cat / airbyte-ci: improve CAT container orchestration

[alafanechere]

What

Closes #31703

This PR aims at improving how we orchestrate CAT through airbyte-ci.
We want to:

• Reduce test cold start duration
• Optionally enable concurrent CAT tests

How

Reducing test cold start duration

• Make the connector_container a session scoped fixture. It means the tests will reuse the same connector container.
• Load the connector container with Dagger container id instead of importing the container image as a tar. Avoiding the use of tar allows us to remove the long and costly docker export and dagger import steps.

Enabling test concurrency

Pytest has pytest-xdist plugin which can distribute tests across multiple CPUs.
With the --numprocesses=auto flag in the pytest command this plugin will distribute tests across all the available cores.
I don't want to enable this by default on Python connectors as I'm concerned it can have a bad impact on the API connectors' rate limit. So I:

• Expose a new --concurrent-cat flag on airbyte-ci connectors test command. Defaulting to false.
• Force concurrency to true on Java connectors. Java sources are connecting to databases that are not affected by rate limits.
• We can eventually enable CAT concurrency by default when we want to make Python connector CAT faster. But a thorough investigation of its effect must be done before.

Performance boost on source-postgres (with concurrency).

This change makes source-postgres CAT execution run in 01mn56s. Which is a ~2mn boost compared to the current 03mn54 execution on master.

Perfomance boost on python connectors (no concurrency)

I used source-google-sheets for testing.
As I said, test concurrency is disabled by default for Python connectors.
We still get a ~1mn CAT boost due to a lower test cold start.
On master: 3mn11s
On this branch: 2mn07

Recommended reading order

1. Changes to CAT: airbyte-integrations/bases/connector-acceptance-test/*
2. Changes to airbyte-ci: airbyte-ci/connectors/pipelines/*

🚨 User Impact 🚨

• Significant CAT performance boost on Java connectors test thanks to concurrency. Tested for source-postgres
• 1mn CAT performance boost on Python connectors thanks to a lower test start-up time. Tested for source-google-sheets

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

1 Ignored Deployment

Name	Status	Preview	Comments	Updated (UTC)
airbyte-docs	⬜️ Ignored (Inspect)	Visit Preview		Oct 24, 2023 8:02am

[alafanechere]

Current dependencies on/for this PR:

• master
  • PR cat / airbyte-ci: improve CAT container orchestration #31699 👈

This comment was auto-generated by Graphite.

[github-actions]

Before Merging a Connector Pull Request

Wow! What a great pull request you have here! 🎉

To merge this PR, ensure the following has been done/considered for each connector added or updated:

• PR name follows PR naming conventions
• Breaking changes are considered. If a Breaking Change is being introduced, ensure an Airbyte engineer has created a Breaking Change Plan.
• Connector version has been incremented in the Dockerfile and metadata.yaml according to our Semantic Versioning for Connectors guidelines
• You've updated the connector's metadata.yaml file any other relevant changes, including a breakingChanges entry for major version bumps. See metadata.yaml docs
• Secrets in the connector's spec are annotated with airbyte_secret
• All documentation files are up to date. (README.md, bootstrap.md, docs.md, etc...)
• Changelog updated in docs/integrations/<source or destination>/<name>.md with an entry for the new version. See changelog example
• Migration guide updated in docs/integrations/<source or destination>/<name>-migrations.md with an entry for the new version, if the version is a breaking change. See migration guide example
• If set, you've ensured the icon is present in the platform-internal repo. (Docs)

If the checklist is complete, but the CI check is failing,

1. Check for hidden checklists in your PR description

2. Toggle the github label checklist-action-run on/off to re-run the checklist CI.

[airbyte-oss-build-runner]

source-google-sheets test report (commit 7e46301bb4) - ✅

⏲️ Total pipeline duration: 01mn53s

Step	Result
Build source-google-sheets docker image for platform(s) linux/amd64	✅
Unit tests	✅
Acceptance tests	✅
Check our base image is used	✅
Code format checks	✅
Validate metadata for source-google-sheets	✅
Connector version semver check	✅
QA checks	✅

🔗 View the logs here

☁️ View runs for commit in Dagger Cloud

Please note that tests are only run on PR ready for review. Please set your PR to draft mode to not flood the CI engine and upstream service on following commits.
You can run the same pipeline locally on this branch with the airbyte-ci tool with the following command

airbyte-ci connectors --name=source-google-sheets test

[airbyte-oss-build-runner]

source-postgres test report (commit bbb4d89b41) - ✅

⏲️ Total pipeline duration: 19mn36s

Step	Result
Build connector tar	✅
Java Connector Unit Tests	✅
Build source-postgres docker image for platform(s) linux/amd64	✅
Acceptance tests	✅
Java Connector Integration Tests	✅
Validate metadata for source-postgres	✅
Connector version semver check	✅
QA checks	✅

🔗 View the logs here

☁️ View runs for commit in Dagger Cloud

Please note that tests are only run on PR ready for review. Please set your PR to draft mode to not flood the CI engine and upstream service on following commits.
You can run the same pipeline locally on this branch with the airbyte-ci tool with the following command

airbyte-ci connectors --name=source-postgres test

[postamar]

LGTM for the parts that I do understand. It makes sense to me to leverage dagger in this way.

[bazarnov]

I still don't understand why we are not getting performance boost for python-based sources?

Please share, if there are more details around this. LGTM!

[alafanechere]

I still don't understand why we are not getting performance boost for python-based sources? Please share, if there are more details around this. LGTM!

@bazarnov As Python sources make API requests that can be rate-limited I'm afraid that running the CAT tests in parallel can easily burst this rate limit. Feel free to try out on python connectors locally by running airbyte-ci connectors --name=<your-connector> tests --concurrent-cat and let me know if you perceive a performance boost or regression 😄

[sentry-io]

Suspect Issues

This pull request was deployed and Sentry observed the following issues:

• ‼️ ImportError: cannot import name 'PipelineContext' from 'pipelines.models.contexts' (/home/runner/.local/pipx/v... pipelines.airbyte_ci.connectors.test.steps.comm... View Issue

_{Did you find this useful? React with a 👍 or 👎}