1 package DBIx::Class::Storage::DBI::Replicated;
4 use Carp::Clan qw/^DBIx::Class/;
6 ## Modules required for Replication support not required for general DBIC
7 ## use, so we explicitly test for these.
9 my %replication_required = (
11 'MooseX::AttributeHelpers' => '0.21',
12 'MooseX::Types' => '0.16',
13 'namespace::clean' => '0.11',
14 'Hash::Merge' => '0.11'
19 for my $module (keys %replication_required) {
20 eval "use $module $replication_required{$module}";
21 push @didnt_load, "$module $replication_required{$module}"
25 croak("@{[ join ', ', @didnt_load ]} are missing and are required for Replication")
30 use DBIx::Class::Storage::DBI;
31 use DBIx::Class::Storage::DBI::Replicated::Pool;
32 use DBIx::Class::Storage::DBI::Replicated::Balancer;
33 use DBIx::Class::Storage::DBI::Replicated::Types qw/BalancerClassNamePart DBICSchema DBICStorageDBI/;
34 use MooseX::Types::Moose qw/ClassName HashRef Object/;
35 use Scalar::Util 'reftype';
36 use Carp::Clan qw/^DBIx::Class/;
37 use Hash::Merge 'merge';
39 use namespace::clean -except => 'meta';
43 DBIx::Class::Storage::DBI::Replicated - BETA Replicated database support
47 The Following example shows how to change an existing $schema to a replicated
48 storage type, add some replicated (readonly) databases, and perform reporting
51 You should set the 'storage_type attribute to a replicated type. You should
52 also define your arguments, such as which balancer you want and any arguments
53 that the Pool object should get.
55 $schema->storage_type( ['::DBI::Replicated', {balancer=>'::Random'}] );
57 Next, you need to add in the Replicants. Basically this is an array of
58 arrayrefs, where each arrayref is database connect information. Think of these
59 arguments as what you'd pass to the 'normal' $schema->connect method.
61 $schema->storage->connect_replicants(
62 [$dsn1, $user, $pass, \%opts],
63 [$dsn2, $user, $pass, \%opts],
64 [$dsn3, $user, $pass, \%opts],
67 Now, just use the $schema as you normally would. Automatically all reads will
68 be delegated to the replicants, while writes to the master.
70 $schema->resultset('Source')->search({name=>'etc'});
72 You can force a given query to use a particular storage using the search
73 attribute 'force_pool'. For example:
75 my $RS = $schema->resultset('Source')->search(undef, {force_pool=>'master'});
77 Now $RS will force everything (both reads and writes) to use whatever was setup
78 as the master storage. 'master' is hardcoded to always point to the Master,
79 but you can also use any Replicant name. Please see:
80 L<DBIx::Class::Storage::DBI::Replicated::Pool> and the replicants attribute for more.
82 Also see transactions and L</execute_reliably> for alternative ways to
83 force read traffic to the master. In general, you should wrap your statements
84 in a transaction when you are reading and writing to the same tables at the
85 same time, since your replicants will often lag a bit behind the master.
87 See L<DBIx::Class::Storage::DBI::Replicated::Instructions> for more help and
92 Warning: This class is marked BETA. This has been running a production
93 website using MySQL native replication as its backend and we have some decent
94 test coverage but the code hasn't yet been stressed by a variety of databases.
95 Individual DB's may have quirks we are not aware of. Please use this in first
96 development and pass along your experiences/bug fixes.
98 This class implements replicated data store for DBI. Currently you can define
99 one master and numerous slave database connections. All write-type queries
100 (INSERT, UPDATE, DELETE and even LAST_INSERT_ID) are routed to master
101 database, all read-type queries (SELECTs) go to the slave database.
103 Basically, any method request that L<DBIx::Class::Storage::DBI> would normally
104 handle gets delegated to one of the two attributes: L</read_handler> or to
105 L</write_handler>. Additionally, some methods need to be distributed
106 to all existing storages. This way our storage class is a drop in replacement
107 for L<DBIx::Class::Storage::DBI>.
109 Read traffic is spread across the replicants (slaves) occuring to a user
110 selected algorithm. The default algorithm is random weighted.
114 The consistancy betweeen master and replicants is database specific. The Pool
115 gives you a method to validate its replicants, removing and replacing them
116 when they fail/pass predefined criteria. Please make careful use of the ways
117 to force a query to run against Master when needed.
