AllegroGraph version 7.0.4 Docker image, Kubernetes and Helm features extended

README-original.md

@@ -0,0 +1,17 @@
+# Welcome to AllegroGraph examples!
+
+This repository contains examples of using AllegroGraph in advanced scenarios.
+
+## Featured examples
+
+* Configuration and setup examples
+  * [Clustering](clustering)
+    * [Load balancing multi-master replication cluster on AWS using Terraform](clustering/terraform-elb/using-terraform.md)
+    * [AllegroGraph Replication using NGINX Load Balancer](clustering/misc/using-nginx-load-balancer.md)
+    * [MMR Replication Clusters in Kubernetes and Docker Swarm](clustering/kubernetes/mmr/kubernetes-mmr.md)
+* Data examples
+  * [Bitcoin RDF model](data/bitcoin)
+
+* Generate AllegroGraph databases
+  * [Weather data](generators/weather/weather.md)
+

README.md

@@ -1,17 +1,96 @@
-# Welcome to AllegroGraph examples!
+# Development/Release Notes
 
-This repository contains examples of using AllegroGraph in advanced scenarios.
+Author:     Robert Wydler </br>
+LinkedIn:   [Profile](https://www.linkedin.com/in/robwyd/) 
 
-## Featured examples
+## Introduction 
+### Based on ###
+ This is an extension of the product "AllegroGraph for Multi-Master Replication" and is based on https://github.com/franzinc/agraph-examples.
 
