[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Draft Minutes F2F 8-10th July 2013
Meeting Minutes Face to Face Meeting 8th thru 10th July 2013, Broomfield, Colorado, USA
Meeting Attendees
US Department of Defense (DoD) Michael Behrens Voting Member
Oracle Martin Chapman Chair
Cloudsoft Corporation Limited Alex Heneveld Member
Cloudsoft Corporation Limited Duncan Johnston-Watt Member
Oracle Anish Karmarkar Voting Member
Oracle Ashok Malhotra Voting Member
Rackspace Hosting, Inc. Adrian Otto Secretary
Vnomic Derek Palma Voting Member
Oracle Gilbert Pilz Voting Member
Fujitsu Limited Tom Rutt Voting Member
Topic: Intro:
Adrian assume scribe duties for the first session of the F2F.
Roll Call: Attendees listed above: Voting Members: 8 of 11 (72%) . Meeting is quorate.
Topic: Agenda
Goal: resolve P1 issues to drive us to public review.
Agenda approved, noting that issues and timeslots will get juggled around.
Topic: Minutes
3rd July 2013: https://www.oasis-open.org/apps/org/workgroup/camp/email/archives/201307/msg00010.html
MOTION: Gil approve minutes from 2013-07-03
second by Tom
No discussion, no objections, minutes approved.
Topic: Editor's Update
no remarks.
Topic: Administrivia
Cloudsoft agrees to present at Madrid event (http://www.cloudplugfest.org/cloud-interoperability-week )
Santa Clara, USA, September 16-18 and Madrid, Spain, September 18-20, 2013
Duncan Johnston-Watt (Cloudsoft): Alex to attend https://sites.google.com/a/cloudplugfest.org/welcome/upcoming-events/next-event-september-18-20-2013
Two events in two locations: Madrid and Santa Clara. We will have TC representatives at both. Santa Clara will be attended by Gil.
Discussion about attendance at Eurocloud event in October.
"Eurocloud Congress" October 15,16 Luxembourg
Duncan Johnston-Watt (Cloudsoft): CFP http://events.linuxfoundation.org/events/cloudopen-europe/program/cfp
Jane will send a request for volunteers to the TC list.
Topic: Review of P1 Issues
7 open P1 issues: CAMP-4, CAMP-9, CAMP-45, CAMP-55, CAMP-56, CAMP-61, CAMP-65
Solutions plans will land for each of these. No apparent showstoppers.
we plan to discuss CAMP-45 on Wed morning
as it's going to be editorial in nature.
Topic: Issue Discussion
CAMP-4: https://tools.oasis-open.org/issues/browse/CAMP-4
Gil presents.
proposal is: https://www.oasis-open.org/apps/org/workgroup/camp/download.php/49858/camp-spec-v1.1-wd12-issue-4-v7.doc
Example 6 changed, some lines were removed
this simplifies the Ruby/Rails DP
solves a bug: added the version node in the DP to each example.
that concludes the summary of changes.
Alex suggests we make the version attribute named CampVersion
"campVersion" ?
s/CampVersion/campVersion/
Gil speaks in opposition
Anish speaks in opposition
Alex agrees.
MOTION: Alex moves to resolve Issue CAMP-4 with https://www.oasis-open.org/apps/org/workgroup/camp/download.php/49858/camp-spec-v1.1-wd12-issue-4-v7.doc with s/version/campVersion/ and examples changed accordingly
Anish seconds
s/version/campVersion/ change is in section 4.3.2.1 version
No Discussion, No Objections
Motion passed w/o
recognition of a major milestone
Topic: CAMP-65: https://tools.oasis-open.org/issues/browse/CAMP-65
Anish takes action item to make an Inline word format proposal based on the current draft proposal and attach it to the issue
Derek asks why tar format, and not zip? Zip is not indexed.
Zip is indexed, but tar is not. tar has metadata concern that's platform specific.
tar allows for a stream to be expanded with a chunked encoding
Zip can not
Alex proceeds with presentation of the proposal draft.
Anish Karmarkar: zip spec reference: http://www.pkware.com/documents/APPNOTE/APPNOTE-6.3.2.TXT
Anish: that's in the draft at the end.
discussion of external spec references required for this proposal
references to TAR and GZIP will be added
Discussion of archive formats concluded
Tom asks for clarification on the camp.yaml
Gil asks why should we even mention Zip, when it's one of many available archive formats
Anish responds indicating that ZIP is very common among small package deployments
Alex suggests that we also require support for ZIP
Derek Palma (Vnomic): +1 to require zip support because 1) it's pervasive (a lot of artifacts and tools use this format), 2) it's not that hard to support (i.e. most everyone will support it anyway)
Members of the F2F will revise the proposal, and present it tomorrow
Topic: CAMP-56: https://tools.oasis-open.org/issues/browse/CAMP-56
General agreement that we use the term CAMP Provider and CAMP Consumer, and that in conformance section we will declare that CAMP Provider and CAMP Consumer shall refer to any mention of "provider" and "consumer" respectively in the spec.
