Re: [Boost-testing] out of date results for release branch

Subject: Re: [Boost-testing] out of date results for release branch
From: Eric Niebler (eric_at_[hidden])
Date: 2010-06-25 22:25:33

On 6/25/2010 8:20 PM, Jim Bell wrote:
>> Of the 15 testers on the release branch, only 4 have results from within
>> the past 5 days. That's awful. What's going on?
> 
> I cycle through five tests on machine(s) whose resources I need during
> waking hours, running one release & trunk test in parallel. So my goal is
> one test per platform per week.

I appreciate your work. I wasn't criticizing any of the testers. I was
trying to find out if there was a systemic problem with our
test/reporting infrastructure that is causing our results to be stale. I
guess not. :-(

Perhaps we need to put out a call for more test resources on the release
branch.

> This brings me to a question I've wondered: if it's the release branch, why
> wouldn't a test result be valid until the next release, unless patches are
> also coming out. (Are they?) The tests shouldn't "age" the same way the
> trunk tests should . should they?

The release branch is constantly getting patches merged from trunk.
Those need to be tested; otherwise, developers have no way of knowing
whether they've broken anything on platforms they don't have access to.

Weeks ago, I broke something in xpressive. I fixed it last week, I
think. I say "I think" because very few testers have refreshed their
results, so it's still showing up as busted. :-/

-- 
Eric Niebler
BoostPro Computing
http://www.boostpro.com