[AIRFLOW-6860] Default ignore_first_depends_on_past to True

[zhongjiajie]

---

Issue link: AIRFLOW-6860

Make sure to mark the boxes below before creating PR: [x]

• Description above provides context of the change
• Commit message/PR title starts with [AIRFLOW-NNNN]. AIRFLOW-NNNN = JIRA ID^*
• Unit tests coverage for changes (not needed for documentation changes)
• Commits follow "How to write a good git commit message"
• Relevant documentation is updated including usage instructions.
• I will engage committers as explained in Contribution Workflow Example.

^* For document-only changes commit message can start with [AIRFLOW-XXXX].

---

In case of fundamental code change, Airflow Improvement Proposal (AIP) is needed.
In case of a new dependency, check compliance with the ASF 3rd Party License Policy.
In case of backwards incompatible changes please leave a note in UPDATING.md.
Read the Pull Request Guidelines for more information.

[zhongjiajie]

ref #7490 (comment)

[zhongjiajie]

CI sad with unrelated reason

tests/test_config_templates.py ....                                      [  0%]
tests/test_configuration.py ...................................          [  0%]
tests/test_core.py ........................                              [  1%]
tests/test_core_to_contrib.py .......................................... [  1%]
........................................................................ [  3%]
........................................................................ [  4%]
........................................................................ [  6%]
........................................................................ [  7%]
........................................................................ [  8%]
........................................................................ [ 10%]
........................................................................ [ 11%]
........................................................................ [ 12%]
........................................................................ [ 14%]
........................................................................ [ 15%]
........................................................................ [ 16%]
..........................................                               [ 17%]
tests/test_example_dags.py ..                                            [ 17%]
tests/test_example_dags_system.py .Timeout (0:08:00)!
Thread 0x00007f63a8dbf400 (most recent call first):

[zhongjiajie]

It's weird, it pass in my local breeze env, but failed in Travis both two of my PR. #7593

[turbaszek]

Yeah, sometimes the tests timeouts. I think @mik-laj was investigating it and we think is related to limited resources on CI (memory swap for example).

[zhongjiajie]

@nuclearpinguin Thanks for the clarification, so did it Travis side or ours?

[potiuk]

It looks like Travis quirks - it looks like the same tests start to fail more often and then after some time they continue to work more reliably. It' super-difficult trace (and know whose fault it is) and figure out what's going on. This is not a simple "We" vs. Travis.

We already fixed (or workarounded) a number of problems with Travis and we have an outstanding task to migrate to Github Actions but we have other high priority tasks.

@zhongjiajie - Maybe you could help with investigating it or migrating to Github Actions (where we believe things will be more stable)?

[zhongjiajie]

@potiuk Like to help, but I don't have much experience in CI/CD. I could see we already have some discuss in dev-list, and does the related AIP is? https://cwiki.apache.org/confluence/display/AIRFLOW/AIP-23+Migrate+out+of+Travis+CI

[potiuk]

Yep. That's the related PR indeed :)

[zhongjiajie]

@potiuk @nuclearpinguin PTAL, thx.

[ashb]

Since the original change/PR hasn't been included in a release yet, this should be targeted against the same JIRA

[zhongjiajie]

Since the original change/PR hasn't been included in a release yet, this should be targeted against the same JIRA

Got it

[zhongjiajie]

I use the same JIRA ticket, change PR title and git commit message, and also close this PR original ticket https://issues.apache.org/jira/browse/AIRFLOW-6976

[ashb]

@zhongjiajie The reason for targetting this against AIRFLOW-6976 and not a new jira is around release-automation: if we have separate Jiras for a "related" fix then it is possible to miss this change if the first PR was backported. By having two commits target a single jira issue our existing release scripts (./dev/airflow-jira) will show both commits need to be pulled in.

[zhongjiajie]

@zhongjiajie The reason for targetting this against AIRFLOW-6976 and not a new jira is around release-automation: if we have separate Jiras for a "related" fix then it is possible to miss this change if the first PR was backported. By having two commits target a single jira issue our existing release scripts (./dev/airflow-jira) will show both commits need to be pulled in.

I haven't noticed that before, thanks Ash for the clarification

[zhongjiajie]

Travis failed due to job name mistake and fix in #7668, I rebase on master on restart the failing test.

[codecov-io]

Codecov Report

Merging #7610 into master will decrease coverage by 0.27%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master    #7610      +/-   ##
==========================================
- Coverage   86.95%   86.67%   -0.28%     
==========================================
  Files         900      900              
  Lines       43605    43605              
==========================================
- Hits        37915    37796     -119     
- Misses       5690     5809     +119

Impacted Files	Coverage Δ
airflow/cli/commands/dag_command.py	85.18% <100%> (ø)	⬆️
airflow/kubernetes/volume_mount.py	44.44% <0%> (-55.56%)	⬇️
airflow/kubernetes/volume.py	52.94% <0%> (-47.06%)	⬇️
airflow/kubernetes/pod_launcher.py	47.18% <0%> (-45.08%)	⬇️
...viders/cncf/kubernetes/operators/kubernetes_pod.py	69.69% <0%> (-25.26%)	⬇️
airflow/kubernetes/refresh_config.py	50.98% <0%> (-23.53%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c93f82b...42da161. Read the comment docs.

[zhongjiajie]

Travis happy, could you please take a look? @ashb @potiuk

[zhongjiajie]

PTAL @ashb @kaxil @potiuk @mik-laj @turbaszek @feluelle

[zhongjiajie]

Fix UPDATING.md

[zhongjiajie]

Thanks everyone