15 Commits

Author SHA1 Message Date
Timofey Kirillov
7aaf2cce6b Merge pull request #57 from werf/upd-readme
[1.1] docs: add notice about fetch-depth into README
2022-10-21 12:35:05 +03:00
Timofey Kirillov
95ff181e6a docs: add notice about fetch-depth into README
Signed-off-by: Timofey Kirillov <timofey.kirillov@flant.com>
2022-10-21 12:33:17 +03:00
Alexey Igrychev
c5e105e95e chore(readme): fix links
Signed-off-by: Alexey Igrychev <alexey.igrychev@flant.com>
2022-08-15 08:06:09 +01:00
Alexey Igrychev
8047e055e2 chore(readme): fix failed to load werf logo 2021-12-07 16:18:17 +01:00
Alexey Igrychev
7425cd9c3c Update README.md 2021-11-10 14:28:18 +01:00
Alexey Igrychev
4033ce9a92 Merge pull request #45 from werf/feat_use_trdl_binaries_1.1
feat: download binaries directly from the trdl repository (1.1)
2021-10-28 12:24:26 +01:00
Alexey Igrychev
6acc213146 feat: download binaries directly from the trdl repository 2021-10-27 22:25:32 +01:00
Alexey Igrychev
f507c4f344 ci(test): add Test workflow 2021-10-27 22:24:02 +01:00
Alexey Igrychev
93f4252af6 build: update dependencies and lint configuration 2021-10-27 22:21:14 +01:00
Alexey Igrychev
bd74d436c7 Output envs loaded by the dotenv instead of printing a source file content 2021-03-18 17:15:21 +00:00
Alexey Igrychev
4e62aa309e [readme] Update links to werf documentation 2021-03-10 16:20:24 +00:00
Alexey Igrychev
697f06e4a1 [readme] Add "Versioning" section 2021-03-10 15:46:06 +00:00
Alexey Igrychev
47e7c4a185 [readme] Fix the "Working with container registry" section for v1.1 2021-03-10 15:31:41 +00:00
Alexey Igrychev
7c8622c3d6 [readme] Update channels, werf release cycle and compatibility promise link 2021-03-10 15:01:01 +00:00
Alexey Igrychev
156c73b693 [readme] Add "Working with container registry" section 2021-03-10 14:59:08 +00:00
40 changed files with 159279 additions and 82722 deletions

View File

@@ -12,7 +12,7 @@ jobs:
runs-on: ${{ matrix.os }}
steps:
- name: Checkout code
uses: actions/checkout@v4
uses: actions/checkout@v2
with:
fetch-depth: 0
@@ -31,13 +31,13 @@ jobs:
steps:
- name: Checkout code
uses: actions/checkout@v4
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Install
uses: ./install
with:
version: v2.31.1
version: v1.1.23+fix50
- run: werf version
- run: werf version

