Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Setup docs on current pipelines #1058

Merged
merged 12 commits into from
Dec 18, 2023
215 changes: 215 additions & 0 deletions docs/current_pipelines.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,215 @@
# Drools 8 / Kogito / OptaPlanner 8 pipelines

Here is a small overview of current pipelines existing around Kogito project.

## Which projects ?

- Drools
- https://github.com/apache/incubator-kie-drools
- https://github.com/kiegroup/drools-website
- https://github.com/apache/incubator-kie-benchmarks
- https://github.com/kiegroup/kie-pmml-integration
- Kogito
- https://github.com/apache/incubator-kie-kogito-pipelines
- https://github.com/apache/incubator-kie-kogito-runtimes
- https://github.com/apache/incubator-kie-kogito-apps
- https://github.com/apache/incubator-kie-kogito-examples
- https://github.com/apache/incubator-kie-kogito-images
- https://github.com/apache/incubator-kie-kogito-operator
- https://github.com/apache/incubator-kie-kogito-docs
- https://github.com/apache/incubator-kie-kogito-serverless-operator
- https://github.com/apache/incubator-kie-docs
- https://github.com/kiegroup/kie-tools
- OptaPlanner
- https://github.com/apache/incubator-kie-optaplanner
- https://github.com/apache/incubator-kie-optaplanner-quickstarts
- https://github.com/kiegroup/optaplanner-website

### Projects' dependencies

- Kogito snapshot depends on Drools snapshot
They remain extremely coupled and some pipelines will need first to run Drools before Kogito, like the `Prepare release branch` and `Release` pipelines
- OptaPlanner depends a Final version of Drools.

All 3 projects can still be released

### Kogito specificities

Kogito contains both Artifacts and Cloud repositories. Those are "decoupled", which means there are:

- 2 main release jobs, one for Artifacts and one for Cloud (but artifacts can call cloud)
- 2 main default nightly jobs, one for Artifacts and one for Cloud (but artifacts can call cloud)
- 2 main setup branch jobs, one for Artifacts and one for Cloud (but artifacts can call cloud)

## Jenkins

radtriste marked this conversation as resolved.
Show resolved Hide resolved
All current Jenkins pipelines can be found in https://ci-builds.apache.org/job/KIE

Each project has there its own folder (`drools`, `kogito`, `optaplanner`) and in there its own main seed job (see after).
### Seed generation

For more information on the seed generation, please look at the [jenkins](./jenkins.md) documentation.

Main information to remember:

- One main seed job will start the whole process and generate all jobs, for all defined branches.
In that process, it will create global root jobs as well as one folder per branch (defined into the `main.yaml` config file).
In each branch, it is created a branch seed job.
Logic of the seed job can be found in [here](../dsl/seed/jenkinsfiles/Jenkinsfile.seed.main)
Definition of the main seed job can found [here](../dsl/seed/jobs/seed_job_main.groovy)
- Each branch seed job will generate the jobs, based on the `.ci/jenkins/dsl/jobs.groovy` file, for each of the defined repository/branch.
For each branch, the information needed is defined into the `branch.yaml` file, being environments or repositories or credentials ...
Logic of the seed job can be found in [here](../dsl/seed/jenkinsfiles/Jenkinsfile.seed.branch)
Definition of a branch seed job can found [here](../dsl/seed/jobs/seed_job_branch.groovy)
- When branching (`prepare-release-branch` job), it should process 3 things:
- Create release branches for all concerned repositories
- Create config branch for the config files repository
- Create seed branch for the seed library (`kogito-pipelines` repository)

### Jenkins jobs

### Job environments

Environments can be found in the branch seed configuration.

Usually you find those:

- default
or the "empty" environment
- native
Execute build with native maven commands
- quarkus-branch
Check the project against the latest updates from the current quarkus branch used
- quarkus-main
Check the project against the latest updates from quarkus main branch (forward compatiblity)
- quarkus-3
Migrate repositories to Quarkus 3
- quarkus-lts
Check the project against the latest updates from quarkus LTS branch (backward compatiblity)
- native-lts
Check the project against the latest updates from quarkus LTS branch (backward compatiblity) in native mode

#### Nightlies

**Main nightly**

or default nightly.

This one building, testing and deploying artifacts/images.

**Environment nightlies**

Those jobs are only for artifacts' repositories.

