New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 725223 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Jun 2017
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux , Windows , Mac
Pri: 2
Type: Task

Blocked on:
issue 630757

Blocking:
issue 725186



Sign in to add a comment

Request slaves for multiple platforms (for tryserver.blink).

Project Member Reported by qyears...@chromium.org, May 22 2017

Issue description

Context:  bug 725186 .

I believe we want to get one new VM slave for each of the following platforms:
 Win7
 Win10
 Mac10.9
 Mac10.10
 Mac10.11
 Mac10.12
 Linux Trusty

And one Mac 10.11 Retina device (like that for bug 630757).

Currently all slaves are "vm*-m4" slaves except for Retina (build115-b4) Current slaves.cfg: https://cs.chromium.org/chromium/build/masters/master.tryserver.blink/slaves.cfg.
 
Blocking: 725186

Comment 2 by vhang@chromium.org, May 23 2017

Owner: friedman@chromium.org
Status: Assigned (was: Unconfirmed)
Hi friedman@ :-) Is there any more information required here, or anything that this is blocked on?
Sorry for the delay.

slave1409-c4 = linux trusty
Project Member

Comment 5 by bugdroid1@chromium.org, May 30 2017

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/infra/infra_internal/+/b6e10c6ed932beca7df1b1d8008e9d954647ae59

commit b6e10c6ed932beca7df1b1d8008e9d954647ae59
Author: Elliott Friedman <friedman@google.com>
Date: Tue May 30 22:23:33 2017

vm1441-m4 = Win7
vm1442-m4 = Win10
vm1443-m4 = Mac10.9
vm1444-m4 = Mac10.10
build2-m4 = Mac10.11
vm1445-m4 = Mac10.12
Project Member

Comment 7 by bugdroid1@chromium.org, May 30 2017

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/chrome-golo/chrome-golo/+/e403a54c5152683ddc9232211782f6485ee53f01

commit e403a54c5152683ddc9232211782f6485ee53f01
Author: Elliott Friedman <friedman@google.com>
Date: Tue May 30 23:43:28 2017

Thanks! build2-m4 is not a Retina device, right?
Project Member

Comment 9 by bugdroid1@chromium.org, May 30 2017

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/chrome-golo/chrome-golo/+/dd4f41fbecb988fe60eae8eb765c9dbd1b1f8c13

commit dd4f41fbecb988fe60eae8eb765c9dbd1b1f8c13
Author: Elliott Friedman <friedman@google.com>
Date: Tue May 30 23:58:55 2017

No, it's a mac mini, but it's giving me some fuss... might be tomorrow before it's ready.
Project Member

Comment 11 by bugdroid1@chromium.org, May 31 2017

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/chrome-golo/chrome-golo/+/aa9fbe675a8822a986de5337165e45f6d2373493

commit aa9fbe675a8822a986de5337165e45f6d2373493
Author: Elliott Friedman <friedman@google.com>
Date: Wed May 31 00:18:42 2017

Project Member

Comment 12 by bugdroid1@chromium.org, May 31 2017

The following revision refers to this bug:
  https://chrome-internal.googlesource.com/chrome-golo/chrome-golo/+/aa9fbe675a8822a986de5337165e45f6d2373493

commit aa9fbe675a8822a986de5337165e45f6d2373493
Author: Elliott Friedman <friedman@google.com>
Date: Wed May 31 00:18:42 2017

Also, I was wondering, if "mac10.11_blink_rel" has one VM slave and one Mac Mini slave (vm641-m4 and build2-m4) might there be a danger of one of them behaving slightly different from the other (in terms of test results)?

I don't know of any specific thing that might be different - but I'm not sure if we should make them both VMs just in case. In general we want this builder to always be consistent with the continuous builder https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Mac10.11 which has a VM slave.
No prob.  Minis are cheaper which is why I gave you one.  I'll get a VM for you.
vm1446-m4 = Mac 10.11
Status: Fixed (was: Assigned)
All set!
Status: Assigned (was: Fixed)
Gah, sorry to re-open this but I think we'll need one more slave for mac10.11_retina_blink_rel - a Mac 10.11 Retina device (like that in bug 630757).

(All of these try bots are used together, so if one of them has only one slave then that's the bottleneck).

mac10.11_retina_blink_rel currently has one slave:
https://build.chromium.org/p/tryserver.blink/buildslaves/build115-b4
  product name: MacBookPro11,5
  architecture: x86_64
  processor count: 8
  os family: Darwin
  os version: OS X 10.11.6 (15G31)

Are there any more MacBook Pros available to be allocated?
Cc: kbr@chromium.org
kbr: I was told last time we had a req like this (for a retina bot) we were allowed to remove it from GPU swarming... can we do the same here or should I get a new one going?

Comment 19 by kbr@chromium.org, May 31 2017

Blockedon: 630757
Is MacBookPro11,5 one of the MacBook Pro Retinas with an AMD GPU? It appears to be. If so, yes, please feel free to grab one from Swarming; there are plenty there.

There is a shortage of the older NVIDIA based MacBook Pros and we're occasionally hitting capacity on them, so I'd appreciate it if we didn't take more out of Swarming.

Status: Fixed (was: Assigned)
Thanks kbr.  Looks like build116-b4 meets those dimensions (AMD GPU).  I'll remove it from swarming.
Excellent, thanks :-)
build116-b4 just setup as a 10.11 buildbot and was deleted from swarming.

Sign in to add a comment