[asterisk-dev] Git Migration

[Russell Bryant]

On Tue, Sep 16, 2014 at 3:48 PM, Matthew Jordan <mjordan at digium.com> wrote:

> "And there was much rejoicing"
>

\o/


> But seriously, we all know that a lot of people have wanted to move to Git
> for some time. For the record, everyone at Digium has wanted to move the
> project to Git for some time. I swore to myself that we wouldn't do another
> Standard release on Subversion - after we spent at least six weeks mucking
> around with merge conflicts during Asterisk 12 - and with Asterisk 14
> looming ever closer, the time is now to start getting something done on
> this.
>
> So!
>
> To that end, a page on the wiki has been made with some initial thoughts:
>
> https://wiki.asterisk.org/wiki/display/AST/Git+Migration
>
> To summarize:
>  * A comparison of management platforms has been done. Barring a giant
> catastrophe or some insane limitation, we're going to go simple here and
> stick with gitolite. Reasoning is on the wiki page.
>  * The first thing to migrate is _not_ the Asterisk project, but the
> Asterisk Test Suite. That will allow us (or force us) to deal with some of
> the tooling and process issues, which will make it easier to tackle
> Asterisk.
>
> I'm sure there are a lot of opinions about all of this, and if you have
> thoughts on technical or process hurdles we may be running into, I'd love
> to hear it. Just remember that like many other things in life and
> development, there's a lot of ways to manage your source code. You may
> really, really, _REALLY_ like the way Project X does it, and you may think
> that the way we are proposing it is clearly inferior. That's great, you may
> be right. But in the interest of this not dragging on for another 5 years,
> I'd like to keep any discussions focussed on getting things done while not
> shooting ourselves in our virtual feet.
>

So, I realize this is pretty much exactly the opposite of what you just
said, but I wanted to offer some comments on the infrastructure.  :-)  I'm
really not deeply invested in what is chosen.  My interest here is just to
provide an overview of another option in the interest of exploring options.

For the last few years, most of my time has been going into OpenStack [1].
 We use git and I have become a big fan of our workflow and infrastructure.
 It's all open source and reusable.

>From a high level, all patches go to a code review system.  *Every* patch
must be peer reviewed (usually by 2 people, but that's a policy decision).
 *Every* patch must also pass tests.  Once a patch passes both tests and
peer review, it is automatically merged into the repository.

I *love* that workflow for several reasons.  If it's appealing, it's
probably much easier to do it now while you're doing a big switch anyway.
 If you're not sold, I'm certainly not hurt.  Like I said, I just wanted to
offer info.  The current plan will be less up front setup for sure.

If you're a hands on kind of person, browse http://review.openstack.org/
for open code reviews.  You can also see patches going through CI pipelines
on http://status.openstack.org/zuul//

The major tools involved are:

 - gerrit for code review and repository management [2][3]
 - jenkins for CI [4]
 - Zuul, A CI job scheduler that automates running things in response to
events on gerrit. [5]
 - CGit, repo hosting [6][7]

Everything we use is managed via puppet and all of the configuration is in
git.  It's designed to be reusable.  The folks that run it have documented
how to re-use it [8] and are quite friendly.  You can find them in
#openstack-infra on freenode.

[1] http://www.openstack.org
[2] https://code.google.com/p/gerrit/
[3] https://review.openstack.org/
[4] http://jenkins-ci.org/
[5] http://ci.openstack.org/zuul.html
[6] http://ci.openstack.org/git.html
[7] http://git.openstack.org
[8] http://ci.openstack.org/running-your-own.html

I'll also try to answer the fields of the comparison chart on the wiki page:

-- Web View

Yes.

-- Project Management

This would replace existing usage of bamboo and reviewboard.  It does not
include issue tracking.  Keeping that is would be fine.

-- Protected Branches

Gerrit supports permissions on a per branch basis.

-- Rewriting history

Not sure the intent here ... wanting to make that can be avoided?

In this system, the merges are automated so you can't accidentally do a bad
push.  An admin can force reset a repo if needed, of course.

-- Team repos

I'd recommend just using your own account on github or whatever.

-- Git hooks

For what, exactly?  It's probably easier to discuss the problem that needs
to be solved.

-- Web hooks

Again, it's probably worth discussing the use case.  Gerrit has an event
stream.  That event stream includes merges.  Tools to do things in
responses to merges (which could be running a web hook) listen and react.

-- Performance

Yes.  :-)

There's lots of stats I could dig up, but as one example, I track code
review stats for a subset of projects on review.openstack.org  In the last
year, for that subset, there have been 266,127 code reviews done (avg > 700
per day).  That gives some sense of scale.  From a quick glance at the CI
status page (http://status.openstack.org/zuul/), it has been launching
about 500-600 jobs per hour today.

-- Process Recommendation

I discussed this a good bit above, but I'm happy to answer questions.

-- 
Russell Bryant
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20140916/93836c1d/attachment.html>