Skip to end of metadata
Go to start of metadata

Date

IRC Minutes

Recording

Full IRC Log

Zoom Chat Log

Zoom Attendance Log


Discussion items 

DurationAgenda ItemRequested byNotes / Links
START RECORDING
5 mHousekeeping

10 mRelease Status
10 mIntegration Status

Known Vulnerability AnalysisStephen Terrill

20 MTSC Composition

Not covered

15 mSubcommittee / Coordinator UpdateCarried forward for 5 weeks (... and still not covered this week)
15 mVersioning & API Documentation Guidelinesseeking TSC approval for Casablanca . Not covered
5m

Self-commit exemption for PTLs

(by request from PTL meet)

Allow PTLs to +2 and merge their own commits

Why?

  • PTL's are now burdened by the +2 rule when they need to make
    • immediate fixes - usually jenkins/sonar related
    • experiment with getting usually jenkins/sonar/release issues going
  • weekend/TZ 2nd committer not available
  • need to triage jenkins issues without a 12h+ turnaround on each commit

Discussed in PTL meet this monday - by consensus

I was going to do a slide - but I find that a read-only way to put up editable content.

Original hardening of the commit rules was proposed by me on TSC 2018-03-15 - at the time a PTL exception was not fully discussed.

Full IRC Log 

13:52:59 <kennypaul> #startmeeting tsc-2018-05-24
13:52:59 <collabot`> Meeting started Thu May 24 13:52:59 2018 UTC.  The chair is kennypaul. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:52:59 <collabot`> Useful Commands: #action #agreed #help #info #idea #link #topic.
13:52:59 <collabot`> The meeting name has been set to 'tsc_2018_05_24'
13:53:39 <kennypaul> #chair phrobb SteveT gildaslanilis
13:53:39 <collabot`> Warning: Nick not in channel: gildaslanilis
13:53:39 <collabot`> Current chairs: SteveT gildaslanilis kennypaul phrobb
13:53:53 <kennypaul> #topic rollcall
13:57:08 <SteveT> #info Stephen Terrill, Ericsson
13:59:07 <kennypaul> #info Lingli Deng, China Mobile
14:00:10 <RajeshGadiyar> #info Rajesh Gadiyar Intel
14:00:34 <Xiaojun> #info Xiaojun Xie, China Telecom
14:00:42 <xinhuili> #info Xinhui Li, Vmware
14:00:44 <Susana> #info Susana Sabater, Vodafone
14:01:15 <EricDebeau> #info Eri Debeau, Orange
14:01:16 <RannyHaiby> #info Ranny Haiby, Nokia
14:01:27 <JasonHunt> #info Jason Hunt, IBM
14:03:30 <gilesheron> #info proxy Giles Heron, Cisco Systems
14:03:39 <gilbert> #info mazin
14:04:01 <kennypaul> #Yoav Kluger, proxy , AMDOCS
14:06:47 <cdonley> #info Chris Donley Huawei
14:07:49 <SteveT> #topic Housekeeping
14:08:17 <SteveT> #info ONAP maillists are planed to move from mailman to groups.io on june 20.
14:09:44 <SteveT> #info Jira federation. Would like to be able cross reference tickets. Call for any concerns?
14:10:38 <kennypaul> #info Murat Tupcu. Turk Telekom
14:12:06 <SteveT> #info The jira federation will be delayed to after the Beijing release (2 weeks)
14:13:56 <SteveT> #info Reminder about request for internships
14:14:41 <kennypaul> #topic release management
14:14:57 <kennypaul> #info @gildaslanilis reviewed his slides
14:16:31 <kennypaul> #info artifacts - email helpdest w/ the link to the daily version
14:17:10 <kennypaul> #info review CSIT and daily build failures
14:17:46 <kennypaul> #info drop in new defects founf
14:21:16 <kennypaul> #info release notes 1 file contains all history. - new info needs to be appended
14:21:50 <kennypaul> #info document new functionality and known bugs
14:22:21 <kennypaul> #info gildas will assist with vulnerability docs
14:22:55 <kennypaul> #info Catherine asks comunnity to verify install instructions.
14:23:20 <kennypaul> #topic integration update
14:23:40 <kennypaul> #info helenychen reviewed her slides
14:24:33 <kennypaul> #info PLEASE free up your lab resources if you are finished or not using them
14:27:59 <kennypaul> #action kennypaul move ptl meeting to tuesday the 29th
14:28:53 <kennypaul> #info helen asks for contact info for all projects for weekend coverage
14:30:38 <kennypaul> #info one each fails HEAT and OOM test
14:30:48 <kennypaul> ?#info both vfc
14:32:56 <kennypaul> #info resiliency testing an track for meeting level 1 & 2 requirements for OOM
14:33:29 <kennypaul> #info 72 hr burn in passed with HEAT
14:35:54 <kennypaul> #info vFW & vDNS passed, vCPE one last test to pass, VoLTE is close
