Rietveld Code Review Tool
Help | Bug tracker | Discussion group | Source code | Sign in
(133)

Issue 29875559: Issue 6651,6220 - add gitlab-ci config (Closed)

Can't Edit
Can't Publish+Mail
Start Review
Created:
1 year, 5 months ago by sergei
Modified:
1 year, 5 months ago
Base URL:
https://github.com/adblockplus/adblockpluscore.git@3fc393a609971e48a4f2e664800da1444d75d970
Visibility:
Public.

Description

# this change depends on the availability of eyeo gitlab-runner # the tag will likely be adjusted # please use https://gitlab.com/eyeo/adblockplus/adblockpluscore/merge_requests/1 to check the CI status of this change and readiness for review

Patch Set 1 #

Total comments: 4
Unified diffs Side-by-side diffs Delta from patch set Stats (+26 lines, -0 lines) Patch
A .gitlab-ci.yml View 1 chunk +26 lines, -0 lines 4 comments Download

Messages

Total messages: 14
sergei
1 year, 5 months ago (2018-09-05 12:32:48 UTC) #1
tlucas
https://codereview.adblockplus.org/29875559/diff/29875560/.gitlab-ci.yml File .gitlab-ci.yml (right): https://codereview.adblockplus.org/29875559/diff/29875560/.gitlab-ci.yml#newcode19 .gitlab-ci.yml:19: tags: You said the tag will likely be adjusted ...
1 year, 5 months ago (2018-09-05 12:45:35 UTC) #2
sergei
https://codereview.adblockplus.org/29875559/diff/29875560/.gitlab-ci.yml File .gitlab-ci.yml (right): https://codereview.adblockplus.org/29875559/diff/29875560/.gitlab-ci.yml#newcode19 .gitlab-ci.yml:19: tags: On 2018/09/05 12:45:35, tlucas wrote: > You said ...
1 year, 5 months ago (2018-09-05 12:49:47 UTC) #3
Sebastian Noack
I wonder whether we could split this up into multiple jobs like we are doing ...
1 year, 5 months ago (2018-09-05 12:49:55 UTC) #4
tlucas
On 2018/09/05 12:49:55, Sebastian Noack wrote: > I wonder whether we could split this up ...
1 year, 5 months ago (2018-09-05 16:22:23 UTC) #5
Sebastian Noack
On 2018/09/05 16:22:23, tlucas wrote: > On 2018/09/05 12:49:55, Sebastian Noack wrote: > > I ...
1 year, 5 months ago (2018-09-05 16:26:49 UTC) #6
Manish Jethani
Not really my area, so if it looks OK to y'all it looks good to ...
1 year, 5 months ago (2018-09-06 07:04:07 UTC) #7
tlucas
LGTM remember to delay the push until our runners are running, i'll let you know ...
1 year, 5 months ago (2018-09-06 07:14:13 UTC) #8
sergei
On 2018/09/05 16:26:49, Sebastian Noack wrote: > On 2018/09/05 16:22:23, tlucas wrote: > > On ...
1 year, 5 months ago (2018-09-06 09:08:50 UTC) #9
tlucas
On 2018/09/06 09:08:50, sergei wrote: > On 2018/09/05 16:26:49, Sebastian Noack wrote: > > On ...
1 year, 5 months ago (2018-09-06 09:33:15 UTC) #10
sergei
OK, since it has successfully passed and the job execution is serialized I would propose ...
1 year, 5 months ago (2018-09-06 19:13:10 UTC) #11
Manish Jethani
On 2018/09/06 19:13:10, sergei wrote: > OK, since it has successfully passed and the job ...
1 year, 5 months ago (2018-09-06 19:36:03 UTC) #12
Sebastian Noack
https://codereview.adblockplus.org/29875559/diff/29875560/.gitlab-ci.yml File .gitlab-ci.yml (right): https://codereview.adblockplus.org/29875559/diff/29875560/.gitlab-ci.yml#newcode19 .gitlab-ci.yml:19: tags: As far as I understand, tags serve to ...
1 year, 5 months ago (2018-09-06 19:36:27 UTC) #13
tlucas
1 year, 5 months ago (2018-09-06 20:06:46 UTC) #14
On 2018/09/06 19:36:27, Sebastian Noack wrote:
> https://codereview.adblockplus.org/29875559/diff/29875560/.gitlab-ci.yml
> File .gitlab-ci.yml (right):
> 
>
https://codereview.adblockplus.org/29875559/diff/29875560/.gitlab-ci.yml#newc...
> .gitlab-ci.yml:19: tags:
> As far as I understand, tags serve to match jobs with runners. So if we'd have
a
> cluster of runners, this would tell the CI to only run this job on a runner
that
> is tagged "nodejs". However, we only have one runner at the moment, and once
we
> add more runners, we don't know yet, how we will tag those. So (unless I
> misunderstood something), I wouldn't configure any tags for now (which would
> also be consistent with the configuration we have for adblockpluschrome).

You understood everything correctly - but there's the following situation right
now:

There is exactly 1 runner registered for
https://gitlab.com/eyeo/adblockplus/adblockpluscore/,
configured with tags "eyeo", "core" (those tags were used in developement iirc)
and "nodejs" (from this very review).

leaving this patch as it is will hence work perfectly fine - however, the runner
is not configured to run untagged jobs (like those in adblockpluschrome are), so
(that's what just happened in #adblockplus) tagged jobs get stuck. We could
change that - but those that know from the top of their head how AND have direct
access (i.e. not me) are in vacation afaict.

Meaning: i'd leave this as it is for now, FWIW "nodejs" is not wrong.
Sign in to reply to this message.

Powered by Google App Engine
RSS Feeds Recent Issues | This issue
This is Rietveld 87257f5