On Fri, 2017-03-17 at 09:47 +0100, Miroslav Suchý wrote: > Hi, > I am in process in writing playbook for retrace.fedoraproject.org > > ABRT team created: > https://github.com/abrt/ansible-role-retrace-server > > Which we can use. I just wonder what is best practise for using such 3rd > party roles? > Should I just copy it into our ansible.git? Or should I use ansible-galaxy > command to sync it? Manually? Or when the > playbook is run?
We ran into this a while back with the openshift ansible stuff. I guess the answer is 'it depends'. If it's pretty small and can be made pretty close to our existing conventions, I'd say just copy it in and try and keep the external and internal ones in sync. If it's more complex or written in a incompatible way or changes too much, then we could do what we did with osbs, which is to make a host, something like: retrace-control and in our ansible playbooks we go to that host and pull the role/setup and run it from there on the retrace machines. This role seems pretty small to me and hopefully won't change too much, so hopefully we can just do the first of these. ;) Thoughts? kevin
signature.asc
Description: This is a digitally signed message part
_______________________________________________ infrastructure mailing list -- infrastructure@lists.fedoraproject.org To unsubscribe send an email to infrastructure-le...@lists.fedoraproject.org
