DBIx::Class::Storage::User:ContrDBIx::Class::Storage::DBI::Replicated::Pool(3)NAMEDBIx::Class::Storage::DBI::Replicated::Pool - Manage a pool of
replicants
SYNOPSIS
This class is used internally by DBIx::Class::Storage::DBI::Replicated.
You shouldn't need to create instances of this class.
DESCRIPTION
In a replicated storage type, there is at least one replicant to handle
the read-only traffic. The Pool class manages this replicant, or list
of replicants, and gives some methods for querying information about
their status.
ATTRIBUTES
This class defines the following attributes.
maximum_lag ($num)
This is a number which defines the maximum allowed lag returned by the
"lag_behind_master" in DBIx::Class::Storage::DBI method. The default
is 0. In general, this should return a larger number when the
replicant is lagging behind its master, however the implementation of
this is database specific, so don't count on this number having a fixed
meaning. For example, MySQL will return a number of seconds that the
replicating database is lagging.
last_validated
This is an integer representing a time since the last time the
replicants were validated. It's nothing fancy, just an integer provided
via the perl time built-in.
replicant_type ($classname)
Base class used to instantiate replicants that are in the pool. Unless
you need to subclass DBIx::Class::Storage::DBI::Replicated::Replicant
you should just leave this alone.
replicants
A hashref of replicant, with the key being the dsn and the value
returning the actual replicant storage. For example, if the $dsn
element is something like:
"dbi:SQLite:dbname=dbfile"
You could access the specific replicant via:
$schema->storage->replicants->{'dbname=dbfile'}
This attributes also supports the following helper methods:
set_replicant($key=>$storage)
Pushes a replicant onto the HashRef under $key
get_replicant($key)
Retrieves the named replicant
has_replicants
Returns true if the Pool defines replicants.
num_replicants
The number of replicants in the pool
delete_replicant ($key)
Removes the replicant under $key from the pool
master
Reference to the master Storage.
METHODS
This class defines the following methods.
connect_replicants ($schema, Array[$connect_info])
Given an array of $dsn or connect_info structures suitable for
connected to a database, create an
DBIx::Class::Storage::DBI::Replicated::Replicant object and store it in
the "replicants" attribute.
connect_replicant ($schema, $connect_info)
Given a schema object and a hashref of $connect_info, connect the
replicant and return it.
_safely_ensure_connected ($replicant)
The standard ensure_connected method with throw an exception should it
fail to connect. For the master database this is desirable, but since
replicants are allowed to fail, this behavior is not desirable. This
method wraps the call to ensure_connected in an eval in order to catch
any generated errors. That way a slave can go completely offline (e.g.
the box itself can die) without bringing down your entire pool of
databases.
_safely ($replicant, $name, $code)
Execute $code for operation $name catching any exceptions and printing
an error message to the "<$replicant-"debugobj>>.
Returns 1 on success and undef on failure.
connected_replicants
Returns true if there are connected replicants. Actually is overloaded
to return the number of replicants. So you can do stuff like:
if( my $num_connected = $storage->has_connected_replicants ) {
print "I have $num_connected connected replicants";
} else {
print "Sorry, no replicants.";
}
This method will actually test that each replicant in the "replicants"
hashref is actually connected, try not to hit this 10 times a second.
active_replicants
This is an array of replicants that are considered to be active in the
pool. This does not check to see if they are connected, but if they
are not, DBIC should automatically reconnect them for us when we hit
them with a query.
all_replicants
Just a simple array of all the replicant storages. No particular order
to the array is given, nor should any meaning be derived.
validate_replicants
This does a check to see if 1) each replicate is connected (or
reconnectable), 2) that is ->is_replicating, and 3) that it is not
exceeding the lag amount defined by "maximum_lag". Replicants that
fail any of these tests are set to inactive, and thus removed from the
replication pool.
This tests "all_replicants", since a replicant that has been previous
marked as inactive can be reactivated should it start to pass the
validation tests again.
See DBIx::Class::Storage::DBI for more about checking if a replicating
connection is not following a master or is lagging.
Calling this method will generate queries on the replicant databases so
it is not recommended that you run them very often.
This method requires that your underlying storage engine supports some
sort of native replication mechanism. Currently only MySQL native
replication is supported. Your patches to make other replication types
work are welcomed.
AUTHOR
John Napiorkowski <john.napiorkowski@takkle.com>
LICENSE
You may distribute this code under the same terms as Perl itself.
perl v5.16.2DBIx::Class::Storage::DBI::Replicated::Pool(3)