# Multiple database connections

**URL:** <https://www.ruby-forum.com/t/multiple-database-connections/154918>\
**Category:** Rails\
**Created:** [December 5, 2008, 6:36am UTC](https://www.ruby-forum.com/t/multiple-database-connections/154918 "2008-12-05T06:36:24Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Mukund](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/e8c25b/32.png) [@Mukund](https://www.ruby-forum.com/u/Mukund)\
**Post date:** [December 5, 2008, 6:36am UTC](https://www.ruby-forum.com/t/multiple-database-connections/154918/1 "2008-12-05T06:36:24Z")

</div>

For multiple database connections, is it better to have a mixin  
included into the model that does a establish connection on included  
event or is it better to have an abstract class that establishes the  
connnection and inherit from the abstract class? In particular, how  
do Rails handle database connections? It is supposed to use a  
connection pool, does the pooling break if you use a mixin?

Mixin:

module CustomConnect

def self.included(base)  
config = ActiveRecord::Base.configurations[“custom\_”+RAILS\_ENV]  
base.establish\_connection(config)  
end

end

and then you do a

class abc \< ActiveRecord::Base

include CustomConnect

end

Inheritance:

class CustomConnect \< ActiveRecord::Base

self.abstract\_class = true  
config = ActiveRecord::Base.configurations[“custom\_”+RAILS\_ENV]  
establish\_connection(config)

end

class abc \< CustomConnect

end

When you have a significant number of models, how are database  
connections handled and which approach is better?

---

<div class="post-metadata">

**Author:** ![Mukund](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/m/e8c25b/32.png) [@Mukund](https://www.ruby-forum.com/u/Mukund)\
**Post date:** [December 7, 2008, 4:51am UTC](https://www.ruby-forum.com/t/multiple-database-connections/154918/2 "2008-12-07T04:51:04Z")

</div>

Does anyone have an input on connection pooling and which approach is  
better?