Problem with ansible recipe about galaxykickstart HOT 8 CLOSED

The problem is that the virtualbox had no candidate on the host system. I will increase the robustness of the install script by testing the distro, by trying to find the correct repo for the virtualbox (must be also compatible with vagrant) and by testing if the steps were terminated correctly.

from galaxykickstart.

The install script and README.md are modified according to the new simpler install.sh (Please, see issue #16).

from galaxykickstart.

Just my 2 cents here, but that install.sh script must die! ;).
What's the point of shell scripting if you need to have ansible on the host.
Use ansible to install ansible, vagrant and so on. ;).
You don't need ansible on the remote side (even if I claimed earlier that you do ...).
You can point ansible at your ec2 instance, and let it figure out the dependency installation.

from galaxykickstart.

Yes, it is not working indeed.
On another note, now that I am digging in it, I find that the galaxy install using ansible should be uncoupled from virtualbox. I don't care about virtualbox when I install galaxy on EC2, and even it bothers me (right now)

from galaxykickstart.

Correct me if I am wrong.
The objective was to deploy a Galaxy instance into a Virtual box, was not?
I will edit again the README.md to cover the deployment direct to the host system, but it will need a correctly configured ansible environment on the host.

from galaxykickstart.

Let me be blunt.
If you don't want the objective(s) being re-evaluated , get your stuff working.
Have a look again to http://www.agilemanifesto.org/principles.html and pay attention to

• Deliver working software frequently, from a couple of weeks to a couple of months, with a preference to the shorter timescale.
• Working software is the primary measure of progress
  just to extract 2 items...

If I see no progress, I feel it's my duty to revisit the objectives.

Plus, sorry to correct, but the objective is to deploy a Galaxy instance, and Virtual box is one mean among others to reach this objective. I said uncouple, I did not say abandon.

from galaxykickstart.

The installation of the VM (as asked in the beginning) is working (tested in a BSD clean) since the middle of October. It did not work because the environment of the node was misconfigured (virtualbox was not correctly installed in the cloud node). In the last two weeks I am doing some improvements suggested by Marius and some of myself in the playbook and improving my NGS knowledge. Thus, I do not not want to source of controversy, so there are now two Install scripts, one for galaxy into a vm (a guest machine) and the other for galaxy on the host machine. The installation on host is not tested yet (tomorrow morning i will). To do so can you give me access to your cloud node, please?

from galaxykickstart.

I am closing this issue, it's redundant with current testing

from galaxykickstart.