airflow-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Björn Pollex <bjoern.pol...@soundcloud.com.INVALID>
Subject Re: Travis CI reporting incorrectly
Date Mon, 07 Jan 2019 07:37:40 GMT
Hijacking this thread for a related problem. A couple of weeks ago I posted to the mailing
list that some of the tests for the plugins manager are failing when I run them locally. Now,
looking at the CI output linked above, it seems none of these tests is even run in CI. I have
no experience with nose for running tests, could it be because the filename doesn’t start
with test_? Is this intentional?

Cheers,

	Björn

> On 6. Jan 2019, at 10:39, Tao Feng <fengtao04@gmail.com> wrote:
> 
> I commit three fixes to the master branch with CI(
> https://travis-ci.org/apache/airflow/builds/475904686?utm_source=github_status&utm_medium=notification).
> Most of the ORMs pass except the mysql one which fails at
> https://github.com/apache/airflow/blob/master/tests/www_rbac/test_views.py#L1465
> introduced in https://github.com/apache/airflow/pull/4407.
> 
> Not sure if it is a flaky test or a real issue.
> 
> On Sat, Jan 5, 2019 at 9:16 PM Tao Feng <fengtao04@gmail.com> wrote:
> 
>> I will submit a pr to fix the test for PR # 4432 .
>> 
>> 
>> 
>> On Sat, Jan 5, 2019 at 7:45 PM Deng Xiaodong <xd.deng.r@gmail.com> wrote:
>> 
>>> Thanks Kaxil for fixing this.
>>> 
>>> Obvious that some errors were introduced into Master branch in the last
>>> few days during which the Travis CI was not working as expected.
>>> 
>>> - Some of the fails were due to a logic error in `delete_dag`, which was
>>> introduced into master branch in PR #4406 six days ago (
>>> https://github.com/apache/airflow/pull/4406, `[AIRFLOW-3599] Removed
>>> DagBag from delete dag`).
>>> 
>>> I already prepared a PR to fix it (
>>> https://github.com/apache/airflow/pull/4445 , [AIRFLOW-3635] Fix
>>> incorrect logic in delete_dag). This fix reduces fails & errors from “8
>>> FAILED, 2 errors” to “3 FAILED, 1 error”. Please review.
>>> 
>>> - Other fails & errors are due to multiple reasons. One of them is that
>>> test was not updated for PR # 4432 ([AIRFLOW-3622] Add ability to pass
>>> hive_conf to HiveToMysqlTransfer
>>> <https://github.com/apache/airflow/commit/089fa65419bb854c1422b2d1fc26cdd1ce7051e9>).
>>> May anyone pick this up?
>>> 
>>> Thanks.
>>> 
>>> 
>>> XD
>>> 
>>> On 5 Jan 2019, at 11:52 PM, Kaxil Naik <kaxilnaik@gmail.com> wrote:
>>> 
>>> It is fixed now. I will merge that PR (
>>> https://github.com/apache/airflow/pull/4441) once the CI build finishes.
>>> 
>>> It is now reporting error correctly. Can someone approve the PR?
>>> 
>>> 
>>> 
>>> On Sat, Jan 5, 2019 at 3:38 PM Naik Kaxil <k.naik@reply.com> wrote:
>>> 
>>> I have that in my PR but it still failed. I made another change and
>>> triggered the CI, let's see
>>> 
>>> 
>>> Kaxil Naik
>>> 
>>> Data Reply
>>> Nova South
>>> 160 Victoria Street, Westminster
>>> London SW1E 5LB - UK
>>> phone: +44 (0)20 7730 6000
>>> k.naik@reply.com
>>> www.reply.com
>>> 
>>> [image: Data Reply]
>>> ------------------------------
>>> *From:* Deng Xiaodong <xd.deng.r@gmail.com>
>>> *Sent:* Saturday, January 5, 2019 3:20:57 PM
>>> *To:* dev@airflow.apache.org
>>> *Cc:* Tao Feng; Driesprong, Fokko
>>> *Subject:* Re: Travis CI reporting incorrectly
>>> 
>>> Hi Kaxil,
>>> 
>>> Not sure if you guys have found the root-cause. If not, I suspect it’s
>>> because of this line “sudo ln -s ${ROOTDIR}
>>> /home/travis/build/apache/incubator-airflow” in
>>> 
>>> https://github.com/apache/airflow/blob/master/scripts/ci/5-run-tests.sh#L97
>>> <https://github.com/apache/airflow/blob/master/scripts/ci/5-run-tests.sh
>>>> .
>>> It was not updated accordingly in PR 4419.
>>> 
>>> Please check.
>>> 
>>> Thanks.
>>> 
>>> 
>>> XD
>>> 
>>> On 5 Jan 2019, at 11:06 PM, Kaxil Naik <kaxilnaik@gmail.com> wrote:
>>> 
>>> An Example is the latest build:
>>> https://travis-ci.org/apache/airflow/jobs/475704671#L7363
>>> 
>>> It had "8 FAILED, 2 errors, 40 skipped (2072 tests passed)"
>>> 
>>> But still the test show as successful.
>>> 
>>> On Sat, Jan 5, 2019 at 1:18 PM Kaxil Naik <kaxilnaik@gmail.com> wrote:
>>> 
>>> Sorry - I meant I think it *is* because of the renaming of the repo.
>>> 
>>> On Sat, Jan 5, 2019 at 1:14 PM Kaxil Naik <kaxilnaik@gmail.com> wrote:
>>> 
>>> Hi guys,
>>> 
>>> Travis CI is reporting the build status incorrectly. Few of the tests
>>> have failed but it still shows the overall test as passed.
>>> 
>>> Request to all the committers to not merge any PR before going to
>>> 
>>> Travis
>>> 
>>> and checking.
>>> 
>>> PS: I don't think it has anything to do with renaming for
>>> incubator-airflow to airflow
>>> 
>>> Regards,
>>> 
>>> *Kaxil Naik*
>>> 
>>> 
>>> 
>>> --
>>> *Kaxil Naik*
>>> *Big Data Consultant *@ *Data Reply UK*
>>> *Certified *Google Cloud Data Engineer | *Certified* Apache Spark &
>>> 
>>> Neo4j
>>> 
>>> Developer
>>> *Phone: *+44 (0) 74820 88992
>>> *LinkedIn*: https://www.linkedin.com/in/kaxil
>>> 
>>> 
>>> 
>>> --
>>> *Kaxil Naik*
>>> *Big Data Consultant *@ *Data Reply UK*
>>> *Certified *Google Cloud Data Engineer | *Certified* Apache Spark & Neo4j
>>> Developer
>>> *Phone: *+44 (0) 74820 88992
>>> *LinkedIn*: https://www.linkedin.com/in/kaxil
>>> 
>>> 
>>> 
>>> 
>>> --
>>> *Kaxil Naik*
>>> *Big Data Consultant *@ *Data Reply UK*
>>> *Certified *Google Cloud Data Engineer | *Certified* Apache Spark & Neo4j
>>> Developer
>>> *Phone: *+44 (0) 74820 88992
>>> *LinkedIn*: https://www.linkedin.com/in/kaxil
>>> 
>>> 
>>> 


Mime
View raw message