285
README.md
View File

@@ -3,47 +3,46 @@
</p>
___
This action allows you to organize CI/CD with GitHub Actions and [werf](https://github.com/werf/werf).
This action set allows you to organize CI/CD with GitHub Actions and [werf](https://github.com/werf/werf). The set consists of several independent and complex actions:
**Ready-to-use GitHub Actions Workflows** for different CI/CD workflows are available [here](https://werf.io/guides/nodejs/400_ci_cd_workflow/040_github_actions.html).
- [werf/actions/converge](https://github.com/werf/actions/tree/v1.1/converge)
- [werf/actions/build-and-publish](https://github.com/werf/actions/tree/v1.1/build-and-publish)
- [werf/actions/build](https://github.com/werf/actions/tree/v1.1/build)
- [werf/actions/publish](https://github.com/werf/actions/tree/v1.1/build)
- [werf/actions/deploy](https://github.com/werf/actions/tree/v1.1/deploy)
- [werf/actions/dismiss](https://github.com/werf/actions/tree/v1.1/dismiss)
- [werf/actions/run](https://github.com/werf/actions/tree/v1.1/run)
- [werf/actions/cleanup](https://github.com/werf/actions/tree/v1.1/cleanup)
## How to use
Each action combines all the necessary steps in itself and logic may be divided into environment setup and launching the corresponding command.
```yaml
converge:
name: Converge
runs-on: ubuntu-latest
steps:
**Ready-to-use GitHub Actions Workflows** for different CI/CD workflows are available [here](https://werf.io/v1.1-alpha/documentation/guides/github_ci_cd_integration.html#complete-set-of-configurations-for-ready-made-workflows).
- name: Checkout code
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Install werf
uses: werf/actions/install@v2
- name: Run script
run: |
. $(werf ci-env github --as-file)
werf converge
env:
WERF_KUBECONFIG_BASE64: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
WERF_ENV: production
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
```
> Also, there is another action — [werf/actions/install](https://github.com/werf/actions/tree/v1.1/install). With this action a user can just install werf and use binary within job steps for own purposes
## Versioning
When using action, select the version corresponding to the required `MAJOR` version of werf.
When using actions, select the version corresponding to the `MAJOR.MINOR` version of werf:
By default, the action installs actual werf version within alpha channel (more details about channels, werf release cycle and compatibility promise [here](https://werf.io/installation.html#all-changes-in-werf-go-through-all-stability-channels)).
```yaml
# run converge with actual werf version for 1.1 alpha channel
- uses: werf/actions/converge@v1.1
# run converge with actual werf version for 1.2 alpha channel
- uses: werf/actions/converge@v1.2
```
## Environment setup in details
### werf binary setup
By default, all actions setup actual werf version for 1.1 alpha channel (more details about channels, werf release cycle and compatibility promise [here](https://werf.io/installation.html#all-changes-in-werf-go-through-all-stability-channels)).
Using the `channel` input the user can switch the release channel.
> This is recommended approach to be up-to-date and to use actual werf version without changing configurations.
> This is recommended approach to be up-to-date and to use actual werf version without changing configurations
```yaml
- uses: werf/actions/install@v2
- uses: werf/actions/converge@v1.1
with:
channel: alpha
```
@@ -51,31 +50,63 @@ Using the `channel` input the user can switch the release channel.
Withal, it is not necessary to work within release channels, and the user might specify certain werf version with `version` input.
```yaml
- uses: werf/actions/install@v2
- uses: werf/actions/converge@v1.1
with:
version: v2.1.0
version: v1.1.23
```
## FAQ
### kubeconfig setup (*optional*)
### werf always rebuilds images on new commit
The _kubeconfig_ may be used for deployment, cleanup, distributed locks and caches. Thus, the configuration should be added before step with the action or passed as base64 encoded data with `kube-config-base64-data` input:
Make sure to use `fetch-depth: 0` setting in the checkout action, like follows:
* Prepare _kubeconfig_ (e.g. `cat ~/.kube/config | base64`) and save in GitHub Project Secrets (e.g. with name `KUBE_CONFIG_BASE64_DATA`).
* Pass secret with `kube-config-base64-data` input:
```yaml
- uses: werf/actions/build-and-publish@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
### werf ci-env
This command performs _docker login_ using `github-token`, sets up predefined variables based on GitHub Workflow context.
**Note** that `github-token` is optional in this action, and the input is there in case you need to use a non-default token.
By default, action will use the token provided to your workflow.
## Working with werf options
Any werf option can be defined with environment variables:
```yaml
- name: Checkout code
uses: actions/checkout@v4
with:
fetch-depth: 0
- uses: werf/actions/build-and-publish@v1.1
env:
WERF_LOG_VERBOSE: "on"
WERF_TAG_CUSTOM_TAG1: tag1
WERF_TAG_CUSTOM_TAG2: tag2
```
By default, fetch-depth set to `1` which disables git history when checking out code. werf cache selection algorithm uses git history to determine whether some image bound to some commit could be used as a cache when building current commit (current commit should be descendant to the cache commit).
## Working with container registry
Setting `fetch-depth` to `0` enables full fetch of git history, and it is a **recommended** approach. It is also possible to limit fetch history with some decent number of commits, which would enable images caching limited to that number of commits, but this would have a negative impact on cache reproducibility.
Due to the fact that the new GitHub container registry (`ghcr.io`) does not currently support removal, all actions default to the old one (`docker.pkg.github.com`).
### Working with container registry
If necessary, the user can define an arbitrary container registry using the `WERF_REPO` and `WERF_REPO_IMPLEMENTATION` environment variables.
If there is a need to perform authorization using custom credentials or in an external container registry, then you have to use a ready-made action tailored to your container registry (or just run `werf cr login`).
```yaml
- uses: werf/actions/converge@v1.1
env:
WERF_REPO: "gcr.io/company/app"
WERF_REPO_IMPLEMENTATION: "gcr"
```
To learn how to work with the different container registries, see the corresponding [article in the werf documentation](https://werf.io/v1.1-alpha/documentation/reference/working_with_docker_registries.html).
## Examples
### converge
```yaml
converge:
@@ -84,28 +115,164 @@ converge:
steps:
- name: Checkout code
uses: actions/checkout@v4
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Install werf
uses: werf/actions/install@v2
- name: cr login
run: werf cr login -u ${{ secrets.REGISTRY_USER }} -p ${{ secrets.REGISTRY_TOKEN }} registry.example.com
- name: converge
run: werf converge
env:
WERF_KUBECONFIG_BASE64: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
WERF_ENV: production
WERF_REPO: registry.example.com/repo
- name: Converge
uses: werf/actions/converge@v1.1
with:
env: production
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
> Environment variables **`WERF_REPO`** and **`GITHUB_TOKEN`** for converge should only be used if building images is required otherwise they can be omitted
### build, publish and deploy
In the simplest case, if an [integrated GitHub Packages-like container registry](https://help.github.com/en/packages/using-github-packages-with-your-projects-ecosystem/configuring-docker-for-use-with-github-packages) is used, then the authorization is performed automatically when the `werf ci-env` command is invoked. This command is run with several required arguments such as GitHub environment variables, the [`GITHUB_TOKEN` secret](https://help.github.com/en/actions/configuring-and-managing-workflows/authenticating-with-the-github_token#about-the-github_token-secret) (you have to explicitly declare it).
```yaml
build-and-publish:
name: Build and Publish
runs-on: ubuntu-latest
steps:
## License
- name: Checkout code
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Build and Publish
uses: werf/actions/build-and-publish@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
deploy:
name: Deploy
needs: build-and-publish
runs-on: ubuntu-latest
steps:
- name: Checkout code
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Deploy
uses: werf/actions/deploy@v1.1
with:
env: production
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
### dismiss
```yaml
dismiss:
name: Dismiss
runs-on: ubuntu-latest
steps:
- name: Checkout code
uses: actions/checkout@v2
- name: Dismiss
uses: werf/actions/dismiss@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
env: production
```
### run
```yaml
run:
name: Run
runs-on: ubuntu-latest
steps:
- name: Checkout code
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Run
uses: werf/actions/run@v1.1
with:
image: backend
args: rails server
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
env:
WERF_DOCKER_OPTIONS: "-d -p 3000:3000"
```
### cleanup
```yaml
cleanup:
name: Cleanup
runs-on: ubuntu-latest
steps:
- name: Checkout code
uses: actions/checkout@v2
- name: Fetch all history for all tags and branches
run: git fetch --prune --unshallow
- name: Cleanup
uses: werf/actions/cleanup@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
### install
```yaml
werf:
name: werf
runs-on: ubuntu-latest
steps:
- name: Checkout code
uses: actions/checkout@v2
- name: Install werf CLI
uses: werf/actions/install@v1.1
# for deploy and distributed locks
- name: Create kube config
run: |
KUBECONFIG=$(mktemp -d)/config
base64 -d <(printf "%s" $KUBE_CONFIG_BASE64_DATA) > $KUBECONFIG
echo KUBECONFIG=$KUBECONFIG >> $GITHUB_ENV
env:
KUBE_CONFIG_BASE64_DATA: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
- name: Run werf commands
run: |
source $(werf ci-env github --as-file)
werf build-and-publish
werf deploy
env:
GITHUB_TOKEN: ${{ github.token }}
WERF_ENV: production
```
# FAQ
## werf always rebuilds images on new commit
Make sure to use `fetch-depth: 0` setting in the checkout action, like follows:
```
- name: Checkout code
uses: actions/checkout@v2
with:
fetch-depth: 0
```
By default fetch-depth set to `1` which disables git history when checking out code. werf cache selection algorithm uses git history to determine whether some image bound to some commit could be used as a cache when building current commit (current commit should be descendant to the cache commit).
Setting `fetch-depth` to `0` enables full fetch of git history and it is a **recommended** approach. It is also possible to limit fetch history with some decent number of commits, which would enable images caching limited to that number of commits, but this would have a negative impact on cache reproducibility.
# License
Apache License 2.0, see [LICENSE](LICENSE)