121 Replicated Storage has additional requirements not currently part of L<DBIx::Class>
124 MooseX::AttributeHelpers => '0.20',
125 MooseX::Types => '0.16',
126 namespace::clean => '0.11',
127 Hash::Merge => '0.11'
129 You will need to install these modules manually via CPAN or make them part of the
130 Makefile for your distribution.
134 This class defines the following attributes.
138 The underlying L<DBIx::Class::Schema> object this storage is attaching
151 Contains the classname which will instantiate the L</pool> object. Defaults
152 to: L<DBIx::Class::Storage::DBI::Replicated::Pool>.
159 default=>'DBIx::Class::Storage::DBI::Replicated::Pool',
161 'create_pool' => 'new',
167 Contains a hashref of initialized information to pass to the Balancer object.
168 See L<DBIx::Class::Storage::DBI::Replicated::Pool> for available arguments.
182 The replication pool requires a balance class to provider the methods for
183 choose how to spread the query load across each replicant in the pool.
187 has 'balancer_type' => (
189 isa=>BalancerClassNamePart,
192 default=> 'DBIx::Class::Storage::DBI::Replicated::Balancer::First',
194 'create_balancer' => 'new',
200 Contains a hashref of initialized information to pass to the Balancer object.
201 See L<DBIx::Class::Storage::DBI::Replicated::Balancer> for available arguments.
205 has 'balancer_args' => (
215 Is a <DBIx::Class::Storage::DBI::Replicated::Pool> or derived class. This is a
216 container class for one or more replicated databases.
222 isa=>'DBIx::Class::Storage::DBI::Replicated::Pool',
233 Is a <DBIx::Class::Storage::DBI::Replicated::Balancer> or derived class. This
234 is a class that takes a pool (<DBIx::Class::Storage::DBI::Replicated::Pool>)
240 isa=>'DBIx::Class::Storage::DBI::Replicated::Balancer',
242 handles=>[qw/auto_validate_every/],
247 The master defines the canonical state for a pool of connected databases. All
248 the replicants are expected to match this databases state. Thus, in a classic
249 Master / Slaves distributed system, all the slaves are expected to replicate
250 the Master's state as quick as possible. This is the only database in the
251 pool of databases that is allowed to handle write traffic.
261 =head1 ATTRIBUTES IMPLEMENTING THE DBIx::Storage::DBI INTERFACE
263 The following methods are delegated all the methods required for the
264 L<DBIx::Class::Storage::DBI> interface.
268 Defines an object that implements the read side of L<BIx::Class::Storage::DBI>.
272 has 'read_handler' => (
285 Defines an object that implements the write side of L<BIx::Class::Storage::DBI>.
289 has 'write_handler' => (
301 deployment_statements
304 build_datetime_parser
318 with_deferred_fk_checks
321 with_deferred_fk_checks
335 has _master_connect_info_opts =>
336 (is => 'rw', isa => HashRef, default => sub { {} });
338 =head2 around: connect_info
340 Preserve master's C<connect_info> options (for merging with replicants.)
341 Also set any Replicated related options from connect_info, such as
342 C<pool_type>, C<pool_args>, C<balancer_type> and C<balancer_args>.
346 around connect_info => sub {
347 my ($next, $self, $info, @extra) = @_;
349 my $wantarray = wantarray;
352 for my $arg (@$info) {
353 next unless (reftype($arg)||'') eq 'HASH';
354 %opts = %{ merge($arg, \%opts) };
358 if (@opts{qw/pool_type pool_args/}) {
359 $self->pool_type(delete $opts{pool_type})
363 merge((delete $opts{pool_args} || {}), $self->pool_args)
366 $self->pool($self->_build_pool)
370 if (@opts{qw/balancer_type balancer_args/}) {
371 $self->balancer_type(delete $opts{balancer_type})
372 if $opts{balancer_type};
374 $self->balancer_args(
375 merge((delete $opts{balancer_args} || {}), $self->balancer_args)
378 $self->balancer($self->_build_balancer)
382 $self->_master_connect_info_opts(\%opts);
386 @res = $self->$next($info, @extra);
388 $res = $self->$next($info, @extra);
391 # Make sure master is blessed into the correct class and apply role to it.
