Revert "WebKit roll 62844:63172."

Revert "WebKit roll 62844:63172."

This reverts commit 39338994e0b3c8d3220df0ca3b47bd655eafa92a.

TBR=pfeldman@chromium.org

[pfeldman, 2010-07-14 09:47:15]

Why was it reverted? We were lagging 300 revisions behind the ToT which is
unacceptable. There has been a single webkit test failure in debug mode that
should have been rolled to (mentioned in the chromium build status message).

[oshima, 2010-07-14 15:41:59]

On Wed, Jul 14, 2010 at 2:47 AM, <pfeldman@chromium.org> wrote:

> Why was it reverted? We were lagging 300 revisions behind the ToT which is
> unacceptable.


webkit has been rolled by ajwong yesterday.


> There has been a single webkit test failure in debug mode that
> should have been rolled to (mentioned in the chromium build status
> message).
>

I rolled it back because the tree has been broken for very long time and
the fix hasn't been rolled out that time (which broke other tests :().
 The status message wasn't helpful to me since it doesn't say who is working
on
nor when it will be fixed, and I couldn't get answer on irc or chat.
I'm sorry for trouble and confusion, but if the breakage is expected, please
send
sheriffs heads-up so that sheriffs know what to do, especially in the
situation like
yesterday where the tree is broken for more than 24 hours.

- oshima


>
> http://codereview.chromium.org/2951010/show
>

[awong, 2010-07-14 17:38:15]

This was an unfortunate mis-communication in handoff from night to morning.
 I think Oshima is right in that we need to be a bit more proactive about
pushing information regarding touchy changes + strange tree state between
shifts.

Between monday to tuesday PST, we had 2 instances where between shifts, the
tree was closed without enough documentation on what's going on leaving
people all of pamg, oshima, & pfeldman, to have to reconstruct what was
going on from whoever happened to be around.  In the course of this, we
ended up having to revert one major grd change, and one major webkit change,
largely blind.  I sent a mail yesterday describing the hazardous changes of
the day (there were a number) to stave off something like this happening
again between Tues -> Wed PST.

Things look good now.  Sorry for the mess, and thanks everyone for helping
me get the webkit rolled done.

-Albert


On Wed, Jul 14, 2010 at 8:41 AM, oshima <oshima@chromium.org> wrote:

>
>
> On Wed, Jul 14, 2010 at 2:47 AM, <pfeldman@chromium.org> wrote:
>
>> Why was it reverted? We were lagging 300 revisions behind the ToT which is
>> unacceptable.
>
>
> webkit has been rolled by ajwong yesterday.
>
>
>> There has been a single webkit test failure in debug mode that
>> should have been rolled to (mentioned in the chromium build status
>> message).
>>
>
> I rolled it back because the tree has been broken for very long time and
> the fix hasn't been rolled out that time (which broke other tests :().
>  The status message wasn't helpful to me since it doesn't say who is
> working on
> nor when it will be fixed, and I couldn't get answer on irc or chat.
> I'm sorry for trouble and confusion, but if the breakage is expected,
> please send
> sheriffs heads-up so that sheriffs know what to do, especially in the
> situation like
> yesterday where the tree is broken for more than 24 hours.
>
> - oshima
>
>
>>
>> http://codereview.chromium.org/2951010/show
>>
>
>