View File

@@ -23,5 +23,5 @@ inputs:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
runs:
using: 'node20'
using: 'node12'
main: 'converge/index.js'

101
build-and-publish/README.md Normal file
View File

@@ -0,0 +1,101 @@
<p align="center">
<img src="https://github.com/werf/werf/raw/master/docs/images/werf-logo.svg?sanitize=true" style="max-height:100%;" height="175">
</p>
___
The action combines all the necessary steps in itself and logic may be divided into environment setup and launching `werf build-and-publish`.
## Action in Details
### werf binary setup
By default, all actions setup actual werf version for [1.1 alpha channel](https://werf.io/releases.html) (more details about channels, werf release cycle and compatibility promise [here](https://github.com/werf/werf#backward-compatibility-promise)).
Using the `channel` input the user can switch the release channel.
> This is recommended approach to be up-to-date and to use actual werf version without changing configurations
```yaml
- uses: werf/actions/build-and-publish@v1.1
with:
channel: alpha
```
Withal, it is not necessary to work within release channels, and the user might specify certain werf version with `version` input.
```yaml
- uses: werf/actions/build-and-publish@v1.1
with:
version: v1.1.23
```
### kubeconfig setup (*optional*)
The _kubeconfig_ may be used for deployment, cleanup, distributed locks and caches. Thus, the configuration should be added before step with the action or passed as base64 encoded data with `kube-config-base64-data` input:
* Prepare _kubeconfig_ (e.g. `cat ~/.kube/config | base64`) and save in GitHub Project Secrets (e.g. with name `KUBE_CONFIG_BASE64_DATA`).
* Pass secret with `kube-config-base64-data` input:
```yaml
- uses: werf/actions/build-and-publish@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
### werf ci-env
This command performs _docker login_ using `github-token`, sets up predefined variables based on GitHub Workflow context.
**Note** that `github-token` is optional in this action, and the input is there in case you need to use a non-default token.
By default, action will use the token provided to your workflow.
## Working with werf options
Any werf option can be defined with environment variables:
```yaml
- uses: werf/actions/build-and-publish@v1.1
env:
WERF_LOG_VERBOSE: "on"
WERF_TAG_CUSTOM_TAG1: tag1
WERF_TAG_CUSTOM_TAG2: tag2
```
## Inputs
```yaml
channel:
description: 'The one of the following channel: alpha, beta, ea, stable, rock-solid'
default: 'alpha'
required: false
version:
description: 'The certain version'
required: false
github-token:
description: 'The GitHub token used to login and to interact with Docker Github Packages'
default: ${{ github.token }}
required: false
kube-config-base64-data:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
```
## Example
```yaml
build-and-publish:
name: Build and Publish
runs-on: ubuntu-latest
steps:
- name: Checkout code
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Build and Publish
uses: werf/actions/build-and-publish@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```

View File

@@ -0,0 +1,24 @@
name: werf build-and-publish
author: 'Flant'
description: 'Prepare the environment and perform image building and publishing with werf'
branding:
color: blue
icon: anchor
inputs:
channel:
description: 'The one of the following channel: alpha, beta, ea, stable, rock-solid'
default: 'alpha'
required: false
version:
description: 'The certain version'
required: false
github-token:
description: 'The GitHub token used to login and to interact with Docker Github Packages'
default: ${{ github.token }}
required: false
kube-config-base64-data:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
runs:
using: 'node12'
main: 'index.js'

