Design Specs: Database Connectivity
Contents
Objectives
Part of the WormBase::API. Needs to be able to connect to multiple databases (multiple instances of the same database in some cases). Should be fault tolerant if a server or database crashes.
Implementation
http://bitbucket.org/tharris/wormbase/src/tip/lib/WormBase/API/Role/Service.pm a Moose Role module which defines the common attributes and behavior for different databases.
eg.
'hosts' array reference attribute which has all the available hosts from configuration file.
reconnect: abstract method for the database connection, fail-over is also implementation here:
- use Moose 'around' method modifier to check on the db handle every time it is accessed
around 'dbh' sub {}
Note that attribute accessors also count as methods for the purposes of satisfying the requirements of a role.
- do "reconnect" if dbh is not set, undef or time-out(AceDB)
- will try to connect for maximum $ReconnectMaxTries(now this is hard coded as 5, will replace by conf) times
- host is picked randomly among the pool
my $host = $self->hosts->[ rand @{$self->hosts} ];
This Role is consumed by each Database Classes(Service module)
- http://bitbucket.org/tharris/wormbase/src/tip/lib/WormBase/API/Service/acedb.pm
- http://bitbucket.org/tharris/wormbase/src/tip/lib/WormBase/API/Service/gff.pm
each has its own "connect" method and "ping" method(mostly just a wrapper)
AceDB
use Ace;
sub connect {
my $self = shift;
my ($host,$port,$user,$pass)=@_;
return Ace->connect(-host => $host,
-port => $self->port,
-user=>$self->user,
-pass=>$self->pass,
);
# @cache);
}
GFF database
use Bio::DB::GFF;
sub connect {
my $self = shift;
my ($host,$port,$user,$pass)=@_;
return Bio::DB::GFF->new( -user => $user,
-pass => $pass,
-dsn => "dbi:mysql:database=".$self->species.";host=" . $host,
);
}
Minor changes in main API http://bitbucket.org/tharris/wormbase/src/tip/lib/WormBase/API.pm
- took out the BUILD subroutine
- modified gff_dsn() method
Testing
prove -lv t/WormBase/API/Object/Gene.t http://bitbucket.org/tharris/wormbase/src/tip/t/WormBase/API/Object/Gene.t
ok 1 - use WormBase::API; ok 2 - Constructed WormBase::API object ok [DEBUG] acedb dbh for specie doesn't exist or is not alive; trying to connect [INFO] trytime 1: Connecting to acedb [DEBUG] using the following parameters: [DEBUG] aceserver.cshl.edu:2005 [INFO] --> succesfully established connection to acedb on aceserver.cshl.edu [DEBUG] acedb dbh for specie exists and is alive! ok 3 - The object isa WormBase::API::Object::Gene ... ... [DEBUG] gff dbh for specie c_elegans doesn't exist or is not alive; trying to connect [INFO] trytime 1: Connecting to gff [DEBUG] using the following parameters: Use of uninitialized value in concatenation (.) or string at /home/xshi/work/project/wormbase/website-2.0/lib/WormBase/API/Role/Service.pm line 106. [DEBUG] aceserver.cshl.edu: [INFO] --> succesfully established connection to gff on aceserver.cshl.edu [DEBUG] gff dbh for specie c_elegans exists and is alive! [DEBUG] gff dbh for specie c_elegans exists and is alive! ...
Question
When Ace time out would the dbh becomes undef? How useful is the Ace reopen method? For now I just re-connect.
Continued Work
setting up attributes such as hosts, user, password, caching for AceDB etc through Configuration file.
Catalyst::Plugin::ConfigLoader
use Ace (); to prevent eg. the connect sub from Ace being imported into our namespaces.
Things to think about
persistent connections
Apache::DBI DBI connect_cached DBIx::Class
Also, because Apache::DBI's connection caching never checks to see if the process and/or thread id changed, it will give you an invalid $dbh from a previous process/thread if you use a connection, then thread/fork, then try to re-connect to the same source, which is a Bad Thing.
DBIx::Class already manages its connections for you, and therefore it cannot benefit from Apache::DBI under any scenario. It makes one connection per-process, and keeps that connection persistent, reconnecting only if the connection appears to have died, or if you fork/thread over to another process/thread-id. The only Apache::DBI issue in DBIx::Class is that Apache::DBI will actually thwart DBIx::Class's connection management code, and cause it to use the same (and invalid) connection in a new process, in cases such as (as mentioned above) if you make a DBI connection before forking in a prefork mod_perl server.
Resources
The current connectivity code is part of the API. The API, in turn, is inside of the lib directory of the WormBase mercurial repository:
http://bitbucket.org/tharris/wormbase/src/tip/lib/WormBase/API/Service/
Currently, each database is a "Service". When the API is instantiated, connections are made to each required service. Database handles are cached. When a database handle is needed, the service first checks to see if it is live, if not it refreshes the connection. This would probably be the ideal location to implement failover -- check the current connection, if dead, try a different database server.
- Note that attribute accessors also count as methods for the purposes of satisfying the requirements of a role.
http://bitbucket.org/tharris/wormbase/src/9009075c4c5a/lib/WormBase/API/Role/Service/gff.pm
around 'dbh' => sub {
my $orig = shift;
my $self = shift;
my $species = $self->species;
# Do we already have a dbh? HOW TO TEST THIS WITH HASH REF?
if ($self->has_dbh) {
$self->log->debug(" gff-dbh for $species exists and is alive!");
return $self->$orig;
} else {
$self->log->debug(" gff-dbh for $species doesn't exist yet; trying to connect");
my $dbh = $self->connect($species);
}
};
Note also that these Services use Moose "Roles" to define shared methods and variables like connect().
Here's the acedb service:
http://bitbucket.org/tharris/wormbase/src/tip/lib/WormBase/API/Service/acedb.pm
The acedb service "consumes" the generic Service role:
http://bitbucket.org/tharris/wormbase/src/9009075c4c5a/lib/WormBase/API/Role/Service.pm
This nomenclature and structure is a bit confusing and probably needs to be streamlined/refined!!
