On 27/03/14 18:10 +1300, Robert Collins wrote:
On 27 March 2014 17:30, Tom Fifield <t...@openstack.org> wrote:

Does anyone disagree?

/me raises hand

When I was an operator, I regularly referred to the sample config files
in the git repository.

If there weren't generated versions of the sample config in the repo, I
would probably grep the code (not an ideal user experience!). Running
some random script that I don't know about the existence and might
depend on having something else installed of is probably not something
that would happen.

So, I think its important you have sample configs to refer to.

Do they need to be in the git repo?

Note that because libraries now export config options (which is the
root of this problem!) you cannot ever know from the source all the
options for a service - you *must* know the library versions you are
running, to interrogate them for their options.

We can - and should - have a discussion about the appropriateness of
the layering leak we have today, but in the meantime this is breaking
multiple projects every time any shared library that uses oslo.config
changes any config option... so we need to solve the workflow aspect.

How about we make a copy of the latest config for each project for
each series - e.g. trunk of everything, Icehouse of servers with trunk
of everything else, etc and make that easily acccessible?

I'd agree with the original proposal if - and only if - something like
what Robert proposed here is done.

I'd say the config file could be generated for each milestone cut and
live in the milestone branch.

As Tom pointed out, referring to the sample configs is very useful
from many points of view (operations, support, development etc).


Flavio Percoco

Attachment: pgp5FItgZgpGb.pgp
Description: PGP signature

OpenStack-dev mailing list

Reply via email to