30418
build-and-publish/index.js Normal file

File diff suppressed because one or more lines are too long

View File

@@ -5,6 +5,63 @@ ___
The action combines all the necessary steps in itself and logic may be divided into environment setup and launching `werf build`.
## Action in Details
### werf binary setup
By default, all actions setup actual werf version for [1.1 alpha channel](https://werf.io/releases.html) (more details about channels, werf release cycle and compatibility promise [here](https://werf.io/installation.html#all-changes-in-werf-go-through-all-stability-channels)).
Using the `channel` input the user can switch the release channel.
> This is recommended approach to be up-to-date and to use actual werf version without changing configurations
```yaml
- uses: werf/actions/build@v1.1
with:
channel: alpha
```
Withal, it is not necessary to work within release channels, and the user might specify certain werf version with `version` input.
```yaml
- uses: werf/actions/build@v1.1
with:
version: v1.1.23
```
### kubeconfig setup (*optional*)
The _kubeconfig_ may be used for deployment, cleanup, distributed locks and caches. Thus, the configuration should be added before step with the action or passed as base64 encoded data with `kube-config-base64-data` input:
* Prepare _kubeconfig_ (e.g. `cat ~/.kube/config | base64`) and save in GitHub Project Secrets (e.g. with name `KUBE_CONFIG_BASE64_DATA`).
* Pass secret with `kube-config-base64-data` input:
```yaml
- uses: werf/actions/build@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
### werf ci-env
This command performs _docker login_ using `github-token`, sets up predefined variables based on GitHub Workflow context.
**Note** that `github-token` is optional in this action, and the input is there in case you need to use a non-default token.
By default, action will use the token provided to your workflow.
## Working with werf options
Any werf option can be defined with environment variables:
```yaml
- uses: werf/actions/build@v1.1
env:
WERF_LOG_VERBOSE: "on"
WERF_TAG_CUSTOM_TAG1: tag1
WERF_TAG_CUSTOM_TAG2: tag2
```
## Inputs
```yaml
@@ -33,12 +90,12 @@ build:
steps:
- name: Checkout code
uses: actions/checkout@v4
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Build
uses: werf/actions/build@v2
uses: werf/actions/build@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```

View File

@@ -20,5 +20,5 @@ inputs:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
runs:
using: 'node20'
using: 'node12'
main: 'index.js'