-* Configuration and setup examples
-  * [Clustering](clustering)
-    * [Load balancing multi-master replication cluster on AWS using Terraform](clustering/terraform-elb/using-terraform.md)
-    * [AllegroGraph Replication using NGINX Load Balancer](clustering/misc/using-nginx-load-balancer.md)
-    * [MMR Replication Clusters in Kubernetes and Docker Swarm](clustering/kubernetes/mmr/kubernetes-mmr.md)
-* Data examples
-  * [Bitcoin RDF model](data/bitcoin)
+### Goals ###
+1. Creating a docker image like [Docker and AllegroGraph 7.0.4](https://franz.com/agraph/support/documentation/current/docker.html) but for [Multi-Master Replication in AllegroGraph 7.0.4](https://franz.com/agraph/support/documentation/current/multi-master.html) based on [Welcome to AllegroGraph examples](https://github.com/franzinc/agraph-examples).
+
+1. Run standardized Multi-master Replication in AllegroGraph 7.0.4 using [Google Kubernetes Engine, GKE](https://cloud.google.com/kubernetes-engine).
+
+1. Make it available for semantic AI research studies with help of [Google's Free Tier ](https://cloud.google.com/free) and [AllegroGraph's Free Editon ](https://allegrograph.com/) and to learn more about both products.
+
+1. Setup a Kubernetes namespace `agraph-ns`,  Kubernetes `Update-Strategy` and the `Horizontal Pod AutoScaler, HPA` e.g. using `Google's Cluster Autoscaler`. 
+
+
+### Affected Parts ###
+
+Therefore, only the clustering parts of **docker** (folders: ``clustering/kubernetes/mmr/agmmr/*``), **kubernetes** and **helm** (folders: ``clustering/kubernetes/mmr/helm/*``) has been changed. 
+
+## Important Constraints !!! ##
+Currently ...
+* ... it's only a **developer study** version and it's **not ready for production!** 
+* ... the prototype was only running in a **very small (free)** Google GKE **test environment**.
+* ... **fundamental changes** had been made (e.g. from ``StateFull`` copy read only ``pods`` to ``Deployment`` copy read only ``pods``) and Kubernetes' ``Horizontal Pod Autoscaler`` had been defined added (``hpa.yaml``).
+
+## Known Issue ##
+After deployment, sometimes it may happen, that the repositories are not reachable by the `agraph-mmr-copy-loadbalancer`. That's because the `agraph-mmr-controlling` stateful set has to run   before the `agraph-mmr-copy` deployment. I haven't figured out yet, how to arrange a sequential start in Helm. 
+
+You can **circumvent** the issue by deleting the `agraph-mmr-copy` pods in the Google Cloud Console or you can perform a graceful pod deletion with the following command:
+
+`kubectl delete pods <pod>`
+
+The `agraph-mmr-copy` pods will be recreated automatically by the deploment.  
+
+---
+## ToDo's for a Reliable Release
+1. Change Helm to always start the stateful set `agraph-mmr-controlling` before the  `agraph-mmr-copy` deployment.  
+
+For a reliable release of the Docker Image 7.0.4 together with Kubernetes and Helm, it's important to validate and test ...
+1. ... the functionality of `copy-ssl.yaml` pods after the change from `StatefulSet` to Kubernetes `Deployment` pods.
+1. ... the reliability and horizontal scalability of the pods (`hpa.yaml`) and of the Kubernetes cluster (e.g. GKE) in an operational environment with high query loads.
+1. ... the interaction of the state-oriented controller pods with the stateless copy pods.
+1. ... etc.
+---
+## Overview of Changes 
+### Affected Parts ###
+Only the clustering parts of **docker** (folders under: ``clustering/kubernetes/mmr/agmmr/*``), **kubernetes** and **helm** (folders under: ``clustering/kubernetes/mmr/helm/*``) were changed. 
+
+### Docker Image, Version 7.0.4
+*.../mmr/agmmr/Dockerfile:*
+* Changed to AllegroGraph version to 7.0.4
+* Testrelease removed
+* Changed access to `chmod +x /app/misc/runrepl.sh`
+
+*.../mmr/agmmr/Makefile:*
+* Changed to AllegroGraph version to 7.0.4
+* Testrelease removed
+
+### Helm 
+*.../mmr/helm/Makefile:*
+* Given application name `agraph-mmr`
+* Kubenretes namespace added `agraph-ns` for all components
+* Changed to AllegroGraph version to 7.0.4
+* Changed to AllegroGraph app version to 7.0.4
+
+*.../mmr/helm/agraphmmr/Chart.yaml:*
+* Changed to AllegroGraph version to 7.0.4
+* Kept the maintainers to "Franz Inc."
+
+*.../mmr/helm/agraphmmr/values.yaml:*
+* CPU limits added
+* Update behaviour added
+* Horizontal Pod Autoscaler criteria added
+
+### Kubernetes templates
+*.../mmr/helm/agraphmmr/templates/controlling-ss.yaml:*
+* Required `selector`added
+* Update strategy added
+*  CPU limits for the Horizontal Pod Autoscaler, HPA added 
+
+*.../mmr/helm/agraphmmr/templates/copy-ss.yaml:*
+* Because of read-only, stateless pods: Changed the  `kind` from `StatefulSet` to `Deploment`
+* Constant `replica` set to `1`; it's under HPA control
+* `matchLabels` and `app`added
+* Update strategy added
+* Values added under `Resources` for HPA
+
+*.../mmr/helm/agraphmmr/templates/hpa.yaml:*
+New file added for Horizontal Pod Autoscaler, HPA concerning ... 
+* the controllers (controlling-ss.yaml)
+* the workers (copy-ss.yaml) 
 
-* Generate AllegroGraph databases
-  * [Weather data](generators/weather/weather.md)
-

clustering/kubernetes/mmr/agmmr/Dockerfile

@@ -8,17 +8,16 @@ FROM centos:7
 # ensure we have a good set of network programs installed
 RUN yum -y install net-tools iputils bind-utils wget hostname openssl
 
-ARG agversion=7.0.0
-ARG testrelease=
+ARG agversion=7.0.4  
 
 ENV agdistfile=agraph-${agversion}-linuxamd64.64.tar.gz
 ENV agversion=${agversion}
 
 ## Show the curl command we're about to execute
 ## This useful for debugging curl problems due to invalid arguments
-RUN echo will do: curl -f -o ${agdistfile} http://franz.com/ftp/pri/acl/ag/ag${agversion}${testrelease}/linuxamd64.64/${agdistfile}
+RUN echo will do: curl -f -o ${agdistfile} https://franz.com/ftp/pri/acl/ag/ag${agversion}/linuxamd64.64/${agdistfile}
 
