ODK Aggregate to Sql or PostgreSql

I already installed odk Aggregate on Ubuntu server and modify ODK collect to send data only to my ODK Aggregate. I also have fusion tables in which ODK Aggregate's data send automaticaly. Now I want to get my data to sql or postgresql to manage them. Also, I read https://github.com/opendatakit/aggregate/blob/master/README.md and I understand how to create database and user for postgre but I have no idea how my data can automaticly set in my table.
Thank you.

My recommendation would be to install Ubuntu base OS (with ssh). Then tomcat, and unit test your tomcat is working. Then MySQL, and unit test MySQL is working. Then mysql-connector (the Tomcat-to-MySQL interface), and test its working. And only then install ODK Aggregate. If you've set everything up correctly, getting your form submission data into the underlying MySQL database should 'just work'. Which is to say, you should probably have MySQL in place and working on your system before trying to install ODK Aggregate - if only to help identify what broke where.

In my experience installing ODK Aggregate on bare metal Linux requires a strict 'recipe' - the above procedure works reliably for me. Just my $0.02.


And just to add what @Xiphware said, if you are a bit technical, you can follow the steps we use to build the Aggregate VM here: https://github.com/opendatakit/aggregate/tree/master/packer/ansible. The VM is Debian-based, but it's close enough to Ubuntu that the commands should work in both.

If there's any interest, I've documented my pretty detailed (that is, command-line by command-line) install procedure I follow whenever setting up a new ODK server on bare metal. Its probably not for the feint of heart, you need to be familiar with Linux shell commands and the like, but if its on any interest I'd be happy to upload it somewhere suitable (@yanokwa, @LN?)

1 Like

@adammichaelwood How do you want to handle docs contributions like this? For the official docs, I worry that it might be too technical, too specific, and likely to fall out of date.

My inclination is that @Xiphware publish it wherever (e.g., Github Gist, Google Doc) and then we can link to that as a tip inside the docs?

1 Like

That would be great if you do @Xiphware !

1 Like

I probably agree with this, though I'd like to see the document. I can imagine it being useful and appropriate for the official docs, but it also might be way out of scope.

If you have a blog or personal website you can publish it on, that would be great. Otherwise, I think a GH gist would be my preferred way to share this sort of thing. (I find Google docs really annoying for this kind of content, but that is just me.)

It would be wonderful if you could:

  • publish what you have somewhere
  • open an issue on the docs repo linking to it, providing whatever bg info is helpful, and mentioning that we might want to either include it in the docs or link to it.
1 Like

Hello Xiphware, yes it is a great idea and a big help.

There's now a pretty thorough HOWTO for installing Aggregate on AWS posted in the ODK docs (I'm not sure who added it?). I need to work thru it myself to determine whether mine still has anything useful to add... I'd suggest following this one, and if you run into any problems please post back and we can look at updating it.

1 Like

Hello @Xiphware, can you share the link of the instalation on ubuntu?
Is it posible and good idea Install Postgres 11 alone in a server and others two instance of aggregate in other server?
What operative system you recomend to install Postgres 11 and the aplication of aggregate?
Note: I have one strong machine so they would be virtualized on several machines.

Hi @yanokwa, How can I access aggregate db and extract or query through psql in Aggregate VM?

1 Like