14:37:21 <kennypaul> #info manual scaling VF-C - requirements gap identified by project team - to be fixed in Casablanca
14:38:18 <kennypaul> #info change management- blocked on SO publish event to DMaaP
14:38:26 <kennypaul> #info having daily calls
14:40:05 <kennypaul> #info manual scaling will not be considered part of Beijing
14:41:08 <kennypaul> #info issue was xfunc requirements regarding info getting passed to APPC not socialized between projeuct
14:41:43 <kennypaul> #info is not an issue with APPC
14:44:50 <kennypaul> #agreed manual scaling with VFC will be in Beijing. manual scaling with APPC will be in Casablanca
14:50:45 <kennypaul> #info HPA testing on HEAT - once done it complete. if issues with OOM, not an HPA issue.
14:51:21 <kennypaul> #info Ramki suggests patch release. Alex agrees.
14:55:37 <kennypaul> #info maturity testing - stability blocked by new SDNC issue
14:55:51 <kennypaul> #info resiliance also
14:56:05 <kennypaul> #info stsability after re,ease
14:56:48 <kennypaul> #topic Known Vulnerability Analysis
14:57:00 <kennypaul> #info all projects have passed review
14:57:17 <kennypaul> #info will be a template for release notes
14:57:44 <kennypaul> #info SteveT reviewed his slides
14:59:47 <kennypaul> #info PTL Self Commit
15:00:29 <kennypaul> #info see comments on TSC meeting minute wiki
15:00:53 <kennypaul> #info need ate least one person that can do a self commit per peoject.
15:15:27 <kennypaul> #info much discussion regarding technical, secutiry, and timing of channging this
15:16:25 <kennypaul> #info may impact CII badging
15:20:23 <kennypaul> #startvote Does the TSC approve temporarily allowing the capability for all registered committers to self commit for 3 weeks with the policy of only being exercised by the PTL under exceptional circumstances for Highest Priority bugs? +1, 0, -1
15:20:23 <collabot`> Begin voting on: Does the TSC approve temporarily allowing  the  capability for all registered committers to  self commit for 3 weeks with the policy of only being exercised by the PTL under exceptional circumstances for Highest Priority bugs? Valid vote options are +1, 0, -1.
15:20:23 <collabot`> Vote using '#vote OPTION'. Only your last vote counts.
15:20:30 <cdonley> #vote +1
15:20:36 <EricDebeau> #vote +1
15:20:40 <kennypaul> #vote +1
15:20:47 <RannyHaiby> #vote +1
15:20:54 <SteveT> #vote +1
15:20:56 <kennypaul> #info my vote representing Catherine
15:20:59 <gilesheron> #vote +1
15:21:16 <Xiaojun> #vote +1
15:22:56 <gildaslanilis> #vote +1
15:23:21 <gildaslanilis> #info Gildas voted for Yoav Kluger
15:24:33 <phrobb> #vote +1
15:24:41 <Srini> #vote +1 (on behalf of Rajesh, Intel)
15:24:41 <collabot`> Srini: +1 (on behalf of Rajesh, Intel) is not a valid option. Valid options are +1, 0, -1.
15:25:00 <phrobb> #info phrobb votes for Murate Turpcu of Turk Telecom
15:25:04 <Srini> #vote +1
15:25:05 <xinhuili> #vote +1
15:25:21 <Susana> #vote +1
15:25:41 <helenychen> #vote +1
15:25:58 <kennypaul> #endvote
15:25:58 <collabot`> Voted on "Does the TSC approve temporarily allowing  the  capability for all registered committers to  self commit for 3 weeks with the policy of only being exercised by the PTL under exceptional circumstances for Highest Priority bugs?" Results are
15:25:58 <collabot`> +1 (13): Susana, Xiaojun, phrobb, gilesheron, xinhuili, SteveT, Srini, helenychen, kennypaul, cdonley, gildaslanilis, RannyHaiby, EricDebeau
15:26:00 <phrobb> #info Helen Chen votes for Lingli Deng of China Mobile
15:26:09 <helenychen> #info Helen's vote is for Lingli (China Mobile)
15:26:28 <kennypaul> #agreed TSC approve temporarily allowing the capability for all registered committers to self commit for 3 weeks with the policy of only being exercised by the PTL under exceptional circumstances for Highest Priority bugs
15:27:17 <kennypaul> #action kennypaul open ticket
15:27:53 <EricDebeau> #info For Integration and OOM project, we face some issues due to a lack of committer sin Europe zone and we may loose time to correct some bugs.
15:29:16 <kennypaul> #topic documentation
15:29:45 <kennypaul> #info getting linkcheck fixed is highly important
15:29:59 <kennypaul> #info jiras being opened
15:33:06 <kennypaul> #topic Casablancadeveloper event
15:33:42 <kennypaul> #info registration is low - please register for the event so that we can plan accordingly.
15:33:51 <kennypaul> #endmeeting