File diff suppressed because one or more lines are too long

View File

@@ -5,6 +5,61 @@ ___
The action combines all the necessary steps in itself and logic may be divided into environment setup and launching `werf cleanup`.
## Action in Details
### werf binary setup
By default, all actions setup actual werf version for [1.1 alpha channel](https://werf.io/releases.html) (more details about channels, werf release cycle and compatibility promise [here](https://werf.io/installation.html#all-changes-in-werf-go-through-all-stability-channels)).
Using the `channel` input the user can switch the release channel.
> This is recommended approach to be up-to-date and to use actual werf version without changing configurations
```yaml
- uses: werf/actions/cleanup@v1.1
with:
channel: alpha
```
Withal, it is not necessary to work within release channels, and the user might specify certain werf version with `version` input.
```yaml
- uses: werf/actions/cleanup@v1.1
with:
version: v1.1.23
```
### kubeconfig setup (*optional*)
The _kubeconfig_ may be used for deployment, cleanup, distributed locks and caches. Thus, the configuration should be added before step with the action or passed as base64 encoded data with `kube-config-base64-data` input:
* Prepare _kubeconfig_ (e.g. `cat ~/.kube/config | base64`) and save in GitHub Project Secrets (e.g. with name `KUBE_CONFIG_BASE64_DATA`).
* Pass secret with `kube-config-base64-data` input:
```yaml
- uses: werf/actions/cleanup@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
### werf ci-env
This command performs _docker login_ using `github-token`, sets up predefined variables based on GitHub Workflow context.
**Note** that `github-token` is optional in this action, and the input is there in case you need to use a non-default token.
By default, action will use the token provided to your workflow.
## Working with werf options
Any werf option can be defined with environment variables:
```yaml
- uses: werf/actions/cleanup@v1.1
env:
WERF_LOG_VERBOSE: "on"
```
## Inputs
```yaml
@@ -33,15 +88,13 @@ cleanup:
steps:
- name: Checkout code
uses: actions/checkout@v4
uses: actions/checkout@v2
- name: Fetch all history for all tags and branches
run: git fetch --prune --unshallow
- name: Cleanup
uses: werf/actions/cleanup@v2
uses: werf/actions/cleanup@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
env:
WERF_REPO_GITHUB_TOKEN: ${{ secrets.WERF_CLEANUP_PAM }}
```

View File

@@ -20,5 +20,5 @@ inputs:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
runs:
using: 'node20'
using: 'node12'
main: 'index.js'

File diff suppressed because one or more lines are too long

View File

@@ -5,6 +5,63 @@ ___
The action combines all the necessary steps in itself and logic may be divided into environment setup and launching `werf converge`.
## Action in Details
### werf binary setup
By default, all actions setup actual werf version for [1.1 alpha channel](https://werf.io/releases.html) (more details about channels, werf release cycle and compatibility promise [here](https://werf.io/installation.html#all-changes-in-werf-go-through-all-stability-channels)).
Using the `channel` input the user can switch the release channel.
> This is recommended approach to be up-to-date and to use actual werf version without changing configurations
```yaml
- uses: werf/actions/converge@v1.1
with:
channel: alpha
```
Withal, it is not necessary to work within release channels, and the user might specify certain werf version with `version` input.
```yaml
- uses: werf/actions/converge@v1.1
with:
version: v1.1.23
```
### kubeconfig setup (*optional*)
The _kubeconfig_ may be used for deployment, cleanup, distributed locks and caches. Thus, the configuration should be added before step with the action or passed as base64 encoded data with `kube-config-base64-data` input:
* Prepare _kubeconfig_ (e.g. `cat ~/.kube/config | base64`) and save in GitHub Project Secrets (e.g. with name `KUBE_CONFIG_BASE64_DATA`).
* Pass secret with `kube-config-base64-data` input:
```yaml
- uses: werf/actions/converge@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
### werf ci-env
This command performs _docker login_ using `github-token`, sets up predefined variables based on GitHub Workflow context.
**Note** that `github-token` is optional in this action, and the input is there in case you need to use a non-default token.
By default, action will use the token provided to your workflow.
## Working with werf options
Any werf option can be defined with environment variables:
```yaml
- uses: werf/actions/converge@v1.1
with:
env: production
env:
WERF_LOG_VERBOSE: "on"
```
## Inputs
```yaml
@@ -36,12 +93,12 @@ converge:
steps:
- name: Checkout code
uses: actions/checkout@v4
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Converge
uses: werf/actions/converge@v2
uses: werf/actions/converge@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
env: production

View File

@@ -12,9 +12,9 @@ inputs:
version:
description: 'The certain version'
required: false
env: # TODO: Remove this deprecated werf option, which is not required anymore (use WERF_ENV variable instead)
env:
description: 'Specific deployment environment'
required: false
required: true
github-token:
description: 'The GitHub token used to login and to interact with Docker Github Packages'
default: ${{ github.token }}
@@ -23,5 +23,5 @@ inputs:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
runs:
using: 'node20'
using: 'node12'
main: 'index.js'

File diff suppressed because one or more lines are too long

105
deploy/README.md Normal file
View File

@@ -0,0 +1,105 @@
<p align="center">
<img src="https://github.com/werf/werf/raw/master/docs/images/werf-logo.svg?sanitize=true" style="max-height:100%;" height="175">
</p>
___
The action combines all the necessary steps in itself and logic may be divided into environment setup and launching `werf deploy`.
## Action in Details
### werf binary setup
By default, all actions setup actual werf version for [1.1 alpha channel](https://werf.io/releases.html) (more details about channels, werf release cycle and compatibility promise [here](https://github.com/werf/werf#backward-compatibility-promise)).
Using the `channel` input the user can switch the release channel.
> This is recommended approach to be up-to-date and to use actual werf version without changing configurations
```yaml
- uses: werf/actions/deploy@v1.1
with:
channel: alpha
```
Withal, it is not necessary to work within release channels, and the user might specify certain werf version with `version` input.
```yaml
- uses: werf/actions/deploy@v1.1
with:
version: v1.1.23
```
### kubeconfig setup (*optional*)
The _kubeconfig_ may be used for deployment, cleanup, distributed locks and caches. Thus, the configuration should be added before step with the action or passed as base64 encoded data with `kube-config-base64-data` input:
* Prepare _kubeconfig_ (e.g. `cat ~/.kube/config | base64`) and save in GitHub Project Secrets (e.g. with name `KUBE_CONFIG_BASE64_DATA`).
* Pass secret with `kube-config-base64-data` input:
```yaml
- uses: werf/actions/deploy@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
### werf ci-env
This command performs _docker login_ using `github-token`, sets up predefined variables based on GitHub Workflow context.
**Note** that `github-token` is optional in this action, and the input is there in case you need to use a non-default token.
By default, action will use the token provided to your workflow.
## Working with werf options
Any werf option can be defined with environment variables:
```yaml
- uses: werf/actions/deploy@v1.1
with:
env: production
env:
WERF_LOG_VERBOSE: "on"
```
## Inputs
```yaml
channel:
description: 'The one of the following channel: alpha, beta, ea, stable, rock-solid'
default: 'alpha'
required: false
version:
description: 'The certain version'
required: false
env:
description: 'Specific deployment environment'
required: true
github-token:
description: 'The GitHub token used to login and to interact with Docker Github Packages'
default: ${{ github.token }}
required: false
kube-config-base64-data:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
```
## Example
```yaml
deploy:
name: Deploy
runs-on: ubuntu-latest
steps:
- name: Checkout code
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Deploy
uses: werf/actions/deploy@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
env: production
```

27
deploy/action.yml Normal file
View File

@@ -0,0 +1,27 @@
name: werf build-and-publish
author: 'Flant'
description: 'Prepare the environment and perform deployment with werf'
branding:
color: blue
icon: anchor
inputs:
channel:
description: 'The one of the following channel: alpha, beta, ea, stable, rock-solid'
default: 'alpha'
required: false
version:
description: 'The certain version'
required: false
env:
description: 'Specific deployment environment'
required: true
github-token:
description: 'The GitHub token used to login and to interact with Docker Github Packages'
default: ${{ github.token }}
required: false
kube-config-base64-data:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
runs:
using: 'node12'
main: 'index.js'

