Thanks Nate! That's very helpful.

-----Original Message-----
From: Nate Coraor [] 
Sent: Friday, June 08, 2012 3:31 PM
To: Dorset, Daniel C
Subject: Re: [galaxy-dev] How does Galaxy access datasets?

On Jun 5, 2012, at 6:08 PM, Dorset, Daniel C wrote:

> I'm trying to troubleshoot why I can't retrieve output from my Galaxy cluster 
> instance. I notice that when I click on any output, the URL is something like:
> http://[root galaxy address]/datasets/[some 16-character 
> hash]/display/[file name]
> I'm not able to find the "datasets" directory on the local machine, and I 
> couldn't figure anything out by searching the paster.log file and the apache 
> access and error logs. Everytime I try to access output, it downloads a 
> zero-byte file. The files that I want to download through Galaxy are in the 
> correct subdirectory of /database/files/... If someone could explain to be 
> what's going on "behind the scenes," it would help me quite a bit. I'm 
> guessing that the absolute path is stored in a database, but beyond that I 
> don't know any specifics. If that's the case, knowing the relevant database 
> tables would be a huge hint.

Hi Dan,

The hash is decoded and converted to an ID using code in 
lib/galaxy/web/security/ and the value of id_secret in 
universe_wsgi.ini.  The decoded id is then passed to code in 
lib/galaxy/objectstore/ to assemble the path underneath 

You may find galaxy-dist/scripts/ in converting history dataset IDs to 
filesystem paths (they are also available directly in the web interface if you 
are an administrator or set expose_dataset_path = True in universe_wsgi.ini).


Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

Reply via email to