Changed asan host detection to use short hostname

Changed asan host detection to use short hostname

mini11-a1.chrome.corp.google.com is actually mini11-a1.perf.chromium.org, but
lets just use short hostnames (hostname -s) so that the domain doesn't matter.

BUG=170629, 174873
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=182815

[Ryan Tseng, 2013-02-15 20:32:10]

Clang versions aren't pinning on lkgr
http://chromegw.corp.google.com/i/chromium.lkgr/builders/Mac%20ASAN%20Debug/b...

[Ryan Tseng, 2013-02-15 20:33:01]

Sent the wrong url
http://build.chromium.org/p/chromium.lkgr/builders/Mac%20ASAN%20Debug/builds/...

[Ilya Sherman, 2013-02-15 20:37:26]

LGTM, thanks.

[commit-bot: I haz the power, 2013-02-15 20:39:01]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/hinoka@chromium.org/12285006/1

[Nico, 2013-02-18 10:30:59]

Did you see https://codereview.chromium.org/12212212/ ?

Looks like things work now, so I guess you guys figured it out somehow?

[Ryan Tseng, 2013-02-18 20:01:49]

Yeah the mini11-a identifies itself as chrome.corp externally, but
perf.chromium internally, which was why it wasn't matching

On Feb 18, 2013, at 2:31 AM, "thakis@chromium.org" <thakis@chromium.org> wrote:

> Did you see https://codereview.chromium.org/12212212/ ?
>
> Looks like things work now, so I guess you guys figured it out somehow?
>
> https://codereview.chromium.org/12285006/

[hans, 2013-02-18 20:20:11]

On 2013/02/18 20:01:49, Ryan T. wrote:
> Yeah the mini11-a identifies itself as chrome.corp externally, but
> perf.chromium internally, which was why it wasn't matching

Thanks for catching and fixing this.

Thinking more about this.. I kind of don't want to change this stuff any more,
but would it be a better idea to check for asan=1 in the GYP_DEFINES for Darwin
builds instead of relying on the hostnames? It would make things a lot less
brittle, and I think it would work for all bots unless I'm missing something?

[Alexander Potapenko, 2013-02-19 15:57:42]

Does update.sh know about GYP_DEFINES at all?

[hans, 2013-02-19 16:04:00]

On 2013/02/19 15:57:42, Alexander Potapenko wrote:
> Does update.sh know about GYP_DEFINES at all?

It's an environment variable, so anyone can look at it. There's already some
code doing that on line 122.

I believe that on all the buildbots, update.sh is run by gclient sync, so
GYP_DEFINES will be set at that point.

[Nico, 2013-02-19 16:13:41]

On Tue, Feb 19, 2013 at 5:04 PM,  <hans@chromium.org> wrote:
> On 2013/02/19 15:57:42, Alexander Potapenko wrote:
>>
>> Does update.sh know about GYP_DEFINES at all?
>
>
> It's an environment variable, so anyone can look at it. There's already some
> code doing that on line 122.
>
> I believe that on all the buildbots, update.sh is run by gclient sync, so
> GYP_DEFINES will be set at that point.

At least some of the bots have a dedicated "update clang" step which
runs update.sh:
http://build.chromium.org/p/chromium.mac/builders/Mac%20Builder%20%28dbg%29/b...

GYP_DEFINES isn't set while this runs.

[hans, 2013-02-19 16:18:57]

On 2013/02/19 16:13:41, Nico wrote:
> On Tue, Feb 19, 2013 at 5:04 PM,  <mailto:hans@chromium.org> wrote:
> > I believe that on all the buildbots, update.sh is run by gclient sync, so
> > GYP_DEFINES will be set at that point.
> 
> At least some of the bots have a dedicated "update clang" step which
> runs update.sh:
>
http://build.chromium.org/p/chromium.mac/builders/Mac%2520Builder%2520%2528db...
> 
> GYP_DEFINES isn't set while this runs.

Ah, ok. My idea won't work then, please ignore. :)