Environment jobs are defined in each concerned repositories. They use the environment definition and make a simple build.
Those jobs are making use of the build-chain via the [build-chain jenkinsfile](../dsl/seed/jenkinsfiles/Jenkinsfile.buildchain).

The run of those builds use the `.ci/environments` folder of each repository.
This allows to have some specific updates for the build.
Updates are usually in a patch folder. But some `before.sh` and `after.sh` scripts can allow more changes.
For example, Quarkus 3 flows uses OpenRewrite + patches to do the migration, as the changes are quite big.

Integration jobs are creating specific branches before perfoming the build. Branch name pattern is `{BASE_BRANCH}-integration-{ENVIRONMENT_NAME}`.

#### PRs

PR jobs are making use of the build-chain via the [build-chain jenkinsfile](../dsl/seed/jenkinsfiles/Jenkinsfile.buildchain).
The environment variables are determining what and how it should be launched for each environments.

Like the nightlies, the PR check use the environments scripts if needed and existing.

#### Setup branch

Those jobs are used:

- when a release branch is created to update the release branch version and the main branch version
- after a release is done, so we can update the release branch to the next snapshot

Like the default nightly, there is a main pipeline which is calling the other jobs from the folder.

#### Release

Contains all release jobs.

Like the default nightly and setup branch processes, there is a main release job which is calling the different deploy/promote jobs.

For more information, you can also have a look to the [release page](./release%20pipeline.md).

There is also a [release checklist](./release%20checklist.md).

#### Tools

Useful tools for the pipelines, like updating an external dependency version, cleaning images ...

### Quarkus Platform nightly

As we need to deploy a specific Quarkus Platform with our snapshot artifacts, there is a nightly existing, which is launched by the main nightly.

### Quarkus 3 jobs

The Quarkus 3 flow is a bit specific as we are preparing for releasing versions in parallel of Quarkus 2.x version.
We end up with 2 streams:

- Drools8/Kogito/OptaPlanner8
- Drools9/Kogito2/OptaPlanner9

#### Quarkus 3 nightly integration jobs

This is done in the `quarkus-3` environment nightly job (that can be found in `nightly.quarkus-3` folder).

This environment has always a `0001_before_sh.patch` which is the result of the `before.sh` script, containing OpenRewrite automation. All other patches are manually maintained.

When the nightly job is in failure, it usually means that the patches could not be correctly applied. Check the logs.
And `UNSTABLE` build, means only test failures.

The nightlies are creating specific branches:

- Drools 9.x
- Kogito 2.x
- OptaPlanner 9.x

**NOTE: The Quarkus 3 integration jobs are not running any testing. They are only performing the code migration. Tests are performed in the nightly jobs of the created branches**

#### Quarkus 3 branches

**DSL config files**

Main Config file for Quarkus 3 are can be found on separate from code:
- https://github.com/apache/incubator-kie-drools/tree/9.x-dsl-config
- https://github.com/apache/incubator-kie-kogito-pipelines/tree/2.x-dsl-config
- https://github.com/apache/incubator-kie-optaplanner/tree/9.x-dsl-config

And jobs would be found in different Jenkins folders:
- https://ci-builds.apache.org/job/KIE/job/drools-9.x/
- https://ci-builds.apache.org/job/KIE/job/kogito-2.x/
- https://ci-builds.apache.org/job/KIE/job/optaplanner-9.x/

NOTE: We keep separate configurations between the 2 streams because it is easier to manage. Jobs configuration can then be read. Compare to "Stream 8/1", some environments are disabled for now.

## GitHub Actions

Historically, GHA was used as a backup of Jenkins, at the time when Jenkins was very unstable.
Since then, we tried to keep both sync.
Also GHA is used to provide external users (outside Red Hat) some feedbacks in case anything fails.

Unlike Jenkins, there is no DSL for GHA but it means job run is much more dynamic. Any change can be applied automatically.

And we only have PR checks with GHA.

### PR checks

See the different workflows. They mostly use the build-chain for artifacts's repositories. For cloud, it is to run simple and quick jobs.

