Skip to main content

Notice: This Wiki is now read only and edits are no longer possible. Please see: https://gitlab.eclipse.org/eclipsefdn/helpdesk/-/wikis/Wiki-shutdown-plan for the plan.

Jump to: navigation, search

TPTP-PMC-20080716

Logistics

Attending: Paul, Chris, Harm, AlexA, Joanna, Oliver

  • Eugene had a doctor appointment
  • Joanna has a hard stop in a half hour

Are there any objections to last weeks summary?

  • No

Finalizing 4.5.2 Schedule

Oliver noted that Joanna was going to poll IBM teams to make sure that IBM was okay with schedule

  • Joanna polled teams about what they needed when
  • One particular team would ideally get a drop sometime around August
  • Oliver asks if we can call accomodate by simply finishing 4.5.1 early
    • Paul notes that we have to integrate with new driver from Ganymede fall maintenance for the actual 4.5.1 release
    • Ideally we would need to make any active changes (Unless some APIs change on Ganymede side)
    • The IBM consumer would like to ensure that there are no other changes to our codebase after that milestone.

More detailed discussion

  • One team is requesting patches for Eclipse 3.4 (TPTP 4.5) before Sept deadline for Ganymede fall maintenance
  • August 8 is the quoted date
  • Harm notes that the foundation project with limited resources (which further shrunk this year) can only accomodate so much flex and providing specialized releases and customized back fixes.
  • About 20 defects were requested to be addressed
    • Most of them do not currently have patches available.
    • If we freeze the tree for 4.5.1 and test on the 25th
    • Consumer could get the result on 8/8/8 for their freeze

Oliver notes that we seem to be geting stuck doing back fixes... There is a question about due diligence of testing

  • They (consumer) will need to take onus of part of the testing
  • Oliver asks if there is duplication between tests that we do in TPTP and tests that consumer dones. If so, can we get economies of scale here if they'll start testing more before it goes out?

Oliver asks how did we get here. This defect count is larger than the defect count that Joanna got when she polled consumers during final milestones of TPTP 4.5.

  • There had been general agreement to list of deferrals, GA of 4.5, and the schedule for aligning 4.5.1 with the fall maintenance stream. Resources from both Intel and IBM were reduced for 4.5.1
    • Apparantly this consumer did not pull in their feedback for Joanna's outreach a while back and we got an unexpected request for a supported release in 5 weeks
    • Harm notes that this consimer will actually be getting the distribution from Eclipse update site which would not accomodate an intirim release like this. The interim release is only really useful for testing purposes.
  • It seems that some consumers are probably not pulling in enough of the preliminary milestones of the TPTP components that will consume for advance testing
  • There seems to be some scheduling mismatch between some consumers and TPTP projects. Further discussion around this may be desirable.

We discussed next steps around this

  • Determine which of the 20 are done and which aren't. Get the unmarked ones marked as P1 and do a quick triage
    • Joanna notes that 6 of the 17 are fixed now
      • One of the 6 has had an outstanding request for a week or so to the consumer to verified if it works
    • Harm notes that in depth triaging of the 20 defects is not too necessary. Since the timeline is so tight, we need to just fix those that can be done next week.
    • Chris noted that it is extremely unlikely we would get all 20 done.
    • Either the timeline needs to shift or the number of defects will change
  • After the triage happens, Joanna will drive a discussion with the consuming team.
    • Tell them when we can get them done with our resources
    • Joanna requested a bit of support from Harm on this
    • Copy Oliver on message if this is acceptable to IBM

Summarizing

  • When should we freeze?
    • There is no real objection to freezing next Friday.
    • Joanna asks if we should move all other P1s (besides these 20) to 4.5.2
  • We will try to give consumer a milestone along the way around their 8/8 request. That milestone will become 4.5.1 with only the minimal changes necessary to import from final Ganymede fall maintenance release.
  • Joanna will tell consumer when we will freeze (next friday), and what defects we will probably have fixed by then
    • That is all as much as TPTP can do. There is not resource flexibility to provide any more than that.
    • We will get done what can by next Friday.
      • We then lock down for 4.5.1 (w/ ganymede pull)
      • Anything else desired will get done in 4.5.2 for Feb.

Harm notes that ensuring that we have milestones for our releases can be useful in these situations

  • Chris underscores that if we have milestones, we can point consumers back at the public milestones where consuming products had the lots of opportunities to engage, test, and raise their concerns/issues early.

Finalizing New Leads and Structure

Oliver asks if all that needs doing was collected in one place as requested

  • Paul went and checked the wiki and had several votes to finalize
    • Duyane Morris moving off project and there is a request to move his components to Paul. There were 3+ votes on test project.
      • Paul agrees to take this on.
      • Unanimous approval
    • Jerome is in France and will be working remotely. It will be easier to coordinate if those components have PAul as their designee.
      • Paul agrees to take this on.
      • Unanimous approval
  • Monitoring finalization
    • We need to move designee for monitoring components from AlexN to Joanna.
      • Harm already cleaned up the portal. Just bugzilla owners need movement.
      • Unanimous approval.
  • AI Paul to take care of updating buzilla owners and the matrix on the web.
  • AI Harm to clean up the portal.

Back to the top