akeshet@ - assigning this to you so that it doesn't sit untriaged forever. It's not urgent but it should also be pretty simple and the continued existence of the defucnt builder provides noise and confison to sheriffs and gardeners. Plus its a resource we can repurpose for something else.

This continues to cause confusion for new gardeners. This should be relatively easy, shouldn't it?

+michaelpg@ - If things are relatively calm this week maybe you can help out here and see if you can find where we add it to the waterfall and remove it at least? 

There will also be an associated buildslave but recovering that is a P3 as far as I am concerned :)

There's some special logic in the master.chromeos.chrome/master.cfg file for the LKGM builder. After some investigation, I think we're safe to remove it.

The function dontMergeLKGMChanges was a fix for issue 227312: When chromeos/CHROMEOS_LKGM was updated, the next build would often include that change among several other Chrome changes. So the Perf builder would be testing Chrome changes and a new Chrome OS version, making it hard to isolate whether failures were caused by the Chrome changes or the new CrOS version.

So the idea was to build on all changes in trunk *except* when they include an LKGM update, in which case it built the LKGM change itself.

This code was added in https://chromereviews.googleplex.com/7806014 and improved in https://chromereviews.googleplex.com/8187013.

It's safe to remove this code because it only affected the perf builders (which were later removed in issue 434765) and the LKGM builder (which we're removing and which only runs nightly, so I'm not even sure how the merging process worked for that).

CL to remove the LKGM builder from the master and slave configs: https://chrome-internal-review.googlesource.com/c/chrome/tools/build/+/528843

We'll need to remove or free up cros-beefy429-c2 once we've verified that things are still working after turning down the bot.

After you remove it from the waterfalls list of build slaves, pass it to be
to free up the GCE machine.

I'm not working on this. I don't think this qualifies as P1. I agree with stevenjb's ownership suggestion.

akeshet@ - Could you please find the time at least to review https://chrome-internal-review.googlesource.com/c/chrome/tools/build/+/528843, or suggest someone else to review it?

michaelpg@ - Thanks for moving this forward. Assuming you can get your CL reviewed, please reassign this to dgarret@ to free cros-beefy429-c2.

> akeshet@ - Could you please find the time at least to review https://chrome-internal-review.googlesource.com/c/chrome/tools/build/+/528843, or suggest someone else to review it?

Sorry for the delay. Please use the Assignee field for timely review.

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/chrome/tools/build/+/21d36a7c9daba6dee8ab14e1b72e8622a0a6da91

commit 21d36a7c9daba6dee8ab14e1b72e8622a0a6da91
Author: Michael Giuffrida <michaelpg@chromium.org>
Date: Thu Jan 04 21:52:05 2018

I've just scheduled a restart, it should be finished in ~ 20 minutes.

It seems to be gone!

Woot! Thanks everyone!