# Active Records - multiple connections question

**URL:** <https://www.ruby-forum.com/t/active-records-multiple-connections-question/174863>\
**Category:** Rails\
**Created:** [September 3, 2009, 7:32am UTC](https://www.ruby-forum.com/t/active-records-multiple-connections-question/174863 "2009-09-03T07:32:07Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![lasastard](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/l/58956e/32.png) [@lasastard](https://www.ruby-forum.com/u/lasastard)\
**Post date:** [September 3, 2009, 7:32am UTC](https://www.ruby-forum.com/t/active-records-multiple-connections-question/174863/1 "2009-09-03T07:32:07Z")

</div>

Hi List,

I figure that rails is the most appropriate board to post this in, tho  
my problem is not necessarily rails-specific, but rather an active  
records issue.

So here goes:  
I am working on an ruby-implementation of an API to a large genome  
database. This database is in reality split into multiple database, one  
for each sequenced genome. There is on ‘central’ database that holds  
comparative data between genomes and references the corresponding  
organism-specific database for fetching the actual data.

Connecting to any of these genomes works by specifying the organism name  
in the connection method  
SomeModule::DBConnection.connect(‘homo\_sapiens’). So the Module/Class is  
always the same, except the organism name which specifies which database  
to connect to.

What happens tho if I try to fetch data from multiple genomes in a row  
is that the ruby process locks up eventually. I am guessing it has  
something to do with active records and how it manages these  
connections. For example, I am not using (and don’t know how) a  
termination signal that would close the old connection as I figure that  
this shouldn’t be an issue if a new one is created to replace the old  
one (?).

Anyhow, I know this may be a little confusing but I guess you get the  
idea.

My question is: What to do about this issue (multiple database  
connections in rapid succession via the same DBConnection class) - is  
there a more sophisticated way to handle this? Also,optional question,  
how does one properly deal with dropped connections (which could be part  
of the issue)?

Cheers,  
Marc

---

<div class="post-metadata">

**Author:** ![lasastard](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/l/58956e/32.png) [@lasastard](https://www.ruby-forum.com/u/lasastard)\
**Post date:** [September 3, 2009, 1:45pm UTC](https://www.ruby-forum.com/t/active-records-multiple-connections-question/174863/2 "2009-09-03T13:45:34Z")

</div>

You could try forcing a garbage collection with something like  
GC.start (I think). You could try doing that after each connection,  
or after every N connections, if it takes too long.

–wpd

---

<div class="post-metadata">

**Author:** ![lasastard](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/l/58956e/32.png) [@lasastard](https://www.ruby-forum.com/u/lasastard)\
**Post date:** [September 4, 2009, 3:50am UTC](https://www.ruby-forum.com/t/active-records-multiple-connections-question/174863/3 "2009-09-04T03:50:29Z")

</div>

The optional question should be pretty easy to solve.  
ActiveRecord::Base has a method called ‘connected?’.

I am not that experienced with multiple connections, though, so I’d  
rather let somebody else give you an more intelligent answer than I  
would.

Pepe

On Sep 3, 1:32 am, Marc H. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid)