Skip to main content

Notice: this Wiki will be going read only early in 2024 and edits will no longer be possible. Please see: for the plan.

Jump to: navigation, search

DTP PMC Meeting, August 12, 2008

Back to DTP PMC Meeting Page


  • Brian Fitzpatrick
  • Linda Chan
  • Sheila Sholars


  • John Graham


  • DTP 1.6.1 M1 issues
    • At the Connectivity Team meeting, we discussed 1.6.1 M1 build and the IMarker breakage that caused the 1.6.1 M2 build to break. Larry has pulled that fix out of the M2 build, but wants us to keep it in the M1 build for now, as the product they are using M1 for is 3.4-based.
    • We (PMC) need to discuss what to do in this case. I believe we need a new build that is 3.3-based, since that's what we're supporting in Ganymede (lowest common denominator). But how we support this with IBM's request is a bit of a quandary.
    • Second issue is that two builds - M1 and M2 were done over the weekend, but only M2 broke with this 3.4 API usage. Have sent an e-mail to Xiaoying to figure out why the M1 build didn't break.
    • Larry also mentioned that he's found a possible bug in DTP related to the change we made to driver definition IDs, but he can't reproduce it in DTP, just with the IBM code. He's still investigating. This may force another respin of 1.6.1 M1
  • DTP 1.6.1 M1 Updates
    • Larry found a fix for the IMarker breakage that works in 3.3 and 3.4 [here]
    • Larry found a fix for the Driver ID issue he found [here]
  • Questions
    • Do we approve these two bugs and say that this is the end of M1? We can't keep extending this indefinitely
    • Ask for one final M1 build from Xiaoying and call it good
  • Open discussion


  • Will approve the two bugs for inclusion in M1 (243829 and 243518)
  • Once fixes are in place, will request a build from Xiaoying for her Wednesday, August 13 (which means we'd see it late today Tuesday, August 12, or early tomorrow Wednesday, August 13)
  • That build will be the "final" build for M1 and undergo IBM testing/approval
  • Unless there is a CRITICAL/STOP-SHIP bug found, we will not do any further M1 builds and will release M1 no later than Wednesday, August 20 (earlier would be better if the IBM testing/approval process could be shortened or include weekend time)
  • If there is a CRITICAL/STOP-SHIP bug, IBM will need to internally patch and go with an internal DTP build
  • DTP 1.6.1 M2 fixes up to August 25, 2008
    • Then we will begin a rampdown phase
    • RC1 - September 2, 2008, with EPP on Sept 8, Server freeze on Sept 9, and public access on Sept 10
    • RC2 - September 9, 2008, with EPP on Sept 15, Server freeze on Sept 16, and public access on Sept 17
    • Ganymede SR1 - September 15, 2008, with EPP on Sept 22, Server freeze on Sept 23, and public access on Sept 24
  • This roughly corresponds to our rampdown suggestion of doing a milestone build (1.6.1 M2) for August 29 the week before SR1 RC1 (probably based on the August 28 integration build).

Action Items

  • Sheila is taking the proposal back to Loic for approval
  • Sheila, Linda, and John should have some high-level goals for DTP Galileo for next week's PMC meeting

Tabled for Later Discussion

Back to the top