Lwood-20160911

September 12, 2016 at 22:12 · Filed under Lwood, OpenStack

Introduction

Welcome to Last week on OpenStack Dev (“Lwood”) for the week just past. For more background on Lwood, please refer here.

Basic Stats for week 5 to 11 September for openstack-dev:

  • ~556 Messages (up about 13% relative to last week)
  • ~160 Unique threads (down about 11% relative to last week)

A pretty typical week on the list, two new and quite long threads (~28 and ~60 messages each) contributing significantly to the Messages up/Threads down result.  A nice moment of levity in the middle of the 60 message thread too…

Notable Discussions – openstack-dev

New OpenStack Security Notices

Host machine exposed to tenant networks via IPv6 [OSSN-0069]

From the summary “New interfaces created by Neutron in the default namespace, were done so without disabling IPv6 link-local addresses. This resulted in instances gaining the ability to directly access the host OS, therefore breaking guest isolation.” More in the original post or SSN itself.

Horizon dashboard leaks internal information through cookies [OSSN-0073]

From the summary (slightly paraphrased) “When horizon is configured, its URL contains the IP address of the internal URL of keystone, as the default value for the identity service is “internalURL”.  This has the effect of exposing the internal IP address to the outside world”. The original post or the SSN has more informatio.

The first Project Teams Gathering (PTG)

Thierry Carrez’s email gives a handy summary of the upcoming Project Teams Gathering (PTG) meeting in Atlanta on February 20-24 next year.

The PTG meetings are in essence the “Design Summit” part of the OpenStack Summit separated off into a standalone event.  There will continue to be a link to the Summit attendees through a “Forum” for community wide discussions.

There is more detail on the what’s changed and the rationale on the Foundation website here as well as a FAQ.  It’s worth a bit of a read as this is one of the more fundamental changes to the way OpenStack development is undertaken for some time.

Ocata Schedule Approved

Doug Hellmann notes that the schedule for Ocata has been approved.  There’s more information in a nice table here or in the original review – release week is February 20, 2017.

Writing down OpenStack Principles

One of the longer threads this week and ongoing at the time of writing was one that Chris Dent kicked off.  In his initial post he draws attention to a review that Thierry Carrez created in concert with other TC members to try and translate some of the “assumed knowledge” that the TC has accumulated over the years into a written form.

The thread has been a little contentious in places (about 5 out of 10 on my newly created lwood-fuss-o-meter) but based on my read that is more about the approach taken than any fundamental disagreement about the end result.  You can view the document in its current form here – it’s worth a quick read.

Future election timing and “Release Stewards”

The longest thread of the week, the conversation kicked off by Thierry Carrez in this post flags some side effects of the change in Summit/PTG arrangements and the timing of release cycles.  As currently defined the timing of elections is reference to the Summit – this would have the effect of meaning PTL and TC positions being renewed mid-cycle.  At first blush this seems a bit suboptimal but Thierry makes the point that it might not necessarily be so.

He goes on to make the case for the creation of a new role – a “Release Steward” who would oversee the complete release “from requirements gathering to post-release-bugfix-backport phase”  This would in practice last longer than the release cycle of six months as many aspects of the release precede and follow the cycle proper.

Thierry suggests that the PTL could elect to be Release Steward as well, or (more likely) someone else in the team.  Having someone doing this important work would free the PTL to continue with strategic pieces knowing the minutiae of the release was in good hands.

The thread goes on to tweak and debate the idea, an alternative first suggested by Sean Dague but with some virtual nods from others along the lines of “yes I was wondering about that”, would instead run the PTL elections early and have the “PTL-next” doing the pre-work for the next release while the PTL took care of landing the current release.

At the time of writing the discussion is continuing – it’s a pretty accessible thread if you want to dig in, if not I’ll endeavour to summarise the eventual outcome when the dust settles :)

Oh and the funny little exchange starting here is worth a few clicks to read as it unfolds :)

Election Season Continues

Following on from his post last week, Tristan Cacqueray advised that nominations are now open for OpenStack PTLs.  Nominations remain open until September 18, 23:45 UTC.  The elections begin 15 minutes later at 00:00 September 19, 2016 (UTC) and run until 23:45 September 25, 2016 (UTC)  If you’re curious the candidates will be listed here as they’re approved.

Subteam and Mid-Cycle Reports

None this week.

Notable Discussions – other OpenStack lists

Nothing that leapt out from the other lists this week.

Upcoming OpenStack Events

Best I can tell no OpenStack related events mentioned this week.  Don’t forget the OpenStack Foundation’s Events Page for a list of general events that is frequently updated.

People and Projects

Core nominations & changes

New, Proposed and Changed OpenStack Projects

Joe Huang noted that the Tricircle project has voted to on the name Trio2o for their api-gateway efforts.

Further Reading & Miscellanea

Don’t forget these excellent sources of OpenStack news – most recent ones linked in each case

This weeks edition of Lwood brought to you by Weather Report (Forecast Tomorrow) and Marillion (Holidays in Eden)

Last but by no means least, thanks, as always, to Rackspace :)

Comments are closed.