-RUN curl -f -o ${agdistfile} http://franz.com/ftp/pri/acl/ag/ag${agversion}${testrelease}/linuxamd64.64/${agdistfile}
+RUN curl -f -o ${agdistfile} https://franz.com/ftp/pri/acl/ag/ag${agversion}/linuxamd64.64/${agdistfile}
 
 ## extract the AllegroGraph distribution
 RUN tar xfz ${agdistfile}
@@ -40,5 +39,5 @@ COPY . /app/misc
 # in the controlling instance only.
 VOLUME ["/app/rootcatalog"]
 
-
+RUN chmod +x /app/misc/runrepl.sh
 ENTRYPOINT ["/app/misc/runrepl.sh"]

clustering/kubernetes/mmr/agmmr/Makefile

@@ -4,13 +4,12 @@
 #
 # e.g.  make agversion=6.5.0 build
 #
-dockeraccount=unspecified
-agversion=7.0.0
-testrelease=
-tag=7.0.0
+dockeraccount = unspecified
+agversion=7.0.4 
+tag=7.0.4
 
 build:
-	docker build --build-arg agversion=$(agversion) --build-arg testrelease=$(testrelease) -t  $(dockeraccount)/agmmr:$(tag) .
+	docker build -t $(dockeraccount)/agmmr:$(tag) --build-arg agversion=$(agversion) --build-arg $(dockeraccount)/agmmr:$(tag) .
 
 push:
 	docker image push $(dockeraccount)/agmmr:$(tag)

clustering/kubernetes/mmr/agmmr/helm/Makefile

@@ -0,0 +1,49 @@
+# running agraph mmr in kubernetes using helm
+#
+# type
+#  % make
+# to see the steps you must perform to install the cluster
+#
+
+name = agraph-mmr
+namespace = agraph-ns
+info:
+	@echo
+	@echo You can install the mmr cluster in kubernetes
+	@echo using a generated release name
+	@echo     % make install
+	@echo
+	@echo or if you wish to choose a name for the cluster:
+	@echo     % make name=foo install
+	@echo
+	@echo 'make install' installs AllegroGraph with no license
+	@echo so its capabilities are limited.
+	@echo If you have a license in file ~/license.txt then
+	@echo use make target  install-with-license
+	@echo for example
+	@echo    % make file=~/license.txt name=foo install-with-license
+	@echo
+
+
+install:
+	helm install $(name) ./agraphmmr --create-namespace --namespace $(namespace)
+
+# rwy added:
+re-install:
+	helm upgrade --install $(name) ./agraphmmr --namespace $(namespace)
+
+
+install-with-license:
+	helm install --set-file license=$(file) $(name) ./agraphmmr --namespace $(namespace)
+
+
+# rwy added:
+uninstall:
+	helm uninstall $(name) --namespace $(namespace)
+
+## make a Helm repository with the agraphmmr chart in it
+repository:
+	helm package ./agraphmmr
+	mkdir -p repo
+	mv agraphmmr-*.tgz repo
+	helm repo index repo

clustering/kubernetes/mmr/agmmr/helm/agraphmmr/.helmignore

@@ -0,0 +1,23 @@
+# Patterns to ignore when building packages.
+# This supports shell glob matching, relative path matching, and
+# negation (prefixed with !). Only one pattern per line.
+.DS_Store
+# Common VCS dirs
+.git/
+.gitignore
+.bzr/
+.bzrignore
+.hg/
+.hgignore
+.svn/
+# Common backup files
+*.swp
+*.bak
+*.tmp
+*~
+# Various IDEs
+.project
+.idea/
+*.tmproj
+.vscode/
+.#*

