Merge branch 'agent-getting-started' into 'master'

Add Kubernetes Agent setup docs

See merge request gitlab-org/gitlab!42431
parents 83740d4e 55cb8631
---
stage: Configure
group: Configure
info: To determine the technical writer assigned to the Stage/Group associated with this page, see https://about.gitlab.com/handbook/engineering/ux/technical-writing/#designated-technical-writers
---
# GitLab Kubernetes Agent
## Goals
The [GitLab Kubernetes Agent](https://gitlab.com/gitlab-org/cluster-integration/gitlab-agent) is an active in-cluster component for solving GitLab and Kubernetes integration tasks in a secure and cloud native way.
Features:
1. Makes it possible to integrate GitLab with a Kubernetes cluster behind a firewall or NAT
1. Enables pull-based GitOps deployments by leveraging the [GitOps Engine](https://github.com/argoproj/gitops-engine)
1. Allows for real-time access to API endpoints within a cluster.
1. Many more features are planned. Please [review our roadmap](https://gitlab.com/groups/gitlab-org/-/epics/3329).
## Architecture
### GitLab Agent GitOps workflow
```mermaid
sequenceDiagram
participant D as Developer
participant A as Application code repository
participant M as Manifest repository
participant K as Kubernetes agent
participant C as Agent configuration repository
K->C: Grab the configuration
D->>+A: Pushing code changes
A->>M: Updating manifest
loop Regularly
K-->>M: Watching changes
M-->>K: Pulling and applying changes
end
```
Please refer to our [full architecture documentation in the Agent project](https://gitlab.com/gitlab-org/cluster-integration/gitlab-agent/-/blob/master/doc/architecture.md#high-level-architecture).
## Getting started with GitOps using the GitLab Agent and the GitLab Cloud Native Helm chart
There are several components that work in concert for the Agent to accomplish GitOps deployments:
1. A Kubernetes cluster that is properly configured
1. A configuration repository that contains a `config.yaml` file. This `config.yaml` tells the Agent which repositories to synchronize with.
1. A manifest repository that contains a `manifest.yaml`. This `manifest.yaml` (which can be autogenerated) is tracked by the Agent and any changes to the file are automatically applied to the cluster.
The setup process involves a few steps that, once completed, will enable GitOps deployments to work
1. Installing the Agent server via GitLab Helm chart
1. Defining a configuration directory
1. Creating an Agent record in GitLab
1. Generating and copying a Secret token used to connect to the Agent
1. Installing the Agent into the cluster
1. Creating a `manifest.yaml`
### Installing the Agent server via Helm
Currently the GitLab Kubernetes Agent can only be deployed via our [Helm chart](https://gitlab.com/gitlab-org/charts/gitlab).
NOTE: We are working quickly to [include the Agent in Official Linux Package](https://gitlab.com/gitlab-org/gitlab/-/issues/223060).
If you don't already have GitLab installed via Helm please refer to our [installation documentation](https://docs.gitlab.com/charts/installation/)
When installing/upgrading the GitLab Helm chart please consider the following Helm 2 example (if using Helm 3 please modify):
```shell
helm upgrade --force --install gitlab . \
--timeout 600 \
--set global.hosts.domain=<YOUR_DOMAIN> \
--set global.hosts.externalIP=<YOUR_IP> \
--set certmanager-issuer.email=<YOUR_EMAIL> \
--set name=gitlab-instance \
--set global.kas.enabled=true
```
`global.kas.enabled=true` must be set in order for the Agent to be properly installed and configured.
### Defining a configuration repository
Next you will need a GitLab repository that will contain your Agent configuration.
The minimal repository layout looks like this:
`.gitlab/agents/<agent-name>/config.yaml`
The `config.yaml` file contents should look like this:
```yaml
gitops:
manifest_projects:
- id: "path-to/your-awesome-project"
```
### Creating an Agent record in GitLab
Next you will need to create an GitLab Rails Agent record so that your GitLab project so that the Agent itself can associate with a GitLab project. This process will also yield a Secret that you will use to configure the Agent in subsequent steps.
There are two ways to accomplish this:
1. Via the Rails console
1. Via GraphQL
To do this you could either run `rails c` or via GraphQL. From `rails c`:
```ruby
project = ::Project.find_by_full_path("path-to/your-awesome-project")
agent = ::Clusters::Agent.create(name: "<agent-name>", project: project)
token = ::Clusters::AgentToken.create(agent: agent)
token.token # this will print out the token you need to use on the next step
```
or using GraphQL:
with this approach, you'll need a premium license to use this feature.
If you are new to using the GitLab GraphQL API please refer to the [Getting started with the GraphQL API page](../../../api/graphql/getting_started.md) or check out the [GraphQL Explorer](https://gitlab.com/-/graphql-explorer).
```json
mutation createAgent {
createClusterAgent(input: { projectPath: "path-to/your-awesome-project", name: "<agent-name>" }) {
clusterAgent {
id
name
}
errors
}
}
mutation createToken {
clusterAgentTokenCreate(input: { clusterAgentId: <cluster-agent-id-taken-from-the-previous-mutation> }) {
secret # This is the value you need to use on the next step
token {
createdAt
id
}
errors
}
}
```
Note that GraphQL will only show you the token once, after you've created it.
