[openstack-dev] [stable][all] Keeping Juno "alive" for longer.

Sean Dague sean at dague.net
Mon Nov 9 14:42:57 UTC 2015


On 11/09/2015 05:30 AM, Hugh Blemings wrote:
> Hiya,
> 
> On 7/11/2015 06:42, Sean Dague wrote:
>> On 11/06/2015 01:15 AM, Tony Breeds wrote:
>>> Hello all,
>>>
>>> I'll start by acknowledging that this is a big and complex issue and I
>>> do not claim to be across all the view points, nor do I claim to be
>>> particularly persuasive ;P
>>>
>>> Having stated that, I'd like to seek constructive feedback on the
>>> idea of
>>> keeping Juno around for a little longer.  During the summit I spoke to a
>>> number of operators, vendors and developers on this topic.  There was
>>> some
>>> support and some "That's crazy pants!" responses.  I clearly didn't
>>> make it
>>> around to everyone, hence this email.
>>>
>>> Acknowledging my affiliation/bias:  I work for Rackspace in the private
>>> cloud team.  We support a number of customers currently running Juno
>>> that are,
>>> for a variety of reasons, challenged by the Kilo upgrade.
>>
>> The upstream strategy has been make upgrades unexciting, and then folks
>> can move forward easily.
>>
>> I would really like to unpack what those various reasons are that people
>> are trapped. Because figuring out why they feel that way is important
>> data in what needs to be done better on upgrade support and testing.
> 
> In reading this thread and Sean's post, I wonder out loud if we're
> seeing something somewhat new to OpenStack here, but perhaps not to
> other FOSS projects.
> 
> Specifically does Kilo happen to mark a point where a much larger number
> of end users have adopted OpenStack and so we're starting to see a much
> greater number of visible and mainstream users facing the "upgrade
> difficulty question" ?
> 
> If Juno us the point where we suddenly got an order of magnitude more
> deployments, then some point later you'll see an order of magnitude more
> end users struggling with how/when to upgrade.
> 
> Really wish I could articulate this better, but perhaps the point can be
> distilled from the ramble...

Honestly, I think that every release has seen a larger number of new
installations than the release before it. The stable EOL thread is
nothing new. The promise that people will show up is nothing new. The
lack of anyone else showing up to help maintain stable is nothing new.

I do think we need to focus on the snags though. Very few upstreams
maintain LTS releases. A big piece of that is it makes upgrades harder.
It means a ton of changes are being inflicted on you all at once.
Especially if you want to get to the point of live upgrading an
installation using live migration to create 0 downtime environments.
Which means that you've got to be able to live upgrade between the
versions of libvirt / ovs across that time frame.

So lets figure out where the snags are. I'm pretty uninterested in
threads that just scream LTS without a list of upgrade bugs that have
been filed to describe why rapid upgrade isn't the right long term
solution.

	-Sean

-- 
Sean Dague
http://dague.net



More information about the OpenStack-dev mailing list