30419
deploy/index.js Normal file

File diff suppressed because one or more lines are too long

View File

@@ -5,6 +5,63 @@ ___
The action combines all the necessary steps in itself and logic may be divided into environment setup and launching `werf dismiss`.
## Action in Details
### werf binary setup
By default, all actions setup actual werf version for [1.1 alpha channel](https://werf.io/releases.html) (more details about channels, werf release cycle and compatibility promise [here](https://werf.io/installation.html#all-changes-in-werf-go-through-all-stability-channels)).
Using the `channel` input the user can switch the release channel.
> This is recommended approach to be up-to-date and to use actual werf version without changing configurations
```yaml
- uses: werf/actions/converge@v1.1
with:
channel: alpha
```
Withal, it is not necessary to work within release channels, and the user might specify certain werf version with `version` input.
```yaml
- uses: werf/actions/converge@v1.1
with:
version: v1.1.23
```
### kubeconfig setup (*optional*)
The _kubeconfig_ may be used for deployment, cleanup, distributed locks and caches. Thus, the configuration should be added before step with the action or passed as base64 encoded data with `kube-config-base64-data` input:
* Prepare _kubeconfig_ (e.g. `cat ~/.kube/config | base64`) and save in GitHub Project Secrets (e.g. with name `KUBE_CONFIG_BASE64_DATA`).
* Pass secret with `kube-config-base64-data` input:
```yaml
- uses: werf/actions/converge@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
### werf ci-env
This command performs _docker login_ using `github-token`, sets up predefined variables based on GitHub Workflow context.
**Note** that `github-token` is optional in this action, and the input is there in case you need to use a non-default token.
By default, action will use the token provided to your workflow.
## Working with werf options
Any werf option can be defined with environment variables:
```yaml
- uses: werf/actions/converge@v1.1
with:
env: production
env:
WERF_LOG_VERBOSE: "on"
```
## Inputs
```yaml
@@ -36,10 +93,10 @@ dismiss:
steps:
- name: Checkout code
uses: actions/checkout@v4
uses: actions/checkout@v2
- name: Dismiss
uses: werf/actions/dismiss@v2
uses: werf/actions/dismiss@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
env: production

View File

@@ -12,9 +12,9 @@ inputs:
version:
description: 'The certain version'
required: false
env: # TODO: Remove this deprecated werf option, which is not required anymore (use WERF_ENV variable instead)
env:
description: 'Specific deployment environment'
required: false
required: true
github-token:
description: 'The GitHub token used to login and to interact with Docker Github Packages'
default: ${{ github.token }}
@@ -23,5 +23,5 @@ inputs:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
runs:
using: 'node20'
using: 'node12'
main: 'index.js'