Anish Karmarkar: "A compliant CAMP Provider MUST comply with all mandatory requirements in this specification which have Provider as a target."
Session recess proposed by chair
Next session 1.30PM MT Tuesday July 9
Session open:
Duncan scribes for the second F2F session.
Topic: Agenda revisit
Today's agenda will cover:
https://tools.oasis-open.org/issues/browse/CAMP-9 Support sensors and effectors on components via API
Updated Agenda: https://www.oasis-open.org/apps/org/workgroup/camp/download.php/49890/camp-spec-v1.1-wd15-issue-9-v4.doc
Camp 62
Planning
Topic: CAMP-9 https://tools.oasis-open.org/issues/browse/CAMP-9
Martin reporting that CAMP-9 has been discussed informally in preparation for this meeting
Adrian presenting v4 CAMP-9 - context necessary primitives for policy-based management of platform EG scheduled maintenance; auto scaling etc
2.1.1.5 introduces concept of operations and sensors and their role supporting policy-based management
Discussion about use of term "policy-based management"
would favor saying less rather than more
the danger of saying too much is that people will misinterpret what our spec says
Sensors and operations are necessary but not sufficient hence "enablers" - Alex
Note: Operations are also known as effectors (para 2)
Mike Behrens - does the spec discuss policy frameworks?
Alex/Martin - Out of scope, can be layered above
CAMP spec is providing bare bones support (sensors/operations) for policy-based management without expressing any view about type of policy framework
Reviewing 5.26
Reviewing 5.27
Reviewing 5.28.4 sensorType
Reviewing 5.28.5 value; 5.28.6 (optional) timestamp and 5.28.7 operations
Adrian - should be operationsUri (corrected)
Adrian is submitting v5 containing edits from this realtime review: https://www.oasis-open.org/apps/org/workgroup/camp/download.php/49890/camp-spec-v1.1-wd15-issue-9-v5.doc
MOTION: tom, s: adrian resolve camp-9 with proposal in v5 https://www.oasis-open.org/apps/org/workgroup/camp/download.php/49890/camp-spec-v1.1-wd15-issue-9-v5.doc (and in JIRA )
no discussion
no objections
CAMP-9 is resolved
Topic: CAMP-62 https://tools.oasis-open.org/issues/browse/CAMP-62
Either close with no action or with recommended best practice (note)
Adrian to take ownership and prepare proposal to present at future date
Topic: CAMP-61 https://tools.oasis-open.org/issues/browse/CAMP-61
Informal discussion to follow
Adrian - Could downgrade ahead of public review?
Topic: CAMP-55 https://tools.oasis-open.org/issues/browse/CAMP-55
Informal discussion to follow
Gil - spec cannot go to public review without clearing up this issue
Adrian - agreed. This needs to be addressed.
MartinC - #1 priority for informal session tomorrow
Tom - CAMP-9 - Editors to add common attributes
Topic: Reviewing P2s
General discussion about relationship between application and platform components
Reviewing https://tools.oasis-open.org/issues/browse/CAMP-29 and https://tools.oasis-open.org/issues/browse/CAMP-5
MartinC - Effectively the same issue
Tom - 5.26 - targetResource should be URI (pseudo schema)
discussing https://tools.oasis-open.org/issues/browse/CAMP-51
MOTION: Gil, to change CAMP-51 from P3 to P2
Tom - seconded
No discussion
No objections
Topic: Planning
1. Timeline to public review
MartinC - Target to vote on PR candidate Wed July 31
Tom - Can work on test assertions during "quiet period"
2. Timeline to complete Test Assertions
MartinC - One month out from PR
MartinC - PR will begin 1-2 weeks after vote then run for 30 days
Which means PR will wrap Friday Sept 13
MartinC - Logically a F2F early Oct to close out all issues/PR comments ahead of final PR
MartinC - Second PR end Oct (15 days)
MartinC - wrap by end of year
ACTION - MartinC to verify who is on the TC comment list
3. Setting date for raising the bar
Anish - End of Sept
(Ahead of F2F early Oct)
F2F week of Oct 7 (Oct 8-10)
Potential location London
Session recess
Wednesday Session:
Gil scribes for the 3rd and final session of the F2F.
The chair noted that Duncan did an excellent job of scribing session 2 that Duncan can scribe more often.
