The Kubernetes installation utilises the upstream Alfresco Content Services Community Deployment Helm chart.
- Kubernetes cluster with Ingress
Several steps are required to get an aldica-enabled Kubernetes installation, namely:
- Building the aldica-enabled docker images
- Pushing the aldica-enabled docker images to a docker registry
- Fetching the upstream Helm chart.
- Patching the upstream Helm chart.
- Installing the Helm chart.
Building the aldica-enabled docker images can be done using the Dockerfiles within the docker sub-folder of the project root, namely the files:
- Dockerfile.repo
- Dockerfile.share
Note that building these images will take quite a while (~35 minutes), depending on available CPU, IO performance and network resources.
Building can be done either directly using docker build .. -f Dockerfile.[repo|share] or indirectly using docker-compose build [alfresco|share].
After this step, it's assumed that the reader has build the docker images, and obtained the IDs for the images.
We assume that a registry (myregistry) has already been configured.
Now we tag the images produced in the previous step;
docker tag {{ repo_id }} {{ myregistry }}/aldica-repo:latest
docker tag {{ share_id }} {{ myregistry }}/aldica-share:latest
And push them to {{ myregistry }}:
docker push {{ myregistry }}/aldica-repo:latest
docker push {{ myregistry }}/aldica-share:latest
Fetch the latest acs-community-deployment release archive.
Extract the archive and navigate to root folder (the folder containing the LICENSE).
This is the folder we want to work with in the next step.
Find the provided patch file, and apply it to the Helm chart folder, by running:
patch -p1 -i /path/to/this/repo/docs/0001-Aldica.patch
At this point, you should insert your own docker images in the values.yml file.
Note: You may have to configure custom registry credentials, assuming your docker registry is not publicly available.
At this point, the Helm chart can be installed according to the documentation in the README.md file found within the Helm chart folder, or according to the documentation on the acs-community-deployment repository.
Currently, the Helm chart only works with Kubernetes version 1.15, so adjustments will be needed for later versions of Kubernetes. Below, an example using Minikube and Kubernetes version 1.15.0 is given.
This example shows how to get an aldica cluster up and running in Kubernetes by using Minikube.
In the following, it will be assumed that Minikube has already been installed using the instructions given here and also that Helm has been installed. See the installation instructions for Helm here.
Start a new Minikube cluster running Kubernetes 1.15.0 in e.g. VirtualBox with an appropriate amount of CPUs and memomy (if an existing Minikube cluster is already running you may need to delete this first):
$ minikube start --driver=virtualbox --cpus=6 --memory=24000 --kubernetes-version=v1.15.0
Run a few kubectl maintenance commands as described in the upstream
acs-community-deployment documentation:
$ kubectl create serviceaccount --namespace kube-system tiller
$ kubectl create clusterrolebinding tiller-cluster-rule --clusterrole=cluster-admin --serviceaccount=kube-system:tiller
$ helm init --service-account tiller
$ helm repo update
$ helm repo list
$ helm install stable/nginx-ingress --set rbac.create=true
Install the required dependencies with Helm:
$ cd /path/to/acs-community-deployment/helm/alfresco-content-services-community
$ helm dependencies update
The Helm chart can be installed with:
$ helm install .
A number of pods and services are now fired up. These can be inspected with:
$ kubectl get pods
$ kubectl get services
which should result in output similar to the below once everything is up and running.
$ kubectl get pods
NAME READY STATUS RESTARTS AGE
melting-alpaca-activemq-6f7bb7d865-xwjw2 1/1 Running 0 94m
melting-alpaca-alfresco-cs-ce-imagemagick-6679c56fd-8v5nr 1/1 Running 0 94m
melting-alpaca-alfresco-cs-ce-libreoffice-86cf4866fc-889z9 1/1 Running 0 94m
melting-alpaca-alfresco-cs-ce-pdfrenderer-585c658875-4mspc 1/1 Running 0 94m
melting-alpaca-alfresco-cs-ce-repository-0 1/1 Running 0 94m
melting-alpaca-alfresco-cs-ce-repository-1 1/1 Running 0 89m
melting-alpaca-alfresco-cs-ce-share-67c644d9cd-tkstc 1/1 Running 1 94m
melting-alpaca-alfresco-cs-ce-tika-5996946446-hwf9v 1/1 Running 0 94m
melting-alpaca-alfresco-cs-ce-transform-misc-5fb6cfb65d-dkmrv 1/1 Running 0 94m
melting-alpaca-alfresco-search-solr-5b56488db9-lcvbg 1/1 Running 0 94m
melting-alpaca-keycl-0 1/1 Running 0 94m
melting-alpaca-nginx-ingress-controller-6bbb7dff44-pp8w4 1/1 Running 0 94m
melting-alpaca-nginx-ingress-default-backend-69776bf6dd-ndhdp 1/1 Running 0 94m
melting-alpaca-postgresql-acs-99776fd5-mb5pj 1/1 Running 0 94m
melting-alpaca-postgresql-id-6b799ddffb-m2d7d 1/1 Running 0 94m
mewing-guppy-nginx-ingress-controller-74b9c9bbdb-v6fmq 1/1 Running 0 95m
mewing-guppy-nginx-ingress-default-backend-7656bd9457-rg9tf 1/1 Running 0 95m
$ kubectl get services
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 110m
melting-alpaca-activemq-broker ClusterIP 10.102.144.199 <none> 61613/TCP,61616/TCP,5672/TCP 95m
melting-alpaca-activemq-web-console NodePort 10.102.224.82 <none> 8161:32087/TCP 95m
melting-alpaca-alfresco-cs-ce-imagemagick ClusterIP 10.111.86.218 <none> 80/TCP 95m
melting-alpaca-alfresco-cs-ce-libreoffice ClusterIP 10.110.57.194 <none> 80/TCP 95m
melting-alpaca-alfresco-cs-ce-pdfrenderer ClusterIP 10.111.189.23 <none> 80/TCP 95m
melting-alpaca-alfresco-cs-ce-repository ClusterIP 10.96.152.212 <none> 80/TCP 95m
melting-alpaca-alfresco-cs-ce-repository-headless ClusterIP None <none> 80/TCP 95m
melting-alpaca-alfresco-cs-ce-share ClusterIP 10.99.175.101 <none> 80/TCP 95m
melting-alpaca-alfresco-cs-ce-tika ClusterIP 10.104.61.91 <none> 80/TCP 95m
melting-alpaca-alfresco-cs-ce-transform-misc ClusterIP 10.106.58.79 <none> 80/TCP 95m
melting-alpaca-alfresco-search-solr ClusterIP 10.102.37.178 <none> 80/TCP 95m
melting-alpaca-keycl-headless ClusterIP None <none> 80/TCP 95m
melting-alpaca-keycl-http ClusterIP 10.98.231.253 <none> 80/TCP 95m
melting-alpaca-nginx-ingress-controller LoadBalancer 10.103.219.102 <pending> 80:31543/TCP,443:31676/TCP 95m
melting-alpaca-nginx-ingress-default-backend ClusterIP 10.103.93.115 <none> 80/TCP 95m
melting-alpaca-postgresql-acs ClusterIP 10.109.115.2 <none> 5432/TCP 95m
melting-alpaca-postgresql-id ClusterIP 10.107.101.58 <none> 5432/TCP 95m
mewing-guppy-nginx-ingress-controller LoadBalancer 10.103.234.4 <pending> 80:32328/TCP,443:31813/TCP 96m
mewing-guppy-nginx-ingress-default-backend ClusterIP 10.103.208.191 <none> 80/TCP 96m