File diff suppressed because one or more lines are too long

View File

@@ -3,13 +3,13 @@
</p>
___
By default, all actions setup actual werf version for 1.2 alpha channel (more details about channels, werf release cycle and compatibility promise [here](https://werf.io/installation.html#all-changes-in-werf-go-through-all-stability-channels)).
By default, all actions setup actual werf version for [1.1 alpha channel](https://werf.io/releases.html) (more details about channels, werf release cycle and compatibility promise [here](https://werf.io/installation.html#all-changes-in-werf-go-through-all-stability-channels)).
Using the `channel` input the user can switch the release channel.
> This is recommended approach to be up-to-date and to use actual werf version without changing configurations
```yaml
- uses: werf/actions/install@v2
- uses: werf/actions/install@v1.1
with:
channel: alpha
```
@@ -17,9 +17,9 @@ Using the `channel` input the user can switch the release channel.
Withal, it is not necessary to work within release channels, and the user might specify certain werf version with `version` input.
```yaml
- uses: werf/actions/install@v2
- uses: werf/actions/install@v1.1
with:
version: v2.1.0
version: v1.1.23
```
## Inputs
@@ -43,12 +43,12 @@ werf:
steps:
- name: Checkout code
uses: actions/checkout@v4
uses: actions/checkout@v2
- name: Install werf CLI
uses: werf/actions/install@v2
uses: werf/actions/install@v1.1
# For deploy and distributed locks.
# for deploy and distributed locks
- name: Create kube config
run: |
KUBECONFIG=$(mktemp -d)/config
@@ -60,9 +60,9 @@ werf:
- name: Run werf commands
run: |
source $(werf ci-env github --as-file)
werf render
werf converge
werf build-and-publish
werf deploy
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
GITHUB_TOKEN: ${{ github.token }}
WERF_ENV: production
```

View File

@@ -1,6 +1,6 @@
name: Install werf CLI
author: 'Flant'
description: 'Install werf CLI'
description: 'Install actual channel werf version, like multiwerf does, or specified one'
branding:
color: blue
icon: anchor
@@ -17,5 +17,5 @@ inputs:
default: '0'
required: false
runs:
using: 'node20'
using: 'node12'
main: 'index.js'

File diff suppressed because one or more lines are too long

View File

@@ -1,3 +1,3 @@
#!/bin/bash -e
for pkg in build cleanup converge dismiss install run; do ncc build src/$pkg.ts -o $pkg; done
for pkg in build build-and-publish cleanup converge deploy dismiss install publish run; do ncc build src/$pkg.ts -o $pkg; done

3942
package-lock.json generated

File diff suppressed because it is too large Load Diff

View File

@@ -1,14 +1,14 @@
{
"name": "werf-actions",
"version": "1.2.1",
"version": "1.1.1",
"description": "",
"main": "lib",
"dependencies": {
"@actions/core": "^1.9.1",
"@actions/github": "^6.0.0",
"@actions/core": "^1.6.0",
"@actions/github": "^5.0.0",
"@actions/tool-cache": "^1.7.1",
"dotenv": "10.0.x",
"semver": "^7.5.2",
"semver": "^7.3.5",
"string-argv": "^0.3.1",
"superagent": "^6.1.0",
"tmp": "^0.2.1",
@@ -25,7 +25,7 @@
"eslint": "^7.32.0",
"eslint-plugin-github": "^4.3.2",
"js-yaml": "^3.13.1",
"node-fetch": ">=3.2.10",
"node-fetch": ">=2.6.1",
"prettier": "^1.19.1",
"typescript": "^3.6.4"
},
@@ -34,7 +34,7 @@
"format": "prettier --write **/*.ts",
"format-check": "prettier --check **/*.ts",
"lint": "eslint src/**/*.ts",
"pack": "./pack.sh",
"pack": "pack.sh",
"all": "npm run build && npm run format && npm run lint && npm run pack"
},
"repository": {

101
publish/README.md Normal file
View File

@@ -0,0 +1,101 @@
<p align="center">
<img src="https://github.com/werf/werf/raw/master/docs/images/werf-logo.svg?sanitize=true" style="max-height:100%;" height="175">
</p>
___
The action combines all the necessary steps in itself and logic may be divided into environment setup and launching `werf publish`.
## Action in Details
### werf binary setup
By default, all actions setup actual werf version for [1.1 alpha channel](https://werf.io/releases.html) (more details about channels, werf release cycle and compatibility promise [here](https://github.com/werf/werf#backward-compatibility-promise)).
Using the `channel` input the user can switch the release channel.
> This is recommended approach to be up-to-date and to use actual werf version without changing configurations
```yaml
- uses: werf/actions/publish@v1.1
with:
channel: alpha
```
Withal, it is not necessary to work within release channels, and the user might specify certain werf version with `version` input.
```yaml
- uses: werf/actions/publish@v1.1
with:
version: v1.1.23
```
### kubeconfig setup (*optional*)
The _kubeconfig_ may be used for deployment, cleanup, distributed locks and caches. Thus, the configuration should be added before step with the action or passed as base64 encoded data with `kube-config-base64-data` input:
* Prepare _kubeconfig_ (e.g. `cat ~/.kube/config | base64`) and save in GitHub Project Secrets (e.g. with name `KUBE_CONFIG_BASE64_DATA`).
* Pass secret with `kube-config-base64-data` input:
```yaml
- uses: werf/actions/publish@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
### werf ci-env
This command performs _docker login_ using `github-token`, sets up predefined variables based on GitHub Workflow context.
**Note** that `github-token` is optional in this action, and the input is there in case you need to use a non-default token.
By default, action will use the token provided to your workflow.
## Working with werf options
Any werf option can be defined with environment variables:
```yaml
- uses: werf/actions/publish@v1.1
env:
WERF_LOG_VERBOSE: "on"
WERF_TAG_CUSTOM_TAG1: tag1
WERF_TAG_CUSTOM_TAG2: tag2
```
## Inputs
```yaml
channel:
description: 'The one of the following channel: alpha, beta, ea, stable, rock-solid'
default: 'alpha'
required: false
version:
description: 'The certain version'
required: false
github-token:
description: 'The GitHub token used to login and to interact with Docker Github Packages'
default: ${{ github.token }}
required: false
kube-config-base64-data:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
```
## Example
```yaml
publish:
name: Publish
runs-on: ubuntu-latest
steps:
- name: Checkout code
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Publish
uses: werf/actions/publish@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```

24
publish/action.yml Normal file
View File

@@ -0,0 +1,24 @@
name: werf publish
author: 'Flant'
description: 'Prepare the environment and perform image publishing with werf'
branding:
color: blue
icon: anchor
inputs:
channel:
description: 'The one of the following channel: alpha, beta, ea, stable, rock-solid'
default: 'alpha'
required: false
version:
description: 'The certain version'
required: false
github-token:
description: 'The GitHub token used to login and to interact with Docker Github Packages'
default: ${{ github.token }}
required: false
kube-config-base64-data:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
runs:
using: 'node12'
main: 'index.js'

30418
publish/index.js Normal file

File diff suppressed because one or more lines are too long

View File

@@ -5,6 +5,63 @@ ___
The action combines all the necessary steps in itself and logic may be divided into environment setup and launching `werf run`.
## Action in Details
### werf binary setup
By default, all actions setup actual werf version for [1.1 alpha channel](https://werf.io/releases.html) (more details about channels, werf release cycle and compatibility promise [here](https://werf.io/installation.html#all-changes-in-werf-go-through-all-stability-channels)).
Using the `channel` input the user can switch the release channel.
> This is recommended approach to be up-to-date and to use actual werf version without changing configurations
```yaml
- uses: werf/actions/run@v1.1
with:
channel: alpha
```
Withal, it is not necessary to work within release channels, and the user might specify certain werf version with `version` input.
```yaml
- uses: werf/actions/run@v1.1
with:
version: v1.1.23
```
### kubeconfig setup (*optional*)
The _kubeconfig_ may be used for deployment, cleanup, distributed locks and caches. Thus, the configuration should be added before step with the action or passed as base64 encoded data with `kube-config-base64-data` input:
* Prepare _kubeconfig_ (e.g. `cat ~/.kube/config | base64`) and save in GitHub Project Secrets (e.g. with name `KUBE_CONFIG_BASE64_DATA`).
* Pass secret with `kube-config-base64-data` input:
```yaml
- uses: werf/actions/run@v1.1
with:
kube-config-base64-data: ${{ secrets.KUBE_CONFIG_BASE64_DATA }}
```
### werf ci-env
This command performs _docker login_ using `github-token`, sets up predefined variables based on GitHub Workflow context.
**Note** that `github-token` is optional in this action, and the input is there in case you need to use a non-default token.
By default, action will use the token provided to your workflow.
## Working with werf options
Any werf option can be defined with environment variables:
```yaml
- uses: werf/actions/run@v1.1
env:
WERF_LOG_VERBOSE: "on"
WERF_TAG_CUSTOM_TAG1: tag1
WERF_TAG_CUSTOM_TAG2: tag2
```
## Inputs
```yaml
@@ -40,12 +97,12 @@ run:
steps:
- name: Checkout code
uses: actions/checkout@v4
uses: actions/checkout@v2
with:
fetch-depth: 0
- name: Run
uses: werf/actions/run@v2
uses: werf/actions/run@v1.1
with:
image: backend
args: rails server

