charter for CI testing subgroup of SIG Node

This commit is contained in:
Sergey Kanzhelev 2025-03-30 02:23:42 +00:00
parent 0c9e91d69e
commit fb36548972
3 changed files with 105 additions and 0 deletions

View File

@ -212,8 +212,40 @@ Running a local cluster
Note: Task 5 requires Linux OS
## Subgroups
### CI testing
CI testing subgroup of SIG Node provides a great way to get started with the SIG.
CI testing concentrate on reliability and ongoing issues, see the [charter](sig-node/sig-node-ci-testing-group-charter.md) for details.
There are many ways to participate:
1. Bring a topic for the discussion at a weekly (see "How to..." section below).
Topic must be aligned with the group charter.
2. Participate in test issues, PRs, and bugs triage during the meeting.
Expertise in various areas is needed when triaging.
So YOUR expertise may be very valuable for some of the topics discussed.
3. Host a section of a meeting. Once you attended a meeting a few times,
you may be ready to host the section of a meeting. It will include
presenting your screen and applying needed labels to the issues.
## How to ...
### add agenda item to the meeting
This applies to both, SIG Node weekly meeting and CI testing subgroup weekly meeting.
1. Find the meeting invite and notes document.
2. Join the group sig-node@kubernetes.io for edit permissions.
3. Add you topic to the end of the list, prefixed with your alias, to the agenda of the next meeting.
4. Show up at the meeting to present your agenda item or find a delegate to present it.
There is generally no restrictions on topics, and topics are added on first come, first served basis.
Chair who host the meeting will try to allocate time and fit as many topics as possible.
However declared order is not guaranteed. Also topics may be moved to the next meeting
if other things like KEP planning must be prioritized.
### upload meeting recording to YouTube (for chairs)
1. Download the video file from zoom account `sig-node-leads@kubernetes.io`

View File

@ -150,4 +150,9 @@ The following topics fall under scope of this SIG.
* Host OS and/or kernel interactions (to a limited extent)
We also work closely with [sig-storage](../sig-storage) and [sig-network](../sig-network). As you can see, this is a very cross-functional team!
## Subgroups
Read the CI testing subgroup [charter](sig-node-ci-testing-group-charter.md).
<!-- END CUSTOM CONTENT -->

View File

@ -0,0 +1,68 @@
# SIG Node CI testing subgroup charters
- **Created:** Jul 24, 2020
- **Last Updated:** Mar 27, 2025
## Overview
SIG Node CI Group is proposed and formed by volunteers from the [SIG Node](charter.md) community.
SIG Node owns a broad range of [code, binaries, and services](charter.md#code-binaries-and-services)
which are fundamental to the success of Kubernetes.
To ensure high quality of these components, SIG Node community has developed a large number of
[end-to-end (e2e) test cases](https://testgrid.k8s.io/sig-node) over the past several years.
With new PR requests merged every day, OS distros being updated,
it has become more and more challenging to maintain these tests and to keep them up to date with new changes.
In order to maintain tests and keep up with the incoming bugs triage, the CI subproject was formed.
The progress of this group can be viewed in the [meeting notes](https://docs.google.com/document/d/1fb-ugvgdSVIkkuJ388_nhp2pBTy_4HEVg5848Xy7n5U/edit#heading=h.badwgoqn6j9e).
## Goal
The goal of this group is to keep a high quality bar for SIG Node and to enable continuous integration of node features of Kubernetes.
More specifically, this group drives all aspects of [SIG Node e2e tests](https://testgrid.k8s.io/sig-node),
and aims to achieve the following goals:
- Maintaining existing node e2e tests healthiness
- Triage and fix failing tests (especially release-blocking tests) in a timely manner
- Remove deprecated tests
- Enhanced test failure troubleshooting and diagnostics tooling and documentation
- Improving test code quality
- Identify and reduce flaky behavior in tests
- Review new e2e test code
- Improving test coverage
- Identify areas missing test coverage, and work with owners to add e2e tests
- OS Image and dependencies support
- Make sure the representative set of OS images are being tested
- Make sure the representative set of container runtime versions are being tested
- Test resource utilization and cost optimization
- Increase ROI of SIG Node tests
- Help spread tests for different clouds for better usage of cloud credits
- Make sure timely response for critical bugs
- Triage bugs regularly
- Identify critical issue and ensure timely follow up
- Review high priority bugs before each release
## Execution Plan
- Regular meet-ups for discussion and triage
- Triage test failures
- Review test related Pull requests and issues backlog
- Triage bugs
- Periodically summarize test coverage (flakes, failures etc) in sig-node meeting for broader visibility
- Image Support
- Maintain owners for supported OS images.
- Process/documentation on lifecycle of OS images used by SIG Node tests.
- Ensure a process and/or automation exists for OS image upgrading process.
- Group Sustainability
- Identify, document release-blocking tests
- Grow node e2e test reviewers and approvers
## History
- Original [charter document](https://docs.google.com/document/d/1yS-XoUl6GjZdjrwxInEZVHhxxLXlTIX2CeWOARmD8tY/)