WebKit^H^H^H^H^H^H Blink PDT Gardening Status 2013-04-04

[Joshua Bell]

As soon as I came in I plopped a roll to 147520 in the CQ to pick up tkent@'s rebaselines.  https://codereview.chromium.org/13461022/  Still waiting on that - it's been an amusing day. :)

src.chromium.org was higglety-pigglety most of the day as the whole world ran "fetch blink" a dozen times, so it wasn't until around 3pm that most try jobs didn't fail on "update". The chromium tree was actually closed most of the day, and the blink tree was unofficially closed (thanks eseidel!) per blink-dev. The only things landing were some test expectation updates, dev tools changes, and some initial dead code cleanup by abarth, leviw and other usual suspects. dglazkov is poking away at garden-o-matic but it's currently blocked by CORS headers. The flakiness dashboard is offline which makes diagnosing failures rather tricky.

Otherwise, spent most of the day looking for failures between "update" fails, and offering advice on #blink.

Blink r147518 (a dev tools change) appears to cause a dev tools test failure in chromium; unfortunately that didn't become clear until 5:00 or so, when the chromium.webkit waterfall had finally cycled mostly green. I've disabled that chromium test to unblock the roll - crbug.com/226862. I'll give the roll to 147520 one more poke into the CQ before I leave (there's no urgency to get it in manually and it's been a good canary for the health of the CQ). 

For the next gardener: it should be safe to dcommit that rev anything later. Just be gentle with the poor tree.

[Adam Barth]

Should we send these updates to blink-dev from now on?

Adam

[Ojan Vafai]

+1 lets just kill this mailing list.