Rietveld: schedule builds on buildbucket

Rietveld: schedule builds on buildbucket

Builds are scheduled on behalf of the logged in user.

R=jrobbins@chromium.org, vadimsh@chromium.org, tandrii@chromium.org
BUG=461620
Committed: https://chromium.googlesource.com/infra/infra/+/f20032cd06f4f04915e08ada331d65c100525b8f

[nodir, 2015-09-16 00:31:12]

Patchset #1 (id:1) has been deleted

[nodir, 2015-09-16 00:31:17]

Patchset #1 (id:20001) has been deleted

[nodir, 2015-09-16 01:01:04]

Patchset #2 (id:60001) has been deleted

[nodir, 2015-09-16 01:01:28]

Patchset #1 (id:40001) has been deleted

[nodir, 2015-09-16 01:03:45]

PTAL

Buildbucket build [1] was scheduled by Rietveld. See its "user_agent" tag

[1]:
https://apis-explorer.appspot.com/apis-explorer/?base=https://cr-buildbucket-...

[nodir, 2015-09-16 01:12:12]

With exact latest changes:

Scheduled build:
https://apis-explorer.appspot.com/apis-explorer/?base=https://cr-buildbucket-...

e3379b2-tainted-dot-chromiumcodereview-hr-staging.appspot.com log:

I 2015-09-15 18:08:18.182  200     149 B  1.99 s D 18:08:18.207 I 18:08:18.210 I
18:08:20.171 /250001/edit_flags
  undefined
    D 18:08:18.207 Scheduling 1 builds on behalf of nodir@google.com
    I 18:08:18.210 PUT
https://cr-buildbucket-dev.appspot.com/_ah/api/buildbucket/v1/builds/batch
    I 18:08:20.171 Scheduled buildbucket builds: '9034563659733818336'

cr-buildbucket-dev.appspot.com log:

I 2015-09-15 18:08:19.961  200     996 B  202 ms I 18:08:20.129
/_ah/spi/BuildBucketApi.put_batch
  undefined
    I 18:08:20.129 Build 9034563659733818336 was created by
user:nodir@google.com

[tandrii(chromium), 2015-09-16 11:02:32]

lgtm % nit

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
File appengine/chromium_rietveld/tests/test_buildbucket.py (right):

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
appengine/chromium_rietveld/tests/test_buildbucket.py:212:
email='johndoe@chromium.org')
this is not necessary, you set it up in setUp() now.

[Vadim Sh., 2015-09-16 21:57:38]

lgtm, though I'm not familiar with this code

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
File appengine/chromium_rietveld/codereview/buildbucket.py (right):

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
appengine/chromium_rietveld/codereview/buildbucket.py:263: raise
BuildBucketError('Could not schedule build(s): %r' % error)
how is this displayed to the user?

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
appengine/chromium_rietveld/codereview/buildbucket.py:390: # See tag conventions
http://cr-buildbucket.appspot.com/#docs/conventions
https://

[nodir, 2015-09-17 04:23:42]

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
File appengine/chromium_rietveld/codereview/views_chromium.py (right):

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
appengine/chromium_rietveld/codereview/views_chromium.py:430: if random.random()
<= buildbucket_roll_out_percentage:
Jason: should we use this or gae version splitting? would it be OK to have
version splitting enabled for a couple of days?

[jrobbins, 2015-09-17 06:03:13]

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
File appengine/chromium_rietveld/codereview/views_chromium.py (right):

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
appengine/chromium_rietveld/codereview/views_chromium.py:430: if random.random()
<= buildbucket_roll_out_percentage:
On 2015/09/17 at 04:23:42, nodir wrote:
> Jason: should we use this or gae version splitting? would it be OK to have
version splitting enabled for a couple of days?

Yes.  I think I would favor GAE version splitting because you could adjust the
percentage or disable it any time without needing to alter source code.  So,
let's try that first.

[nodir, 2015-09-17 17:44:17]

Works on staging

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
File appengine/chromium_rietveld/codereview/buildbucket.py (right):

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
appengine/chromium_rietveld/codereview/buildbucket.py:263: raise
BuildBucketError('Could not schedule build(s): %r' % error)
On 2015/09/16 21:57:38, Vadim Sh. wrote:
> how is this displayed to the user?

very rough: on any 500 the old UI shows an alert that does not depend on
response body. It is about the same as opening a CL today. We have retries
though, so it does not happen often

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
appengine/chromium_rietveld/codereview/buildbucket.py:390: # See tag conventions
http://cr-buildbucket.appspot.com/#docs/conventions
On 2015/09/16 21:57:38, Vadim Sh. wrote:
> https://

Updated all doc links to point to the newer MD doc with https

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
File appengine/chromium_rietveld/codereview/views_chromium.py (right):

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
appengine/chromium_rietveld/codereview/views_chromium.py:430: if random.random()
<= buildbucket_roll_out_percentage:
On 2015/09/17 06:03:13, Jason Robbins wrote:
> On 2015/09/17 at 04:23:42, nodir wrote:
> > Jason: should we use this or gae version splitting? would it be OK to have
> version splitting enabled for a couple of days?
> 
> Yes.  I think I would favor GAE version splitting because you could adjust the
> percentage or disable it any time without needing to alter source code.  So,
> let's try that first.

Done.

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
File appengine/chromium_rietveld/tests/test_buildbucket.py (right):

https://codereview.chromium.org/1344253002/diff/80001/appengine/chromium_riet...
appengine/chromium_rietveld/tests/test_buildbucket.py:212:
email='johndoe@chromium.org')
On 2015/09/16 11:02:32, tandrii(chromium) wrote:
> this is not necessary, you set it up in setUp() now.

Done.

[nodir, 2015-09-17 18:15:50]

The CQ bit was checked by nodir@chromium.org

[nodir, 2015-09-17 18:15:50]

The patchset sent to the CQ was uploaded after l-g-t-m from
vadimsh@chromium.org, tandrii@chromium.org
Link to the patchset: https://codereview.chromium.org/1344253002/#ps120001
(title: "remove -dev")

[commit-bot: I haz the power, 2015-09-17 18:16:07]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1344253002/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1344253002/120001

[commit-bot: I haz the power, 2015-09-17 18:19:51]

Committed patchset #3 (id:120001) as
https://chromium.googlesource.com/infra/infra/+/f20032cd06f4f04915e08ada331d6...