Zoom Chat Log 

06:58:35 From Lingli to Kenny Paul (Linux Foundation) (Privately) : Hi Kenny
06:59:42 From Lingli to Kenny Paul (Linux Foundation) (Privately) : thank you Kenny
07:01:51 From Jason Hunt : Unfortunately, I can only be on for 45 minutes… happy to respond to any votes via e-mail.
07:04:53 From Murat Turpcu ( Turk Telekom) : hi kenny i am here
07:14:13 From Kenny Paul (Linux Foundation) : I’ve logged you Murat
07:15:44 From Kenny Paul (Linux Foundation) : link to groups.io slides https://wiki.onap.org/download/attachments/25434845/What%20is%20Groups.io_%20Quick%20feature%20guide.pdf?version=1&modificationDate=1527165725000&api=v2
07:17:41 From Michael O'Brien (Amdocs, Log) : I moved a couple defects to C*
07:19:45 From Pamela Dragosh : Release jobs will fail because STAGING is NOT available anymore. When you upgrade a SNAPSHOT version to get ready for artifact release this is a side effect.
07:27:10 From Michael O'Brien (Amdocs, Log) : 9-10 EDT GMT-4
07:27:57 From Bin Yang (Wind River) : Pausing the instance does not release any resource for others, please remove them whenever possible to release resource to Integration team
07:30:25 From Brian : Also - if you dont need your openvpn session - log off - we have hit the max sessions several times this week
07:30:49 From Brian : blocks develoeprs dropping in to help troubleshoot/debug
08:00:46 From Lusheng Ji (ATT) : I tried to shelve VM instances from CLI successfully, seeing the state becoming shelved and offloaded. I understand this would release resources but still keep the VM. But I never succeeded in unshelving these VMs. Any suggestions?
08:07:07 From Marcus Williams (Intel) : I think self-commit is a horrible idea. It's a substitute for relationships between committers, or not having enough committers and opens up a lot of code quality issues for expediency sake.
08:07:45 From Kenny Paul (Linux Foundation) : #startvote Does the TSC approve allowing a PTL committer the capability for a self commit?
08:07:56 From Kenny Paul (Linux Foundation) : NOT A VOTE YTET
08:11:05 From Don Levy : will there be any checks on the process?
08:11:27 From Kenny Paul (Linux Foundation) : Does the TSC approve temporarily allowing PTLs the capability for a self commit for 3 weeks ?
08:17:42 From Kenny Paul (Linux Foundation) : Does the TSC approve temporarily allowing the capability for all registered committers self commit for 3 weeks?
08:18:27 From Kenny Paul (Linux Foundation) : #startvote Does the TSC approve temporarily allowing the capability for all registered committers to self commit for 3 weeks with the policy of only being exercized by the PTL?
08:19:33 From Kenny Paul (Linux Foundation) : Does the TSC approve temporarily allowing the capability for all registered committers to self commit for 3 weeks with the policy of only being exercized by the PTL under exceptional circumstances for Highest Priority bugs?
08:20:14 From Murat Turpcu ( Turk Telekom) : can create a policy to check the self commits in a week or so
08:20:30 From Murat Turpcu ( Turk Telekom) : or after this 3 weeks
08:22:32 From Michael O'Brien (Amdocs, Log) : +1
08:22:36 From Yoav Kluger (Amdocs) : +1
08:24:06 From Murat Turpcu ( Turk Telekom) : +1
08:24:32 From Susana (VF) : +1
08:24:50 From Lingli : +1
08:30:23 From Michael O'Brien (Amdocs, Log) : dropping for other call - thanks guys


