Rebalance Mac slaves in preparation of switching to 8-core VMs.

Rebalance Mac slaves in preparation of switching to 8-core VMs.

For release, looking at the stats pages the two testers for each config should be roughly equal. For debug, it appears that this simplified configuration will make the second tester ~10 minutes slower than the first one. However they should still cycle faster the current config on the new machine.

This rebalancing should hopefully be a temporary step until we use swarming for the main waterfall.

BUG=373847
NOTRY=true

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=290714

[iannucci, 2014-08-19 21:05:20]

lgtm, though it's pretty hard to follow this diff.

[jam, 2014-08-19 21:11:35]

On 2014/08/19 21:05:20, iannucci wrote:
> lgtm, though it's pretty hard to follow this diff.

I had checked it three times, checked again now :)

So should I land this now and it'll take effect after Peter commits his change
and someone (Peter? trooper?) restarts the master?

I'll keep an eye on the running time after and rebalance as necessary.

Thanks

[iannucci, 2014-08-19 21:16:41]

On 2014/08/19 21:11:35, jam wrote:
> On 2014/08/19 21:05:20, iannucci wrote:
> > lgtm, though it's pretty hard to follow this diff.
> 
> I had checked it three times, checked again now :)
> 
> So should I land this now and it'll take effect after Peter commits his change
> and someone (Peter? trooper?) restarts the master?
> 

I think this should take effect as soon as it lands :)


> I'll keep an eye on the running time after and rebalance as necessary.
> 
> Thanks

[jam, 2014-08-19 21:20:44]

On 2014/08/19 21:16:41, iannucci wrote:
> On 2014/08/19 21:11:35, jam wrote:
> > On 2014/08/19 21:05:20, iannucci wrote:
> > > lgtm, though it's pretty hard to follow this diff.
> > 
> > I had checked it three times, checked again now :)
> > 
> > So should I land this now and it'll take effect after Peter commits his
change
> > and someone (Peter? trooper?) restarts the master?
> > 
> 
> I think this should take effect as soon as it lands :)

oh, well that's good in that it'll make rebalancing easier!

but in that case we probably don't want to land this until right before Peter
lands his cl, otherwise the cycle times will get really slow. Does his change
require a master restart?

[iannucci, 2014-08-19 21:22:30]

On 2014/08/19 21:20:44, jam wrote:
> On 2014/08/19 21:16:41, iannucci wrote:
> > On 2014/08/19 21:11:35, jam wrote:
> > > On 2014/08/19 21:05:20, iannucci wrote:
> > > > lgtm, though it's pretty hard to follow this diff.
> > > 
> > > I had checked it three times, checked again now :)
> > > 
> > > So should I land this now and it'll take effect after Peter commits his
> change
> > > and someone (Peter? trooper?) restarts the master?
> > > 
> > 
> > I think this should take effect as soon as it lands :)
> 
> oh, well that's good in that it'll make rebalancing easier!
> 
> but in that case we probably don't want to land this until right before Peter
> lands his cl, otherwise the cycle times will get really slow. Does his change
> require a master restart?

Yes, unfortunately, moving slaves around still requires a master restart :/

[aneeshm, 2014-08-20 00:08:12]

The CQ bit was checked by aneeshm@chromium.org

[commit-bot: I haz the power, 2014-08-20 00:10:42]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/jam@chromium.org/486923004/40001

[jam, 2014-08-20 00:25:29]

The CQ bit was unchecked by jam@chromium.org

[jam, 2014-08-20 00:25:30]

The CQ bit was checked by jam@chromium.org

[commit-bot: I haz the power, 2014-08-20 00:25:57]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/jam@chromium.org/486923004/40001

[commit-bot: I haz the power, 2014-08-20 00:27:03]

Committed patchset #3 (40001) as 290714