392 my $master = $self->master;
393 $master->_determine_driver;
394 Moose::Meta::Class->initialize(ref $master);
395 DBIx::Class::Storage::DBI::Replicated::WithDSN->meta->apply($master);
397 $wantarray ? @res : $res;
402 This class defines the following methods.
406 L<DBIx::Class::Schema> when instantiating its storage passed itself as the
407 first argument. So we need to massage the arguments a bit so that all the
408 bits get put into the correct places.
413 my ($class, $schema, $storage_type_args, @args) = @_;
424 Lazy builder for the L</master> attribute.
430 my $master = DBIx::Class::Storage::DBI->new($self->schema);
436 Lazy builder for the L</pool> attribute.
442 $self->create_pool(%{$self->pool_args});
445 =head2 _build_balancer
447 Lazy builder for the L</balancer> attribute. This takes a Pool object so that
448 the balancer knows which pool it's balancing.
452 sub _build_balancer {
454 $self->create_balancer(
456 master=>$self->master,
457 %{$self->balancer_args},
461 =head2 _build_write_handler
463 Lazy builder for the L</write_handler> attribute. The default is to set this to
468 sub _build_write_handler {
469 return shift->master;
472 =head2 _build_read_handler
474 Lazy builder for the L</read_handler> attribute. The default is to set this to
479 sub _build_read_handler {
480 return shift->balancer;
483 =head2 around: connect_replicants
485 All calls to connect_replicants needs to have an existing $schema tacked onto
486 top of the args, since L<DBIx::Storage::DBI> needs it, and any C<connect_info>
487 options merged with the master, with replicant opts having higher priority.
491 around connect_replicants => sub {
492 my ($next, $self, @args) = @_;
495 $r = [ $r ] unless reftype $r eq 'ARRAY';
497 croak "coderef replicant connect_info not supported"
498 if ref $r->[0] && reftype $r->[0] eq 'CODE';
500 # any connect_info options?
502 $i++ while $i < @$r && (reftype($r->[$i])||'') ne 'HASH';
505 $r->[$i] = {} unless $r->[$i];
507 # merge if two hashes
508 my @hashes = @$r[$i .. $#{$r}];
510 croak "invalid connect_info options"
511 if (grep { reftype($_) eq 'HASH' } @hashes) != @hashes;
513 croak "too many hashrefs in connect_info"
516 my %opts = %{ merge(reverse @hashes) };
519 splice @$r, $i+1, ($#{$r} - $i), ();
521 # make sure master/replicants opts don't clash
522 my %master_opts = %{ $self->_master_connect_info_opts };
523 if (exists $opts{dbh_maker}) {
524 delete @master_opts{qw/dsn user password/};
526 delete $master_opts{dbh_maker};
529 %opts = %{ merge(\%opts, \%master_opts) };
535 $self->$next($self->schema, @args);
540 Returns an array of of all the connected storage backends. The first element
541 in the returned array is the master, and the remainings are each of the
548 return grep {defined $_ && blessed $_} (
550 values %{ $self->replicants },
554 =head2 execute_reliably ($coderef, ?@args)
556 Given a coderef, saves the current state of the L</read_handler>, forces it to
557 use reliable storage (ie sets it to the master), executes a coderef and then
558 restores the original state.
564 $schema->resultset('User')->create({name=>$name});
565 my $user_rs = $schema->resultset('User')->find({name=>$name});
569 my $user_rs = $schema->storage->execute_reliably($reliably, 'John');
571 Use this when you must be certain of your database state, such as when you just
572 inserted something and need to get a resultset including it, etc.
576 sub execute_reliably {
577 my ($self, $coderef, @args) = @_;
579 unless( ref $coderef eq 'CODE') {
580 $self->throw_exception('Second argument must be a coderef');
583 ##Get copy of master storage
584 my $master = $self->master;
586 ##Get whatever the current read hander is
587 my $current = $self->read_handler;
589 ##Set the read handler to master
590 $self->read_handler($master);
592 ## do whatever the caller needs
594 my $want_array = wantarray;
598 @result = $coderef->(@args);
599 } elsif(defined $want_array) {
600 ($result[0]) = ($coderef->(@args));
606 ##Reset to the original state
607 $self->read_handler($current);
609 ##Exception testing has to come last, otherwise you might leave the
610 ##read_handler set to master.