clustering/kubernetes/mmr/agmmr/helm/agraphmmr/Chart.yaml

@@ -0,0 +1,23 @@
+apiVersion: v2
+name: agraphmmr
+description: An AllegroGraph MMR cluster Helm chart for Kubernetes
+
+type: application
+home: https://github.com/franzinc/agraph-examples/tree/master/clustering/kubernetes/mmr
+
+# This is the chart version. This version number should be incremented each time you make changes
+# to the chart and its templates, including the app version.
+version: 7.0.4
+
+# This is the version number of the application being deployed. This version number should be
+# incremented each time you make changes to the application.
+appVersion: 7.0.4
+
+icon: https://franz.com/images/logo.png
+
+
+maintainers:
+ - name: "Franz Inc"
+   email: "[email protected]"
+
+

clustering/kubernetes/mmr/agmmr/helm/agraphmmr/templates/NOTES.txt

@@ -0,0 +1,64 @@
+ {{ .Chart.Name }} -  Helm chart for running an AllegroGraph cluster with Multi-Master Replication
+
+The full details on this chart can be found at
+
+  https://github.com/franzinc/agraph-examples/blob/master/clustering/kubernetes/mmr/kubernetes-mmr.md
+
+That contains a description of how the chart works as well as how to modify what it does.
+
+
+
+Once the chart is installed you can access the controlling instance via a load balancer
+
+Run
+
+% kubectl get svc
+
+To find the address of the controlling-loadbalancer.
+
+It will take a few minutes for the external IP address to be allocated and
+show by this kubectl command.
+
+For example:
+
+NAME                            TYPE           CLUSTER-IP    EXTERNAL-IP      PORT(S)           AGE
+sample-controlling-loadbalancer LoadBalancer   10.12.8.217   104.155.148.58   10035:32240/TCP   113s
+
+Given the above you can access the controlling instance through the load balancer
+at:
+
+http://104.155.148.58:10035
+
+The default username is "test" and the default password is "xyzzy"
+
+You can do read and write access of the MMR cluster through the controlling-loadbalancer.
+
+There is also a copy-loadbalancer.  This distributes work to all the copy nodes.
+The load balancer does not yet have session support so you should only
+use the copy-loadbalancer for queries to the repository.
+
+We also have a created a copy-0-loadbalancer as a demonstration of how to
+communicate with a single specified copy instance.  Since this load
+balancer is always talking to the same instance one can do read and write accesses through
+this load balancer.   However the copy instances can be deleted by
+kubernetes if you reduce the number of replicas defined.
+
+One way to change the number of replicas is to edit values.yaml
+and specify the desired number of replicas and then
+
+% helm upgrade releasename chart
+
+where releasename and chart are the same as
+in the command you used to install the chart
+
+% helm install releasename chart
+
+
+
+
+
+
+
+
+
+

clustering/kubernetes/mmr/agmmr/helm/agraphmmr/templates/controlling-lb.yaml

@@ -0,0 +1,19 @@
+#
+# This will create a service with an external IP through which
+# one can access the controlling instance.
+# Since there's just one controlling instance we don't need to worry
+# about the fact that the load balancer has no support for sessions.
+#
+# The external IP will be chosen by the cloud provider
+#
+apiVersion: v1
+kind: Service
+metadata:
+  name: {{ .Release.Name }}-controlling-loadbalancer
+spec:
+  type: LoadBalancer
+  ports:
+  - port: {{ .Values.port }}
+    targetPort: {{ .Values.port }}
+  selector:
+    app: {{ .Release.Name }}-controlling

clustering/kubernetes/mmr/agmmr/helm/agraphmmr/templates/controlling-service.yaml

@@ -0,0 +1,23 @@
+#
+# this service exposes the internal IP address
+# of the controlling instance and makes it
+# available via dns at the
+# name 'releasename-controlling'
+# using the dns search list
+#
+apiVersion: v1
+kind: Service
+metadata:
+ name: {{ .Release.Name }}-controlling
+spec:
+ clusterIP:  None
+ selector:
+   app: {{ .Release.Name }}-controlling
+ ports:
+ - name: http
+   port: {{ .Values.port }}
+   targetPort: {{ .Values.port }}
+
+
+
+