potiuk commented on issue #4938: [AIRFLOW-4117] Multi-staging Image - Travis CI 
tests [Step 3/3]
URL: https://github.com/apache/airflow/pull/4938#issuecomment-507757213
 
 
   Re: changes in tests/* I think some of them were because of the pylint 
started to detect them (but it is likely it was because of the pre-commit 
hook). I will try to revert all of those and see if they fail in the new 
environment (luckily it is super easy to exact-reproduce it with the pylint 
docker environment using ci-slim docker). Previously some of the local pylint 
errors were not showing up in CI but they were showing up locally due to 
different pylint version and local configuration. 
   
   I am not sure where this came from 
https://github.com/apache/airflow/pull/4938/files/b32a836fac5eeea85f09ad5460287c122911be64#diff-9753408ee5d93e5869f8ad8a047776e0R64
 :). Will revert.
   
   Some other changes are actually related to the new environment changes:
    
   - Change of the run_unit_tests.sh script name: 
https://github.com/apache/airflow/pull/4938/files/b32a836fac5eeea85f09ad5460287c122911be64#diff-63f689d4fbfcc88aeebc63c85611ef41R83
  - I changed the "run_unit_tests.sh" name "run_tests.sh" which is far more 
appropriate (those are not only unit tests). So I figured updating source code 
comment that refers to it together with name change is a good practice :).
   
   -  AIRFLOW_SOURCES_DIR was a wrong name in this comment. I changed it (but 
wrongly) it should be AIRFLOW_SOURCES instead (I cleaned up and standardised 
use of AIRFLOW_SOURCE/AIRFLOW_ROOT variables as it was quite messy). It is a 
comment only but I will fix it -   
https://github.com/apache/airflow/pull/4938/files/b32a836fac5eeea85f09ad5460287c122911be64#diff-e5927ea2fff7d7b3c1de8d69fffd1f35R43
   
   
   - the change with impersonation : 
https://github.com/apache/airflow/pull/4938/files/b32a836fac5eeea85f09ad5460287c122911be64#diff-86b0f172baeb73219800925605a34b07R42
 
   
   This was  is the problem was that in the new environment this test fails 
because the whole test environment was modified to support this case. It was 
one of the first failing test in the new CI environment which I fixed. 
   
   I could not easily do exactly the same approach in the exact same way as it 
was done in the old environment. For one in the new environment everything runs 
as root. This is another fix (along the way) because of the way local sources 
are mounted on Linux when you try to run tests locally. Current way when 
sources are mounted and airflow user is used, works on Mac but when you try to 
run it on Linux with docker it breaks in cases of generated npm code etc. In 
the new environment starting docker with mounted local sources works also on 
Linux .
   
   Originally it was done in test_run script and it was very difficult to find 
out that it is related to these impersonation tests. And it was actually quite 
bad approach as it was modifying the dag folder for o+ access - ALWAYS - just 
to handle this one case. 
   
   I did not want to propagate the problem to the new environment - I instead 
decided to fix it in the tests - it seems much more appropriate. I can try to 
make separate PR out of that but then it would mean I will have to do quite 
some work to re-introduce the bad environment configuration  back in the new 
one, apply the fix and remove the workaround. I think it is not worth it - 
really. WDYT?
   
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

Reply via email to