When deploying Quay in a Secure AWS environment, we can't use IAM Access Keys or Secrets since these credentials are often blocked for multiple reasons (credentials are long-lived, can be shared / stolen, etc.). So the preferred deployment method is to use an alternative method, like the Web Identity Token files that are automatically created in a Kubernetes cluster that has a federation link with IAM using the OIDC provider federation. The current code of Quay force the use of an IAM account that is then used to assume another role that has S3 access to store the image files. The current pull request removes the need to use that IAM account and allows to directly assume the correct role using Web Identity Tokens while retaining compatibility with the old method of using IAM credentials. The code relies on the automatic detection of the correct configurations using environment variables where possible. The code has been tested on an OpenShift cluster deployed using manual mode with AWS STS.
Config Tool
The Quay Config Tool implements several features to capture and validate configuration data based on a predefined schema.
This tool includes the following features:
- Validate Quay configuration using CLI tool
- Generate code for custom field group definitions (includes structs, constructors, defaults)
- Validation tag support from Validator
- Built-in validator tags for OAuth and JWT structs
Installation
Build from Source
Install using the Go tool:
go get -u github.com/quay/quay/config-tool/...
This will generate files for the Quay validator executable and install the config-tool
CLI tool.
Build from Dockerfile
Clone this repo and build an image:
$ git clone https://github.com/quay/quay.git
$ cd quay/config-tool
$ sudo podman build -t config-tool .
Start the container and execute command:
$ sudo podman run -it -v ${CONFIG_MOUNT}:/conf config-tool ...
Note that you must mount in your config directory in order for the config-tool to see it.
Note: By default, this tool will generate an executable from a pre-built Config definition. For usage on writing a custom Config definition see here
Usage
The CLI tool contains two main commands:
The print
command is used to output the entire configuration with defaults specified
{
"HostSettings": (*fieldgroups.HostSettingsFieldGroup)({
ServerHostname: "quay:8081",
PreferredURLScheme: "https",
ExternalTLSTermination: false
}),
"TagExpiration": (*fieldgroups.TagExpirationFieldGroup)({
FeatureChangeTagExpiration: false,
DefaultTagExpiration: "2w",
TagExpirationOptions: {
"0s",
"1d",
"1w",
"2w",
"4w"
}
}),
"UserVisibleSettings": (*fieldgroups.UserVisibleSettingsFieldGroup)({
RegistryTitle: "Project Quay",
RegistryTitleShort: "Project Quay",
SearchResultsPerPage: 10,
SearchMaxResultPageCount: 10,
ContactInfo: {
},
AvatarKind: "local",
Branding: (*fieldgroups.BrandingStruct)({
Logo: "not_a_url",
FooterIMG: "also_not_a_url",
FooterURL: ""
})
})
}
The validate
command is used to show while field groups have been validated succesully
$ config-tool validate -c <path-to-config-dir>
+---------------------+--------------------+-------------------------+--------+
| FIELD GROUP | FIELD | ERROR | STATUS |
+---------------------+--------------------+-------------------------+--------+
| HostSettings | - | - | 🟢 |
| TagExpiration | - | - | 🟢 |
| UserVisibleSettings | BRANDING.Logo | Field enforces tag: url | 🔴 |
| | BRANDING.FooterIMG | Field enforces tag: url | 🔴 |
+---------------------+--------------------+-------------------------+--------+
The editor
command will bring up an interactive UI to reconfigure and validate a config bundle.
$ config-tool editor -c <path-to-config-dir> -p <editor-password> -e <operator-endpoint>
This command will bring up an interactive UI in which a user can modify, validate, and download a config. In addition, Swagger documentation can be reached by going to {{host}}/swagger/index.html
Using HTTPS
You can deploy the config editor using TLS certificates by passing environment variables to the runtime. The public and private keys must contain valid SANs for the route that you wish to deploy the editor on.
The paths can be specifed using CONFIG_TOOL_PRIVATE_KEY
and CONFIG_TOOL_PUBLIC_KEY
.
NOTE: If running from a container, the CONFIG_TOOL_PRIVATE_KEY
and CONFIG_TOOL_PUBLIC_KEY
values are the locations of the certs INSIDE the container. This might look something like the following:
$ docker run -p 7070:8080 \
-v ${PRIVATE_KEY_PATH}:/tls/localhost.key \
-v ${PUBLIC_KEY_PATH}:/tls/localhost.crt \
-e CONFIG_TOOL_PRIVATE_KEY=/tls/localhost.key \
-e CONFIG_TOOL_PUBLIC_KEY=/tls/localhost.crt \
-e DEBUGLOG=true \
-ti config-app:dev