Hey Everyone,

Last week the Devtools Deploy team had a meeting to discuss our goals around 
Ansible 
and Atomic App, based on the requirements coming from our PM and engineering 
teams. We 
are concerned that we have too many tools in this space - in particular we 
think that 
we end up spending a lot of time explaining the differences between these 
tools, rather 
than working on solving real use cases.

We think that Ansible itself has an interesting end-to-end story for people 
looking to deploy 
applications, and we think that working with Ansible, and in particular the 
ansible-container 
project could solve a number of the use cases we want to look at. As one of our 
first goals
we want to bring some of the deployment support we have in Atomic App to 
ansible container - 
we intend to make PRs based on our experience. We are hoping to become tightly 
integrated 
with their team and share the same roadmap when it comes to that tool.

We also want to get more involved in existing upstream communities that are 
centered around 
Kubernetes/OpenShift. We have already started to get involved with the Kompose 
project [1] 
and we want to evaluate Helm [2] packaging and how it relates to Kompose and 
see where the good 
things about Nulecule can be brought to the table there.

As a result of focusing on Ansible and Kompose/Helm for the next 6 months, it 
does mean we will slow 
down our work on Atomic App. One thing we do want to do is merge the Nulecule 
and Atomic 
App repositories so that it is less confusing to approach that project. We'll 
be merging the
Nulecule repository into the Atomic App repository and Nulecule will become an 
implementation 
detail for Atomic App. After the merge we'll let it rest for a while and 
re-evaluate it in some
time.

Questions/Concerns?

Dusty

[1] - https://github.com/skippbox/kompose
[2] - https://github.com/kubernetes/helm

_______________________________________________
Container-tools mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/container-tools

Reply via email to