[Container-tools] Why 'atomic' cli when 'atomicapp' serves the purpose

Ratnadeep Debnath rtnpro at redhat.com
Mon Feb 8 09:31:02 UTC 2016


Hey Suraj,

On 02/05/2016 05:40 PM, Suraj Deshmukh wrote:
> What I mean here is referring [1] in the seventh step, why do I build
> a docker image,
> and then use 'atomic' cli to run the docker container, while same
> thing I can do with 'atomicapp'?
> And that also leaves me with one less container image? How does this
> all fall into place?

You got it right! We can use atomicapp directly to run Nulecule 
applications,
either from source or from container images. However, as part of the current
use case of Atomic Host, atomicapp command is not available on it, but the
atomic CLI, and it's part of the product decision, AFAIK, to use atomic CLI
on atomic host alone.

That being said, atomicapp, is a stand alone product, and does not depend on
atomic CLI at all. It's a reference implementation of the Nulecule spec. 
atomicapp
gives you more flexibility to play with Nulecule. So, people who just 
want to work
with Nulecule applications are free to use atomicapp directly.

I hope that this answers your query. Let me know if you have some question.

>
> [1] http://www.rtnpro.com/nuleculizing-an-docker-image/
>

Thanks,
rtnpro




More information about the Container-tools mailing list