GeoServer Configuration Guide

This guide covers essential configuration tasks for GeoServer in the SDL platform, including GeoWebCache setup, data stores, and security settings.

GeoWebCache S3 BlobStore Configuration

GeoWebCache in SDL is configured to use MinIO (S3-compatible storage) for distributed tile caching. This configuration enables multiple GeoServer instances to share the same tile cache and provides better scalability.

Configuration Steps

  1. Navigate to Tile Caching > BlobStores in the GeoServer admin interface

  2. Click Add new BlobStore

  3. Select S3 BlobStore from the Type dropdown

  4. Fill in the configuration fields as shown below:

S3 BlobStore Configuration

Configuration Field Reference

Field Description

Identifier

Name for the BlobStore

Enabled

Self-explanatory

Default

Make this the default BlobStore for new layers

Bucket

S3 bucket name for storing tiles

AWS Access Key

MinIO access key (from df-minio secret, get from k9s or see commands below)

AWS Secret Key

MinIO secret key (from df-minio secret, get from k9s or see commands below)

S3 Object Key Prefix

Minio directory name within geoserver bucket

Endpoint

MinIO service URL, e.g. http://df-minio:9000

Maximum Connections

Connection pool size

Use HTTPS

Enable HTTPS for S3 connections, off if e.g. http://df-minio:9000

Use Gzip

Enable gzip compression for transfers

Access Type

Public or Private bucket access

Retrieving MinIO Credentials

The MinIO credentials are stored in the df-minio Kubernetes secret:

# Get MinIO access key
kubectl get secret -n data-fabric df-minio -o jsonpath='{.data.rootUser}' | base64 -d

# Get MinIO secret key
kubectl get secret -n data-fabric df-minio -o jsonpath='{.data.rootPassword}' | base64 -d

Verifying Configuration

After saving the BlobStore configuration:

  1. Navigate to Tile Caching > Tile Layers

  2. Select a layer to cache

  3. In the layer configuration, set BlobStore to sdi-minio

  4. Save the layer configuration

  5. Test by going to map.yourURL/geoserver/gwc/demo, clicking Seed this layer, submitting a seed job, and checking if tiles appear in MinIO.

NASA Blue Marble S3 Cache Configuration
GeoWebCache Seeding Demo
Cached S3 Tiles

PostGIS Datastore Configuration

Creating a PostGIS Datastore

  1. Navigate to Data > Stores in GeoServer

  2. Click Add new Store

  3. Select PostGIS under Vector Data Sources

  4. Configure the connection:

Parameter Description Value

Workspace

Workspace for this datastore

nasa

Data Source Name

Unique name for this datastore

df-postgis

Description

Optional description

PostGIS spatial database

host

Database server hostname

df-postgres

port

Database server port

5432

database

Database name

geoserver_data

schema

Database schema

public

user

Database username

datahub

passwd

Database password

Retrieved from postgres-secrets (see command below)

Validate connections

Test connections before use

✓ (checked)
Total connections across all GeoServer instances should not exceed PostgreSQL’s max_connections setting. 
# Get Postgres Datahub Password
kubectl get secret -n data-fabric postgres-secrets -o jsonpath='{.data.postgresDatahubSecret}' | base64 -d
PostGIS Store Configuration

Security Configuration

Keycloak Integration Settings

The OIDC plugin configuration for Keycloak integration:

Setting Value

Client ID

df-geoserver

Client Secret

From keycloak-realm-init secret

Discovery URL

http://df-keycloak/auth/realms/data-fabric/.well-known/openid-configuration

Roles Claim

resource_access.df-geoserver.roles

Send Client Secret

IN_PARAMS

Logout URI

/geoserver/web/logout

Performance Tuning

JVM Settings

Optimal JVM settings for production:

env:
  - name: JAVA_OPTS
    value: >-
      -Xms4g -Xmx8g
      -XX:+UseG1GC
      -XX:MaxGCPauseMillis=200
      -XX:ParallelGCThreads=4
      -XX:ConcGCThreads=2
      -Dfile.encoding=UTF-8
      -Djava.awt.headless=true
      -Djavax.servlet.request.encoding=UTF-8
      -Djavax.servlet.response.encoding=UTF-8
      -Duser.timezone=UTC
      -Dorg.geotools.coverage.jaiext.enabled=true

Resource Limits

Configure Kubernetes resource limits appropriately:

resources:
  requests:
    cpu: "2"
    memory: "4Gi"
  limits:
    cpu: "4"
    memory: "8Gi"

Next Steps