## How-tos ?

### Update Quarkus version

1) Create an issue on [kie-issues](https://github.com/apache/incubator-kie-issues)
2) Launch `main/tools/update-quarkus-all` job with
* [Drools](https://ci-builds.apache.org/job/KIE/job/drools/job/main/job/tools/job/update-quarkus-all/)
* [Kogito](https://ci-builds.apache.org/job/KIE/job/kogito/job/main/job/tools/job/update-quarkus-all/)
* [OptaPlanner](https://ci-builds.apache.org/job/KIE/job/optaplanner/job/main/job/tools/job/update-quarkus-all/)
3) Wait for the PRs to arrive on [kogito-ci](https://kie.zulipchat.com/#narrow/stream/236603-kogito-ci) and [optaplanner-ci](https://kie.zulipchat.com/#narrow/stream/354864-optaplanner-ci) streams
4) Review the changes.
**Make sure there is no downgrade of dependency version !!!!**

25 changes: 0 additions & 25 deletions docs/nightly pipeline.md
Original file line number Diff line number Diff line change
Expand Up @@ -9,27 +9,11 @@
- [Nightly pipeline is failing](#nightly-pipeline-is-failing)
- [Nightly pipeline is unstable](#nightly-pipeline-is-unstable)
- [Build \& Deploy job is failing](#build--deploy-job-is-failing)
- [Promote job is failing](#promote-job-is-failing)
- [Testing the Nightly Pipeline](#testing-the-nightly-pipeline)

In order to perform, Nightly and Release pipelines need to call some deploy and promote jobs for runtimes, examples, images and operator.
Those jobs should be present at the same level as the nightly and/or release job, so they can be found when called.

Here is the list of jobs and link to Jenkinsfiles:

* [kogito-runtimes-deploy](https://github.com/apache/incubator-kie-kogito-runtimes/blob/main/Jenkinsfile.deploy)
* [kogito-runtimes-promote](https://github.com/apache/incubator-kie-kogito-runtimes/blob/main/Jenkinsfile.promote)
* [kogito-apps-deploy](https://github.com/apache/incubator-kie-kogito-apps/blob/main/Jenkinsfile.deploy)
* [kogito-apps-promote](https://github.com/apache/incubator-kie-kogito-apps/blob/main/Jenkinsfile.promote)
* [kogito-examples-deploy](https://github.com/apache/incubator-kie-kogito-examples/blob/main/Jenkinsfile.deploy)
* [kogito-examples-promote](https://github.com/apache/incubator-kie-kogito-examples/blob/main/Jenkinsfile.promote)
* [kogito-images-deploy](https://github.com/apache/incubator-kie-kogito-images/blob/main/Jenkinsfile.deploy)
* [kogito-images-promote](https://github.com/apache/incubator-kie-kogito-images/blob/main/Jenkinsfile.promote)
* [kogito-examples-images-deploy](https://github.com/apache/incubator-kie-kogito-operator/blob/main/Jenkinsfile.examples-images.deploy)
* [kogito-examples-images-promote](https://github.com/apache/incubator-kie-kogito-operator/blob/main/Jenkinsfile.examples-images.promote)
* [kogito-operator-deploy](https://github.com/apache/incubator-kie-kogito-operator/blob/main/Jenkinsfile.deploy)
* [kogito-operator-promote](https://github.com/apache/incubator-kie-kogito-operator/blob/main/Jenkinsfile.promote)

## Nightly pipeline Architecture

The Nightly Pipeline is composed of many steps, calling different other jobs to perform the build&test of runtimes/examples/images/operator as well as the deployment of jar artifacts and nightly container images.
Expand Down Expand Up @@ -96,15 +80,6 @@ Here are some problems which can occur on a `Build & Deploy` job:
* Test errors
* Deployment problem (credentials)

### Promote job is failing

In that case, identify the error (Groovy script error or test problem), correct it and rebuild the failing job.

Here are some problems which can occur on a `Promote` job:

* Groovy script error (Jenkinsfile.deploy has some problems)
* Deployment problem (credentials)

## Testing the Nightly Pipeline

In order to test the full Nightly Pipeline, and in order to avoid any problem, you will need to change some env in [Jenkinsfile.nightly](../Jenkinsfile.nightly), create jobs in Jenkins and setup some credentials.
Expand Down
6 changes: 1 addition & 5 deletions docs/release checklist.md
Original file line number Diff line number Diff line change
Expand Up @@ -2,11 +2,7 @@

## Cut-off

- Prepare release branch
- Retrieve PR created from the job
- Once PR is merged, execute `0-seed-job` (automation not yet fully available ...)

Jobs should then be created for the branch and nightly are automatically activated.
Call the `0-prepare-release-branch` and wait for all the preparation.

## Before Release day

Expand Down