613 $self->throw_exception("coderef returned an error: $@");
615 return $want_array ? @result : $result[0];
619 =head2 set_reliable_storage
621 Sets the current $schema to be 'reliable', that is all queries, both read and
622 write are sent to the master
626 sub set_reliable_storage {
628 my $schema = $self->schema;
629 my $write_handler = $self->schema->storage->write_handler;
631 $schema->storage->read_handler($write_handler);
634 =head2 set_balanced_storage
636 Sets the current $schema to be use the </balancer> for all reads, while all
637 writea are sent to the master only
641 sub set_balanced_storage {
643 my $schema = $self->schema;
644 my $balanced_handler = $self->schema->storage->balancer;
646 $schema->storage->read_handler($balanced_handler);
651 Check that the master and at least one of the replicants is connected.
658 $self->master->connected &&
659 $self->pool->connected_replicants;
662 =head2 ensure_connected
664 Make sure all the storages are connected.
668 sub ensure_connected {
670 foreach my $source ($self->all_storages) {
671 $source->ensure_connected(@_);
677 Set the limit_dialect for all existing storages
683 foreach my $source ($self->all_storages) {
684 $source->limit_dialect(@_);
686 return $self->master->quote_char;
691 Set the quote_char for all existing storages
697 foreach my $source ($self->all_storages) {
698 $source->quote_char(@_);
700 return $self->master->quote_char;
705 Set the name_sep for all existing storages
711 foreach my $source ($self->all_storages) {
712 $source->name_sep(@_);
714 return $self->master->name_sep;
719 Set the schema object for all existing storages
725 foreach my $source ($self->all_storages) {
726 $source->set_schema(@_);
732 set a debug flag across all storages
739 foreach my $source ($self->all_storages) {
743 return $self->master->debug;
748 set a debug object across all storages
755 foreach my $source ($self->all_storages) {
756 $source->debugobj(@_);
759 return $self->master->debugobj;
764 set a debugfh object across all storages
771 foreach my $source ($self->all_storages) {
772 $source->debugfh(@_);
775 return $self->master->debugfh;
780 set a debug callback across all storages
787 foreach my $source ($self->all_storages) {
788 $source->debugcb(@_);
791 return $self->master->debugcb;
796 disconnect everything
802 foreach my $source ($self->all_storages) {
803 $source->disconnect(@_);
809 set cursor class on all storages, or return master's
814 my ($self, $cursor_class) = @_;
817 $_->cursor_class($cursor_class) for $self->all_storages;
819 $self->master->cursor_class;
824 Due to the fact that replicants can lag behind a master, you must take care to
825 make sure you use one of the methods to force read queries to a master should
826 you need realtime data integrity. For example, if you insert a row, and then
827 immediately re-read it from the database (say, by doing $row->discard_changes)
828 or you insert a row and then immediately build a query that expects that row
829 to be an item, you should force the master to handle reads. Otherwise, due to
830 the lag, there is no certainty your data will be in the expected state.
832 For data integrity, all transactions automatically use the master storage for
833 all read and write queries. Using a transaction is the preferred and recommended
834 method to force the master to handle all read queries.
836 Otherwise, you can force a single query to use the master with the 'force_pool'
839 my $row = $resultset->search(undef, {force_pool=>'master'})->find($pk);
841 This attribute will safely be ignore by non replicated storages, so you can use
842 the same code for both types of systems.
844 Lastly, you can use the L</execute_reliably> method, which works very much like
847 For debugging, you can turn replication on/off with the methods L</set_reliable_storage>
848 and L</set_balanced_storage>, however this operates at a global level and is not
849 suitable if you have a shared Schema object being used by multiple processes,
850 such as on a web application server. You can get around this limitation by
851 using the Schema clone method.
853 my $new_schema = $schema->clone;
854 $new_schema->set_reliable_storage;
856 ## $new_schema will use only the Master storage for all reads/writes while
857 ## the $schema object will use replicated storage.
861 John Napiorkowski <john.napiorkowski@takkle.com>
863 Based on code originated by:
865 Norbert Csongrádi <bert@cpan.org>
866 Peter Siklósi <einon@einon.hu>
870 You may distribute this code under the same terms as Perl itself.
874 __PACKAGE__->meta->make_immutable;