Do not manually mark this bug as "Fixed" when builder is working as intended (WAI) on LUCI. Migration is not complete until builder is solely on LUCI.

The LUCI migration app will mark the bug status to "Fixed" automatically when the BuildBot builder equivalent is turned down. 

Compile takes too long, the swarming job times out.

Likely needs a beefier bot. Jao-ke, do you mind taking a quick look at this? 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/667971446598f2b44cab5848115bd97d9a812b92

commit 667971446598f2b44cab5848115bd97d9a812b92
Author: Jao-ke Chin-Lee <jchinlee@chromium.org>
Date: Wed Jun 06 22:36:58 2018

[cr-buildbucket] Clear cores for linux_chromium_cfi_rel_ng to test beefier machines.

BUG= 731417 

Change-Id: Ic2e8071ce052e2d8ef5b6f0c508f6f62da7624fc
Reviewed-on: https://chromium-review.googlesource.com/1089913
Reviewed-by: Ryan Tseng <hinoka@chromium.org>
Commit-Queue: Jao-ke Chin-Lee <jchinlee@chromium.org>
Cr-Commit-Position: refs/heads/master@{#565065}
[modify] https://crrev.com/667971446598f2b44cab5848115bd97d9a812b92/infra/config/global/cr-buildbucket.cfg

Preliminary tests indicate that this builder is now behaving correctly, mod running out of swarming capacity. Any chance we could boost that? Otherwise it'll be a pain for assessment of other mirrors.

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/infra/infra_internal/+/221d4cdfc3f05e9027f1bd67cf3c845f5b0a5934

commit 221d4cdfc3f05e9027f1bd67cf3c845f5b0a5934
Author: Jao-ke Chin-Lee <jchinlee@google.com>
Date: Wed Jun 13 00:17:25 2018

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/infra/infra_internal/+/221d4cdfc3f05e9027f1bd67cf3c845f5b0a5934

commit 221d4cdfc3f05e9027f1bd67cf3c845f5b0a5934
Author: Jao-ke Chin-Lee <jchinlee@google.com>
Date: Wed Jun 13 00:17:25 2018

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/2fac1e71717c1d222bae96b391ac896a2be52a09

commit 2fac1e71717c1d222bae96b391ac896a2be52a09
Author: Jao-ke Chin-Lee <jchinlee@chromium.org>
Date: Wed Jun 13 00:27:16 2018

[cr-buildbucket] Beef up linux_chromium_cfi_rel_ng, linux_upload_clang.

BUG= 731417 , 731456 

Change-Id: I9fbb2c34f149380a9b98dfc1bfc7ad56041f6fd6
Reviewed-on: https://chromium-review.googlesource.com/1098307
Reviewed-by: Ryan Tseng <hinoka@chromium.org>
Commit-Queue: Jao-ke Chin-Lee <jchinlee@chromium.org>
Cr-Commit-Position: refs/heads/master@{#566647}
[modify] https://crrev.com/2fac1e71717c1d222bae96b391ac896a2be52a09/infra/config/global/cr-buildbucket.cfg

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/infra/infra_internal/+/221d4cdfc3f05e9027f1bd67cf3c845f5b0a5934

commit 221d4cdfc3f05e9027f1bd67cf3c845f5b0a5934
Author: Jao-ke Chin-Lee <jchinlee@google.com>
Date: Wed Jun 13 00:17:25 2018

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/2fac1e71717c1d222bae96b391ac896a2be52a09

commit 2fac1e71717c1d222bae96b391ac896a2be52a09
Author: Jao-ke Chin-Lee <jchinlee@chromium.org>
Date: Wed Jun 13 00:27:16 2018

[cr-buildbucket] Beef up linux_chromium_cfi_rel_ng, linux_upload_clang.