Zoom Attendee log 



1 Comment

  1. Uploading notes from newsgroup traffic on the self commit issue - prior to TSC meeting start - for context

    +1

    Guys,

       Very good responses – we can edit the proposal below up until the meeting together

    https://wiki.onap.org/display/DW/TSC+2018-05-24+Meeting+Agenda

        bottom line is I think we should have left the PTLs out of all the restrictions – I was wrong about the blanket +2 rule - and when I started doing more commit work, gerrit, Jenkins work and butting up against the LF infrastructure - I realized the overhead burden of not being able to do things right away.

        I would say the #1 issue is the inability to verify an infrastructure change or kick off a build/experiment – waiting days to verify something that could be done in 30 min.

         Based on all the issues/traffic I have seen – we should trust PTLs and exempt from this rule all the time – so at least 1 person is “Agile”


    I do not think there should be limitations on the PTL at least.

    In the end we are responsible for the project and we are the ones who usually get called to fix high priority bugs and build issue and versioning issues.

    As such the ptl should have the freedom to self-commit.

    If a ptl abuses this right he will be called on it by his community and the TCS.

    As such I think.

    “PTL’s should be excluded from the self-commit rule because of their position and to allow the projects to progress successfully and not block changes needed for the project operation. ”


     think maybe we need to group self-commits into 2 categories:


    • Self-commits related to managing the releases (version changed due to dependencies rolling the version, changes needed because LF or Integration did something and now application builds, CSIT, etc.,, are failing)  - these should always be allowed, not just in the last 3 weeks.
    • Self-commits to address code fixes for blocking issues, those we can try to time-box during critical points in the release timeline.


    I do, however, think that more generally, we PTLs – the community’s elected technical leads – should be empowered to make these kinds of tactical decisions, with the TSC taking more of an advise and consent role. 

    “Within the last 3 weeks of a release, enable PTLs to do self commits. This should be used for small changes to both quickly resolve release blocking bugs and to deal with release packaging reference changes as we build for relesae instead of SNAPSHOT.”

    “Allow all PTLs to do self commits for small changes like test blocking bugs and fixes for broken builds.”