::Replicated is the type of software load
127 balancer you will use to split up traffic among all your replicants.
128 Right now we have two options, "::Random" and "::First". You can review
129 documentation for both at:
130 DBIx::Class::Storage::DBI::Replicated::Balancer::First,
132 DBIx::Class::Storage::DBI::Replicated::Balancer::Random.
133 In this case we will have three replicants, so the ::Random option is
135 the only one that makes sense.
136 'balancer_args' get passed to the balancer when it's instantiated. All
138 balancers have the 'auto_validate_every' option. This is the number of
139 seconds we allow to pass between validation checks on a load balanced
140 replicant. So the higher the number, the more possibility that your
141 reads to the replicant may be inconsistent with what's on the master.
142 Setting this number too low will result in increased database loads, so
143 choose a number with care. Our experience is that setting the number
144 around 5 seconds results in a good performance / integrity balance.
145 'master_read_weight' is an option associated with the ::Random
147 balancer. It allows you to let the master be read from. I usually
148 leave this off (default is off).
149 The 'pool_args' are configuration options associated with the replicant
151 pool. This object (DBIx::Class::Storage::DBI::Replicated::Pool)
152 manages all the declared replicants. 'maximum_lag' is the number of
153 seconds a replicant is allowed to lag behind the master before being
154 temporarily removed from the pool. Keep in mind that the Balancer
155 option 'auto_validate_every' determines how often a replicant is tested
156 against this condition, so the true possible lag can be higher than the
157 number you set. The default is zero.
158 No matter how low you set the maximum_lag or the auto_validate_every
160 settings, there is always the chance that your replicants will lag a
161 bit behind the master for the supported replication system built into
162 MySQL. You can ensure reliable reads by using a transaction, which
163 will force both read and write activity to the master, however this
164 will increase the load on your master database.
165 After you've configured the replicated storage, you need to add the
167 connection information for the replicants:
168 $schema->storage->connect_replicants(
170 [$dsn1, $user, $pass, \%opts],
171 [$dsn2, $user, $pass, \%opts],
172 [$dsn3, $user, $pass, \%opts],
173 );
174 These replicants should be configured as slaves to the master using the
176 instructions for MySQL native replication, or if you are just learning,
177 you will find MySQL::Sandbox an easy way to set up a replication
178 cluster.
179 And now your $schema object is properly configured! Enjoy!
181
183 Check the list of additional DBIC resources.
184
186 This module is free software copyright by the DBIx::Class (DBIC)
187 authors. You can redistribute it and/or modify it under the same terms
188 as the DBIx::Class library.
189perl v5.36.0 DBIx::Clas2s0:2:3S-t0o1r-a2g0e::DBI::Replicated::Introduction(3)