ingest.py will not parse the downloaded data about bibserver HOT 2 CLOSED

Hello there,

when you run bibserver via gunicorn you need to follow the production
deployment methods. The main point here is that the ingest service is not
run for you automatically, because it runs as a stand-alone asynchronous
service that picks up jobs as they appear on the queue. This allows you to
easily scale your frontend and backend processes separately and as required.

So if you are doing production deployment and using gunicorn to manage your
bibserver frontend, you need to do something similar to start and manage
the ingest service too.

I note that our deployment docs do not explicitly state that you need to do
the setup for the ingest method too in this sort of case, so I will update
those to make it more obvious.

Thanks for checking out bibserver!

Mark

On Wed, Aug 8, 2012 at 11:50 PM, fy [email protected] wrote:

I followed the deployment instructions on
https://github.com/okfn/bibserver/blob/master/doc/deploy.rst
I installed bibserver on Ubuntu 12.04 and everything seemed to work just
fine. I could access bibserver from both localhost and my domain.

However, when I tried to upload a json file that is 2.7mb in size, the
ticket page showed the status of the ticket was "downloaded". I kept
refreshing the page and waited at least 10 minutes, but the status was
still "downloaded". I did not have this problem when I ran the server with
"python bibserver/web.py" instead of running it with gunicorn.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/246.

from bibserver.

Great, thanks!
Uploading worked after I added another program in supervisor:

[program:ingest]
command=python bibserver/ingest.py -d
user=www-data
directory=.........

from bibserver.