Just to clarify, you're looking for actual linux builder and tester bots, not linux machines with Android devices, correct?

bugs with Infra label are triaged by chrome-infra team, but the request (add builder) is not something chrome-infra does because it is project-specific (as opposed to e.g. get slave machines). Remove "Infra" label?

nodir: I believe this is a request for the entire bot to be set up. This would include (but wouldn't be limited to) slave allocation. Leaving the Infra label on for now until skyostil confirms.

Yes, this includes a request for slaves; and yes linux builder and tester bots are required. No Android devices.

That's correct, thanks for confirming Juan.

Which waterfall do you want these bots to be on?

I'm guessing https://build.chromium.org/p/chromium.linux/console on the "linux" side. Maybe named "Headless Linux" (like our current "Cast Linux"). WDYT Sami?

there's no way we can jump directly onto the main waterfall.

Then https://build.chromium.org/p/chromium.fyi/console ? Or what do you suggest?

Yeah, chromium.fyi would be fine.

Starting with fyi sgtm.

and "headless linux" sounds fine too.

nodir@/jbudorick@, I assume getting the slave machines would be the first step here, followed by configuring them to make the right kind of builds? I'm just wondering who can help get things in motion here? :)

Sorry, meant to file a blocking bug to infra-labs.

skyostil@ to get progress:

1. please file a new bug for new hardware to label:Infra-Labs (OS (Linux?), Bitness, Waterfall (chromium.fyi?)). Ideally find existing slave that'd fit your needs and point to it (see https://build.chromium.org/p/chromium.fyi/buildslaves)
2. create or edit existing recipe to execute tests you want. You don't have to wait for the bot to be allocated to start this, but you can start with dummy recipe which say just compiles chromium and does nothing to iterate later (after #3).
3. wait for your bot to be allocated, make a change to master config (see https://chromium.googlesource.com/chromium/tools/build/+/master/masters/master.chromium.fyi) to add your slave (slaves.cfg) and assign it a new builder which would execute your recipe.
4. iterate with your recipe until you get the bot build what you want.

filed the hardware bug on Friday: https://bugs.chromium.org/p/chromium/issues/detail?id=596097

Assigning to skyostil to get this off the Untriaged Infra queue.

I can take this to help on the master/1st-recipe side.

'is_headless' isn't necessarily a great choice for a build flag. We're trying to move away from large cross-cutting flags (like is_chromecast) and to flags that are more focused on individual features. There is another mechanism (//build/args template files) that we can use to specify that you want to build a particular configuration that we're using for blimp, and we should look into doing that instead.

Good point. I'll look into replacing is_headless with a template. Do we need to block adding the tester on that though? It'd be good to have some coverage in the mean time.

Patch for removing is_headless: https://codereview.chromium.org/1845473003/

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/tools/build.git/+/7d1003c8cd72682b2d907f3c5534e5fbcf206706

commit 7d1003c8cd72682b2d907f3c5534e5fbcf206706
Author: perezju@chromium.org <perezju@chromium.org>
Date: Fri Apr 01 10:53:27 2016

Recipe for Headless Linux bot

BUG= 594108 

Review URL: https://codereview.chromium.org/1840063002

git-svn-id: svn://svn.chromium.org/chrome/trunk/tools/build@299628 0039d316-1c4b-4281-b951-d872f2087c98

[modify] https://crrev.com/7d1003c8cd72682b2d907f3c5534e5fbcf206706/masters/master.chromium.fyi/master.cfg
[modify] https://crrev.com/7d1003c8cd72682b2d907f3c5534e5fbcf206706/masters/master.chromium.fyi/slaves.cfg
[modify] https://crrev.com/7d1003c8cd72682b2d907f3c5534e5fbcf206706/scripts/slave/recipe_modules/chromium_tests/chromium_fyi.py
[add] https://crrev.com/7d1003c8cd72682b2d907f3c5534e5fbcf206706/scripts/slave/recipes/chromium.expected/full_chromium_fyi_Headless_Linux__dbg_.json

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/b8b863a043b50ec783dc9cacd683700802a27130

commit b8b863a043b50ec783dc9cacd683700802a27130
Author: perezju <perezju@chromium.org>
Date: Fri Apr 15 09:20:00 2016

Add tests for Headless Linux bot

BUG= 594108 

Review URL: https://codereview.chromium.org/1843623002

Cr-Commit-Position: refs/heads/master@{#387563}

[add] https://crrev.com/b8b863a043b50ec783dc9cacd683700802a27130/build/args/bots/chromium.fyi/headless_linux_dbg.gn
[modify] https://crrev.com/b8b863a043b50ec783dc9cacd683700802a27130/testing/buildbot/chromium.fyi.json
[modify] https://crrev.com/b8b863a043b50ec783dc9cacd683700802a27130/testing/buildbot/gn_isolate_map.pyl
[modify] https://crrev.com/b8b863a043b50ec783dc9cacd683700802a27130/tools/mb/mb_config.pyl

The build config for the bot landed and we can already see the first couple of attempts (after being totally broken due to no build config) here:
https://build.chromium.org/p/chromium.fyi/builders/Headless%20Linux%20%28dbg%29

But it fails to compile with:
ninja: error: unknown target 'headless_libs', did you mean 'headless_lib'?

Oops, probably my bad. Will fire up a CL to get this fixed.

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/1add7215b690ec6a34145c8a74a95714efa9829a

commit 1add7215b690ec6a34145c8a74a95714efa9829a
Author: perezju <perezju@chromium.org>
Date: Fri Apr 15 14:19:54 2016

Fix build config of Headless bot

BUG= 594108 

Review URL: https://codereview.chromium.org/1891153003

Cr-Commit-Position: refs/heads/master@{#387596}

[modify] https://crrev.com/1add7215b690ec6a34145c8a74a95714efa9829a/testing/buildbot/chromium.fyi.json

And the results are in: https://build.chromium.org/p/chromium.fyi/builders/Headless%20Linux%20%28dbg%29/builds/968

Thanks for the help everyone!

Awesome!