Skip to content

README.md

Latest commit

 

History

History
185 lines (133 loc) · 8.29 KB

README.md

File metadata and controls

185 lines (133 loc) · 8.29 KB

GitLab MCP Helm Chart

This Helm chart deploys the GitLab MCP (Model Context Protocol) server based on the iwakitakuma/gitlab-mcp Docker image to a Kubernetes cluster.

Prerequisites

  • Kubernetes 1.16+
  • Helm 3.2.0+

Installing the Chart

To install the chart with the release name gitlab-mcp:

helm install gitlab-mcp ./charts/gitlab

The command deploys the GitLab MCP server on the Kubernetes cluster in the default configuration. The Parameters section lists the parameters that can be configured during installation.

Uninstalling the Chart

To uninstall/delete the gitlab-mcp deployment:

helm delete gitlab-mcp

The command removes all the Kubernetes components associated with the chart and deletes the release.

Parameters

Global parameters

Name Description Value
replicaCount Number of GitLab MCP replicas to deploy 1
nameOverride String to partially override gitlab-mcp.fullname ""
fullnameOverride String to fully override gitlab-mcp.fullname ""

Image parameters

Name Description Value
image.repository GitLab MCP image repository iwakitakuma/gitlab-mcp
image.tag GitLab MCP image tag (immutable tags are recommended) 2.0.7
image.pullPolicy GitLab MCP image pull policy IfNotPresent
imagePullSecrets GitLab MCP image pull secrets []

Service Account parameters

Name Description Value
serviceAccount.create Specifies whether a service account should be created true
serviceAccount.annotations Annotations to add to the service account {}
serviceAccount.name The name of the service account to use ""

Service parameters

Name Description Value
service.type GitLab MCP service type ClusterIP
service.port GitLab MCP service HTTP port 3002

Health probe parameters

Name Description Value
livenessProbe.enabled Enable liveness probe true
livenessProbe.port Port used by the liveness probe nil
livenessProbe.initialDelaySeconds Delay before the liveness probe starts 60
livenessProbe.periodSeconds Frequency of the liveness probe 30
livenessProbe.timeoutSeconds Timeout for the liveness probe 10
livenessProbe.failureThreshold Number of consecutive failures before marking the pod as unhealthy 3
readinessProbe.enabled Enable readiness probe true
readinessProbe.port Port used by the readiness probe nil
readinessProbe.initialDelaySeconds Delay before the readiness probe starts 15
readinessProbe.periodSeconds Frequency of the readiness probe 10
readinessProbe.timeoutSeconds Timeout for the readiness probe 10
readinessProbe.failureThreshold Number of consecutive failures before marking the pod as unready 3

Both probes use the /health endpoint by default and inherit service.port.

Ingress parameters

Name Description Value
ingress.enabled Enable ingress record generation for GitLab MCP false
ingress.className IngressClass that will be be used to implement the Ingress ""
ingress.annotations Additional annotations for the Ingress resource {}
ingress.path Base path for the service /
ingress.pathType Path matching behavior Prefix
ingress.hosts List of hostnames ["gitlab-mcp.local"]
ingress.tls TLS configuration for ingress []

When ingress.path is not /, the annotation nginx.ingress.kubernetes.io/use-regex: "true" is automatically added.

Environment variables

Name Description Value
env Environment variables to be set on the container {}
envSecrets Environment variables from external secrets {}
secretEnv Environment variables to be set from created secret {}

Autoscaling parameters

Name Description Value
autoscaling.enabled Enable Horizontal Pod Autoscaler (HPA) false
autoscaling.minReplicas Minimum number of GitLab MCP replicas 1
autoscaling.maxReplicas Maximum number of GitLab MCP replicas 100
autoscaling.targetCPUUtilizationPercentage Target CPU utilization percentage 80
autoscaling.targetMemoryUtilizationPercentage Target Memory utilization percentage ""

Configuration and installation details

Setting up GitLab access

The GitLab MCP server requires access to GitLab. You can configure this using environment variables:

  1. Using plain environment variables (not recommended for production):

    env:
  GITLAB_URL: "https://gitlab.com"
  GITLAB_TOKEN: "your-gitlab-token"

  2. Using secrets (recommended for production):

    secretEnv:
  data:
    GITLAB_TOKEN: "your-gitlab-token"
env:
  GITLAB_URL: "https://gitlab.com"

Exposing the application

To access the GitLab MCP server from outside the cluster, you can:

  1. Use port forwarding (for development):

    kubectl port-forward svc/gitlab-mcp 8080:8080

  2. Enable ingress (for production):

    ingress:
  enabled: true
  className: "nginx"
  path: /
  hosts:
    - gitlab-mcp.your-domain.com

  3. Use LoadBalancer service type:

    service:
  type: LoadBalancer

Troubleshooting

Check pod status

kubectl get pods -l app.kubernetes.io/name=mcp-gitlab-helm

Check logs

kubectl logs -l app.kubernetes.io/name=mcp-gitlab-helm

Test connection

helm test gitlab-mcp