Topic: Agenda revisit
https://tools.oasis-open.org/issues/browse/CAMP-65
Proposal in Jira: https://www.oasis-open.org/apps/org/workgroup/camp/document.php?document_id=49910
https://tools.oasis-open.org/issues/browse/CAMP-55 Lifecycle Diagram is conflating two levels of states, leading to inconsistencies
proposal: https://www.oasis-open.org/apps/org/workgroup/camp/document.php?document_id=49914
TOPIC: CAMP-55 https://tools.oasis-open.org/issues/browse/CAMP-55
(Alex presents proposal)
https://www.oasis-open.org/apps/org/workgroup/camp/download.php/49914/camp-spec-issue-55-wd15-v3.docx
Alex: discussion of the lifecycle was not in synch with current thinking
... in addition, we repeated ourselves at least three times through sections 2 and 3
... renamed Section 2 to "Concepts and Types"
... Section 2.1 renamed to "Application Assemblies"
... changed text to change self-referential definitions like "An Application Component is a component of an application".
... Adrian added a new diagram to show how Assembly Templates and Assemblies are created
... broke "metadata" section into more descriptive sections
... extensions, parameters, common types, etc.
Martin: Adrian's diagram is a mini-lifecycle
Alex: Section 3 now "Application Management Lifecycle"
... initially there is a PC with some PCTs and, maybe, some PCs
... register an app and you get Assembly Templates, Application Component Templates, etc.
... more information on the use of "representationSkew"
Martin: do we have suspend and resume
Alex: no
... but we have operations
Martin: don't think we want to create common operations
Alex: finally we have removing Assemblies and AssemblyTemplates
... at the end, the resources should look just like you found them before you registered the application
Martin: this proposal is a new direction (not necessarily bad), we need to read it and think about it
Tom: we took away the state chart?
Martin: Yes
Anish: what is attractive is that we are reducing the degree of abstraction in this part of the spec
Tom: I like it. Do we need to wait to accept this?
(misc. discussion about the need)
Gil: Adrian can be assumed to be in favor of it.
Martin: I want to read it. We'll put it on the agenda for next week.
Anish: People on the phone?
Ashok: It looks OK, though I haven't read all of it.
Derek: It feels OK - but I need to study it in more detail.
Put on next weeks agenda.
TOPIC: CAMP-65 https://tools.oasis-open.org/issues/browse/CAMP-65
Martin: We went over the proposal yesterday. This is that text in the context of the spec.
Anish: (presents proposal)
Alex Heneveld (Cloudsoft): https://docs.google.com/document/d/1r-BZymi8QZ9JVwJSLsosGQlBSfu5u98N5yQimIKnxhE/edit?usp=sharing
Alex Heneveld (Cloudsoft): ^^^ this is the input text to Anish's proposal
(some small changes to proposal text)
(small changes degenerate into a full-on joint editing session)
Anish: Changes to Section 6. We now speak in terms of registering an application, not a PDP. An application can be registered using either a PDP (which contains a DP) or by using a bare DP.
... one change I do want to point out.
... If you provided a PDP, the AssemblyTemplate "pdpURI" attribute had to point to a copy of the PDP.
... We added the "dpURI" attribute to the AssemblyTemplate to mirror this behavior for DPs.
(misc. discussion on the use of "pdpURI" and "dpURI")
Anish: One attribute or two attributes?
Tom: If you could possibly have both, use two attributes.
Anish: Outstanding issue around may/should/shall for media types of different packaging formats.
(joint editing session continues)
Anish: We've just made some minor, material changes. Are people OK with those?
... Does this look reasonable?
All:: (assenting noises)
Anish: Next question, are people comfortable in accepting this?
All: (assenting noises)
MOTION: Resolve CAMP-65 with the proposal at https://www.oasis-open.org/committees/document.php?document_id=49916&wg_abbrev=camp
SECOND: Anish
RESOLUTION: motion approved by unanimous consent
Anish Karmarkar: approved proposal: https://www.oasis-open.org/committees/document.php?document_id=49916&wg_abbrev=camp
TOPIC: Appreciations
The TC thanks Cloudsoft for their gracious hospitality in hosting this F2F
TOPIC: New Issues
https://tools.oasis-open.org/issues/browse/CAMP-72 "Error Response Message Resource looks like relic of a bygone age"
Gil: (describes issue)
MOTION: Tom: I move to open CAMP-72
Gil: Seconds
RESOLUTION: Motion approved by unanimous consent
TOPIC: Int'l Cloud Symposium (ICS)
Martin: If anybody can/wants to cover this, send email in response to Jane.
TOPIC: AOB
No stragglers.
MEETING ADJOURNED
----
One outstanding ACTION:
ACTION - MartinC to verify who is on the TC comment list
-------------
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]