BUG= 731417 , 731456 

Change-Id: I9fbb2c34f149380a9b98dfc1bfc7ad56041f6fd6
Reviewed-on: https://chromium-review.googlesource.com/1098307
Reviewed-by: Ryan Tseng <hinoka@chromium.org>
Commit-Queue: Jao-ke Chin-Lee <jchinlee@chromium.org>
Cr-Commit-Position: refs/heads/master@{#566647}
[modify] https://crrev.com/2fac1e71717c1d222bae96b391ac896a2be52a09/infra/config/global/cr-buildbucket.cfg

This can be closed right? If so, please close. Thanks!

linux_chromium_cfi_rel_ng seems to be MIA after this, see the purple box onhttps://chromium-review.googlesource.com/c/chromium/src/+/1232590

I'm not sure I see what you're seeing --- the Buildbot linux_chromium_cfi_rel_ng is purple because it's been deleted (and I should remove it from the Gerrit plugin), but the LUCI linux_chromium_cfi_rel_ng seems to be green? Could you describe a bit more what you see? Thank you!

If I go to https://chromium-review.googlesource.com/c/chromium/src/+/1232590 , click CHOOSE TRYJOBS, type "cfi" into the search box, pick the linux one, and click "ok", I used to get a working CFI try job. Now I just get a purple box. It's possible you just need to update the gerrit config to use the new bot.

Thanks for the description!

It indeed seems like there's something wonky with how the plugin is interfacing with the builders --- it /should/ be selecting the builder marked as production (which is indeed LUCI in this case), but isn't for some reason.

cc'ing Nodir as plugin owner

can you still reproduce this? I suspect your localStorage cache was stale. It has 10m timeout though...

when you type "cfi", is linux_chromium_cfi_rel_ng displayed under "luci.chromium.try" or "master.tryserver.chromium.linux"? Try deleting "luci-migration.appspot.com:tryserver.chromium.linux" entry from localStorage of your browser

Hm, it seems to work now. (I'm on a different computer though.) I guess I was just very unlucky if I hit that 10m window -- otoh I saw it 7 days after this bug was marked fixed -- doesn't that make your theory somewhat unlikely?

i suspected a bug in the plugin, that would explain the theory.

i've dug into this more. The only purple buildbot linux_chromium_cfi_rel_ng build on that CL is http://cr-buildbucket.appspot.com/b/8934938421250097376
note "user_agent:git_cl_try" tag. That build was triggered by git-cl-try and gerrit plugin has nothing to do with it. Indeed, the theory was incorrect.

there may be different reasons for why git-cl-try scheduled a build on buildbot. One of them is -m flag was specified. 

if -m is not specified, git-cl-try derives a bucket for a builder based on
https://builders-map.appspot.com/
currently it says that the cfi builder is mapped to luci.chromium.try

Thanks for pointing this out, updating our docs at https://chromium-review.googlesource.com/c/chromium/src/+/1237113

Sounds like builders-map.appspot.com also needs updating, but we passed an explicit -m

Hm, I sent a try job for that change with `git cl try -B luci.chromium.try -b linux_chromium_cfi_rel_ng`. It's here: https://ci.chromium.org/p/chromium/builders/luci.chromium.try/linux_chromium_cfi_rel_ng/608

https://ci.chromium.org/p/chromium/builders/luci.chromium.try/linux_chromium_cfi_rel_ng claims the builder has two bots, and that one of them is idle. Yet the task has been pending for 3.5 minutes already -- is that expected?

this is not expected.  crbug.com/887620 

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/f218afd820c1e990c1dd01b13efd9d60f1d11420

commit f218afd820c1e990c1dd01b13efd9d60f1d11420
Author: Nico Weber <thakis@chromium.org>
Date: Thu Sep 20 19:11:21 2018

Update clang update docs.

- mention chromium.clang tot waterfall
- linux cfi bot is now on luci

Bug:  731417 
Change-Id: I01b4b01ac35741f66a63f23af78a09d9c1e83495
Reviewed-on: https://chromium-review.googlesource.com/1237113
Reviewed-by: Nodir Turakulov <nodir@chromium.org>
Commit-Queue: Nico Weber <thakis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#592894}
[modify] https://crrev.com/f218afd820c1e990c1dd01b13efd9d60f1d11420/docs/updating_clang.md