entando-solr

Docker image for Apache Solr used by Entando, based on the official solr image.

What this image adds

An init script (entando-init.sh) that creates an entando Solr core on first startup (standalone mode).

Base image

Component	Version
Solr	8.11.4
JDK	Eclipse Temurin 11.0.27
OS	Ubuntu 20.04

Build

docker build -t entando/entando-solr:8 .

Run (standalone)

docker run -d --name solr -p 8983:8983 entando/entando-solr:8

The entando core can be created via the init script:

docker exec solr /opt/docker-solr/entando-init.sh

Standalone deployment (Kubernetes / OpenShift)

A single-node Solr deployment with persistent storage. Suitable for development, testing, or small production environments.

A sample manifest is provided in samples/entando-solr-standalone.yaml.

Deploy

kubectl apply -f samples/entando-solr-standalone.yaml -n YOUR-NAMESPACE
kubectl get pods -l app=solr -w

Create the Entando core

kubectl port-forward service/solr 8983:8983

In another terminal:

curl "http://localhost:8983/solr/admin/cores?action=CREATE&name=entando&instanceDir=entando&dataDir=data"

Configure the EntandoApp

kubectl scale deploy/YOUR-APP-NAME-deployment --replicas=0 -n YOUR-NAMESPACE

Add the following environment variables to the deployment:

spec:
  containers:
    - env:
      - name: SOLR_ACTIVE
        value: "true"
      - name: SOLR_ADDRESS
        value: http://solr:8983/solr

Scale back up:

kubectl scale deploy/YOUR-APP-NAME-deployment --replicas=1 -n YOUR-NAMESPACE

SolrCloud deployment (Kubernetes / OpenShift)

A clustered Solr deployment with Zookeeper, suitable for production environments requiring high availability and replication.

A sample manifest is provided in samples/entando-solrCloud.yaml.

Prerequisites

• A working Entando instance
• Helm
• Cluster-level permissions for CRDs

Install Solr Operator

helm repo add apache-solr https://solr.apache.org/charts
helm repo update

kubectl create -f https://solr.apache.org/operator/downloads/crds/v0.5.0/all-with-dependencies.yaml
helm install solr-operator apache-solr/solr-operator --version 0.5.0

Deploy SolrCloud

Adjust resource settings (memory, CPU, storage, replicas) in the manifest as needed, then:

kubectl apply -f samples/entando-solrCloud.yaml -n YOUR-NAMESPACE
kubectl get solrclouds -w

Create collections

Set up port forwarding (leave it running):

kubectl port-forward service/solr-solrcloud-common 8983:80

In another terminal, create the Solr collection. The first collection (primary tenant, or the only one in single-tenant setups) must use entando as the collection name:

curl "http://localhost:8983/solr/admin/collections?action=CREATE&name=entando&numShards=1&replicationFactor=3&maxShardsPerNode=2"

Multitenancy

For multitenant environments, create an additional collection for each secondary tenant using the tenant name as the collection name:

curl "http://localhost:8983/solr/admin/collections?action=CREATE&name=YOUR-TENANT-NAME&numShards=1&replicationFactor=3&maxShardsPerNode=2"

Placeholder	Description
YOUR-TENANT-NAME	The identifying name of the tenant. In most cases, it will also be used to determine the base URL of the tenant (e.g., yoursite results in yoursite.your-domain.com).

The number of shards and shards per node should be adjusted for very large quantities of content (50k+). In such cases, adjustments to replicas and other resources may also be needed.

Configure the EntandoApp

kubectl scale deploy/YOUR-APP-NAME-deployment --replicas=0 -n YOUR-NAMESPACE

Add the following environment variables to the deployment:

spec:
  containers:
    - env:
      - name: SOLR_ACTIVE
        value: "true"
      - name: SOLR_ADDRESS
        value: http://solr-solrcloud-common/solr

Scale back up:

kubectl scale deploy/YOUR-APP-NAME-deployment --replicas=1 -n YOUR-NAMESPACE

Generate the Solr schema

The schema is generated automatically for the primary (entando) collection. For secondary tenant collections, trigger it manually: go to App Builder > Content > Solr Configuration and click Refresh next to each content type.

Then reindex: App Builder > Content > Settings > Reload the indexes.

Why not the previous image?

The previous entando/entando-solr:8 was based on solr:8.10.1 with OpenJDK 11.0.13, which lacks cgroup v2 support (added in JDK 11.0.16). On modern Linux kernels and OpenShift clusters using cgroup v2, the JVM fails to detect container CPU and memory limits, leading to resource over-allocation.

Rebasing on solr:8.11.4 (JDK 11.0.27) fixes this and also includes security patches (including the Log4Shell fix shipped in Solr 8.11.1).

License

This project is licensed under the Apache License 2.0 — see the LICENSE file for details.

Apache Solr is licensed under the Apache License 2.0. Eclipse Temurin is licensed under the GPL v2 with Classpath Exception.