View File

@@ -26,5 +26,5 @@ inputs:
description: 'Base64 encoded kubeconfig data used for deployment, cleanup and distributed locks'
required: false
runs:
using: 'node20'
using: 'node12'
main: 'index.js'

24359
run/index.js

File diff suppressed because one or more lines are too long

12
src/build-and-publish.ts Normal file
View File

@@ -0,0 +1,12 @@
import * as core from '@actions/core'
import {PrepareEnvironAndRunWerfCommand} from './common'
async function run(): Promise<void> {
try {
await PrepareEnvironAndRunWerfCommand(['build-and-publish'])
} catch (error) {
core.setFailed(error.message)
}
}
run()

View File

@@ -7,6 +7,8 @@ import {String} from 'typescript-string-operations'
import {Manager} from './manager'
import * as werf from './werf'
const minimalWerfVersion = 'v1.1.17'
export async function PrepareEnvironAndRunWerfCommand(
args: string[]
): Promise<void> {
@@ -45,17 +47,13 @@ export async function SetupKubeConfig(
export function ProcessGitHubContext(): void {
if (context.eventName === 'pull_request') {
if (context.payload.pull_request) {
// Do nothing if virtual merge variable is already set
if (process.env.WERF_VIRTUAL_MERGE) {
return
}
const baseSha = context.payload.pull_request.base.sha
const headSha = context.payload.pull_request.head.sha
process.env.WERF_VIRTUAL_MERGE = '1'
process.env.WERF_VIRTUAL_MERGE_FROM_COMMIT = headSha
process.env.WERF_VIRTUAL_MERGE_INTO_COMMIT = baseSha
core.exportVariable('WERF_VIRTUAL_MERGE', '1')
core.exportVariable('WERF_VIRTUAL_MERGE_FROM_COMMIT', headSha)
core.exportVariable('WERF_VIRTUAL_MERGE_INTO_COMMIT', baseSha)
@@ -65,15 +63,27 @@ export function ProcessGitHubContext(): void {
export function ValidateWerfVersion(version: string): void {
const ver = semver.coerce(version)
if (ver && ver.major === werf.MAJOR) {
return
if (ver) {
if (ver.major !== werf.MAJOR || ver.minor !== werf.MINOR) {
throw new Error(
String.Format(
'The arbitrary version ({0}) must be within the MAJOR.MINOR ({1})',
version.trim(),
werf.MAJOR_MINOR_GROUP
)
)
}
if (semver.gte(ver, minimalWerfVersion)) {
return
}
}
throw new Error(
String.Format(
'werf version {0} is not supported: the version must be semver and within the MAJOR ({1})',
'werf version {0} is not supported (expected version must be equal or greater than {1})',
version.trim(),
werf.GROUP
minimalWerfVersion
)
)
}

13
src/deploy.ts Normal file
View File

@@ -0,0 +1,13 @@
import * as core from '@actions/core'
import {PrepareEnvironAndRunWerfCommand} from './common'
async function run(): Promise<void> {
try {
process.env.WERF_ENV = core.getInput('env')
await PrepareEnvironAndRunWerfCommand(['deploy'])
} catch (error) {
core.setFailed(error.message)
}
}
run()

View File

@@ -1,11 +1,8 @@
import * as core from '@actions/core'
import {Manager} from './manager'
import {ProcessGitHubContext} from './common'
async function run(): Promise<void> {
try {
ProcessGitHubContext()
const m = new Manager()
await m.Install()
} catch (error) {

View File

@@ -83,11 +83,6 @@ export class Manager {
const tmpFilePath = tmpFile.name
await this.Exec(['ci-env', 'github', '--as-env-file', '-o', tmpFilePath])
const res = dotenv.config({path: tmpFilePath})
if (res.parsed) {
for (const [key, value] of Object.entries(res.parsed)) {
core.exportVariable(key, value)
}
}
console.log(res.parsed)
tmpFile.removeCallback()
}
@@ -125,7 +120,7 @@ export class Manager {
return this._constructReleaseUrl(version)
}
const url = `${WERF_TUF_SERVER_URL}/targets/channels/${werf.GROUP}/${this.channel}`
const url = `${WERF_TUF_SERVER_URL}/targets/channels/${werf.MAJOR_MINOR_GROUP}/${this.channel}`
try {
const resp = await request
.get(url)

12
src/publish.ts Normal file
View File

@@ -0,0 +1,12 @@
import * as core from '@actions/core'
import {PrepareEnvironAndRunWerfCommand} from './common'
async function run(): Promise<void> {
try {
await PrepareEnvironAndRunWerfCommand(['publish'])
} catch (error) {
core.setFailed(error.message)
}
}
run()

View File

@@ -1,2 +1,3 @@
export const GROUP = '2'
export const MAJOR = 2
export const MAJOR_MINOR_GROUP = '1.1'
export const MAJOR = 1
export const MINOR = 1