QA/Release Meeting, October 2nd 2006
Tinderbox action plans
See Dan's meeting report and suggestions
for references to point numbers. See also Bear's answers
- point 1: we agree to put that policy in place for the intermittent/unreliably failing tests. For all reliably failing/100% failures, we will still apply the "roll back" policy
- point 2:
- implement sub points B and E: Rewrite ChandlerTestLib so that most of the validation is done automatically by the framework. Big work item though. QA is going to write a proposal for this.
- point F: written, needs to be checked in
- points C and D: we need faster hardware to test for that without it taking too long.
- point G: logged and have a patch. Will commit in alpha5.
- point 3:
- point A: yes. Will do.
- point B: Heikki and Stearns did work on this, mostly alleviated.
- point C: Need to log a bug on that.
- point D: need discussion on the list
- point E: need discussion on the list
- point 4: app_ns magic will need work from John, the scripting layer is rather confusing
- point 5: get faster PC would help (5xtimes faster is possible). The PPC Mac will be become the bottleneck. Heikki working on a proposal with IT.
- point 6: logged, assigned to Bear, certainly alpha5
Chandler alpha4 update
- Bug Council this afternoon to trim the bug count
- We moved the date by 2 weeks
- Cosmo 0.5 ready for testing by the end of the week
- Started to do checkpoints
- 02 Oct 2006