### Creating the Kubernetes secret
Once the token has been generated it needs to be applied to the Kubernetes cluster.
If you didn't previously define or create a namespace you need to do that first:
```shell
kubectl create namespace <YOUR-DESIRED-NAMESPACE>
```
Run the following command to create your Secret:
```shell
kubectl create secret generic -n <YOUR-DESIRED-NAMESPACE> gitlab-agent-token --from-literal=token='YOUR_AGENT_TOKEN'
```
### Installing the Agent into the cluster
Next you are now ready to install the in-cluster component of the Agent. The below is an example YAML file of the Kubernetes resources required for the Agent to be installed.
Let's highlight a few of the details in the example below:
1. You can replace `gitlab-agent` with <YOUR-DESIRED-NAMESPACE>
1. For the `kas-address` (Kubernetes Agent Server), you can replace `grpc://host.docker.internal:5005` with the address of the kas agent that was initialized via your Helm install.
1. If you defined your own secret name, then replace `gitlab-agent-token` with your secret name.
`./resources.yml`
```yaml
apiVersion: v1
kind: ServiceAccount
metadata:
name: gitlab-agent
---
apiVersion: apps/v1
kind: Deployment
metadata:
name: gitlab-agent
spec:
replicas: 1
selector:
matchLabels:
app: gitlab-agent
template:
metadata:
labels:
app: gitlab-agent
spec:
serviceAccountName: gitlab-agent
containers:
- name: agent
image: "registry.gitlab.com/gitlab-org/cluster-integration/gitlab-agent/agentk:latest"
args:
- --token-file=/config/token
- --kas-address
- grpc://host.docker.internal:5005 # {"$openapi":"kas-address"}
volumeMounts:
- name: token-volume
mountPath: /config
volumes:
- name: token-volume
secret:
secretName: gitlab-agent-token
strategy:
type: RollingUpdate
rollingUpdate:
maxSurge: 0
maxUnavailable: 1
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
name: gitlab-agent-write
rules:
- resources:
- '*'
apiGroups:
- '*'
verbs:
- create
- update
- delete
- patch
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
name: gitlab-agent-write-binding
roleRef:
name: gitlab-agent-write
kind: ClusterRole
apiGroup: rbac.authorization.k8s.io
subjects:
- name: gitlab-agent
kind: ServiceAccount
namespace: gitlab-agent
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
name: gitlab-agent-read
rules:
- resources:
- '*'
apiGroups:
- '*'
verbs:
- get
- list
- watch
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
name: gitlab-agent-read-binding
roleRef:
name: gitlab-agent-read
kind: ClusterRole
apiGroup: rbac.authorization.k8s.io
subjects:
- name: gitlab-agent
kind: ServiceAccount
namespace: gitlab-agent
```
```shell
kubectl apply -n gitlab-agent -f ./resources.yml
```
```plaintext
$ kubectl get pods --all-namespaces
NAMESPACE NAME READY STATUS RESTARTS AGE
gitlab-agent gitlab-agent-77689f7dcb-5skqk 1/1 Running 0 51s
kube-system coredns-f9fd979d6-n6wcw 1/1 Running 0 14m
kube-system etcd-minikube 1/1 Running 0 14m
kube-system kube-apiserver-minikube 1/1 Running 0 14m
kube-system kube-controller-manager-minikube 1/1 Running 0 14m
kube-system kube-proxy-j6zdh 1/1 Running 0 14m
kube-system kube-scheduler-minikube 1/1 Running 0 14m
kube-system storage-provisioner 1/1 Running 0 14m
```
### Creating a `manifest.yaml`
In the above step, you configured a `config.yaml` to point to which GitLab projects the Agent should synchronize. Within each one of those projects, you need to create a `manifest.yaml` file which the Agent will monitor. This `manifest.yaml` can be autogenerated by a templating engine or other means.
Example `manifest.yaml`:
```yaml
apiVersion: apps/v1
kind: Deployment
metadata:
name: nginx-deployment
spec:
selector:
matchLabels:
app: nginx
replicas: 2
template:
metadata:
labels:
app: nginx
spec:
containers:
- name: nginx
image: nginx:1.14.2
ports:
- containerPort: 80
```
The above file creates a simple NGINX deployment.
Each time you commit and push a change to the `manifest.yaml` the Agent will observe the change. Example log:
```plaintext
2020-09-15_14:09:04.87946 gitlab-k8s-agent : time="2020-09-15T10:09:04-04:00" level=info msg="Config: new commit" agent_id=1 commit_id=e6a3651f1faa2e928fe6120e254c122451be4eea
```
## Example projects
Basic GitOps example deploying NGINX: [Configuration repository](https://gitlab.com/gitlab-org/configure/examples/kubernetes-agent), [Manifest repository](https://gitlab.com/gitlab-org/configure/examples/gitops-project)
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment