Add a Julia enabled Jupyter Notebook cloud shell tutorial

cloud-console-tutorials/compute_julia_jupyter_notebook/createserver.sh

@@ -0,0 +1,31 @@
+#!/usr/bin/env bash
+
+PROJECT=${1:-$(gcloud config get-value core/project)}
+SERVERNAME=${2:-julia-nb-server}
+ZONE=${3:-$(gcloud config get-value compute/zone)}
+SOURCECIDR=${4:-"0.0.0.0/0"}
+
+gcloud beta compute --project=$PROJECT instances create $SERVERNAME \
+    --zone=$ZONE \
+    --machine-type=n1-standard-2 \
+    --subnet=default \
+    --network-tier=PREMIUM \
+    --maintenance-policy=MIGRATE \
+    --scopes=https://www.googleapis.com/auth/devstorage.read_only,https://www.googleapis.com/auth/logging.write,https://www.googleapis.com/auth/monitoring.write,https://www.googleapis.com/auth/servicecontrol,https://www.googleapis.com/auth/service.management.readonly,https://www.googleapis.com/auth/trace.append \
+    --tags=jupyter-server \
+    --image=debian-9-drawfork-v20180810 \
+    --image-project=eip-images \
+    --boot-disk-size=256GB \
+    --boot-disk-type=pd-standard \
+    --boot-disk-device-name=$SERVERNAME \
+    --metadata-from-file=startup-script=$PWD/startup.sh
+
+gcloud compute --project=$PROJECT firewall-rules create default-allow-jupyter \
+    --direction=INGRESS \
+    --priority=1000 \
+    --network=default \
+    --action=ALLOW \
+    --rules=tcp:8089\
+    --source-ranges=$SOURCECIDR \
+    --target-tags=jupyter-server
+

cloud-console-tutorials/compute_julia_jupyter_notebook/startup.sh

@@ -0,0 +1,90 @@
+#
+# Install required packages
+#
+apt update -y
+apt install git bzip2 libgl1-mesa-glx -y
+
+#
+# Create a jupyter user and switch to that context
+#
+adduser jupyter --uid 2001 --disabled-password --gecos "" --quiet
+
+#
+# Install Anaconda
+#
+cd /tmp
+wget https://repo.anaconda.com/archive/Anaconda3-5.2.0-Linux-x86_64.sh
+bash Anaconda3-5.2.0-Linux-x86_64.sh -b -p /opt/anaconda3
+
+#
+# Create a Jupyter configuration or the jupyter user
+#
+cd /home/jupyter
+mkdir .jupyter
+git clone https://github.com/JuliaComputing/JuliaBoxTutorials.git notebooks
+
+openssl req -x509 -nodes -days 365 -newkey rsa:2048 -keyout nbkey.key -out nbcert.pem --batch
+
+cat <<JUPYTER_CONFIG > /home/jupyter/.jupyter/jupyter_notebook_config.py
+c.NotebookApp.certfile = u'/home/jupyter/nbcert.pem'
+c.NotebookApp.keyfile = u'/home/jupyter/nbkey.key'
+c.NotebookApp.allow_origin = '*'
+c.NotebookApp.ip = '*'
+c.NotebookApp.port = 8089
+c.NotebookApp.password = u'HASHED_PASSWD'
+JUPYTER_CONFIG
+
+chown jupyter -R /home/jupyter
+
+#
+# Install Julia
+#
+cd /opt
+mkdir julia
+cd julia
+wget https://julialang-s3.julialang.org/bin/linux/x64/1.0/julia-1.0.0-linux-x86_64.tar.gz
+tar zxf julia-1.0.0-linux-x86_64.tar.gz
+rm julia-1.0.0-linux-x86_64.tar.gz
+ln -s /opt/julia/julia-1.0.0/bin/julia /usr/local/bin
+
+#
+# Add the Julia kernel to the Jupyter setup
+#
+cat <<IJULIA > /tmp/ijulia.jl
+using Pkg
+Pkg.add("IJulia")
+IJULIA
+
+chown jupyter /tmp/ijulia.jl
+su - jupyter -c 'julia /tmp/ijulia.jl'
+rm /tmp/ijulia.jl
+
+#
+# Start the notebook server
+#
+cat <<JUPYTER_SERVICE > /lib/systemd/system/jupyter.service
+[Unit]
+Description=Jupyter notebook server
+After=network.target
+
+[Service]
+ExecStart=/opt/anaconda3/bin/jupyter notebook \
+    --config=/home/jupyter/.jupyter/jupyter_notebook_config.py \
+    --no-browser \
+    --port=8089 \
+    --notebook-dir=/home/jupyter/notebooks
+
+User=jupyter
+Group=jupyter
+
+Restart=no
+PrivateTmp=true
+ProtectSystem=true
+
+[Install]
+WantedBy=multi-user.target
+JUPYTER_SERVICE
+
+systemctl enable juypter.service
+systemctl daemon-reload
+systemctl restart jupyter.service

cloud-console-tutorials/compute_julia_jupyter_notebook/tutorial.md

@@ -0,0 +1,103 @@
+# Run a private Julia Jupyter Notebook server on Google Cloud Platform
+
+This tutorial shows you how to set up a Jupyter Notebook server with a 
+[Julia](https://julialang.org/) kernel installed in a Google Cloud Platform project.
+
+[![button](http://gstatic.com/cloudssh/images/open-btn.png)](https://console.cloud.google.com/cloudshell/open?git_repo=https://github.com/wardharold/julia-notebook&page=editor&tutorial=tutorial.md)
+
+If you want to create a new project for your notebook server proceed to 
+[Task 0](#task-0-optional-create-a-project-with-a-billing-account-attached). 
+Otherwise you can skip that task and use and existing project.
+
+## Task 0 (OPTIONAL) Create a project with a billing account attached 
+**(you can also use an existing project and skip to the next step)**
+```sh
+ORG=[YOUR_ORG]
+BILLING_ACCOUNT=[YOUR_BILLING_ACCOUNT_NAME]
+ZONE=[COMPUTE ZONE YOU WANT TO USE]
+gcloud projects create $PROJECT" --organization=$ORG
+gcloud beta billing projects link $PROJECT --billing-account=$(gcloud beta billing accounts list | grep $BILLING_ACCOUNT | awk '{print $1}')
+gcloud config configurations create -- activate $PROJECT
+gcloud config set compute/zone $ZONE
+```
+
+## Task 1 Set up your environment variables 
+Ensure you are working with the project you want to use to run your notebook server.
+
+Set the environment variables for your project.
+```sh
+PROJECT=$(gcloud config get-value core/project)
+ZONE=$(gcloud config get-value compute/zone)
+```
+
+Choose a name for your server and set the SERVERNAME envrionment variable.
+```sh
+SERVERNAME=your-preferred-servername
+```
+
+Make sure none of these values is empty
+```sh
+echo $PROJECT $SERVERNAME $ZONE
+```
+
+Go to the browser you will use to connect to your notebook server and browse to [Your Address As a Service](https://v4.ident.me). Copy the resulting IP address.
+
+```sh
+SOURCECIDR=your-browsers-ip-address
+```
+
+## Task 2 Set up a password for your notebook server
+The ```startup.sh``` script creates a Jupyter Notebook configuration on your notebook server.
+Part of that configuration is a hashed password. In this task you will use the Jupyter python
+library to create a hashed password for your notebook server.
+
+Install the jupyter python modules
+```sh
+pip3 install jupyter
+```
+
+Generate a hashed password.
+```
+PASSWD=your-preferred-password
+HASHED_PASSWD=$(python3 -c "from notebook.auth import passwd; print(passwd(\"${PASSWD}\"))")
+```
+
+Use your password in the Jupyter Notebook configuration set up by the startup.sh script.
+```sh
+sed -i 's/HASHED_PASSWD/'"${HASHED_PASSWD}"'/' startup.sh
+```
+
+## Task 3 Create your notebook server
+Run the ```createserver.sh``` script to setup your notebook server.
+```sh
+. ./createserver.sh
+```
+
+The script creates a Google Cloud Platform compute instance. The instance's name is the value you gave
+the ```SERVERNAME``` environment variable in [Task 1](#task-1-set-up-your-environment-variables). The 
+compute instance executes the ```startup.sh``` script after it boots and does the following:
+
+* Installs Jupyter via the [Anaconda Distribution](https://www.anaconda.com/download/#linux)
+* Creates a jupyter user to run the notebook server
+* Installs the [Julia](https://julialang.org/) kernel in the Jupyter notebook server
+* Clones some [Julia notebooks](https://github.com/JuliaComputing/JuliaBoxTutorials)
+* Starts the Jupyter Notebook server
+
+Installing the Julia kernel is a time consuming process. Your notebook server will not be ready
+for five to ten minutes.
+
+## Task 4 Connect to your notebook server
+Determine the public IP address of your notebook server.
+```sh
+gcloud compute instances list --format="value(name, networkInterfaces.accessConfigs[0].natIP)" --filter="$SERVERNAME" | awk '{print $}'
+```
+
+Your notebook server runs on port 8089 and is configured to require TLS (https). The ```startup.sh``` script creates a self-signed certificate to secure communications with your notebook server.
+
+In your browser navigate to port 8089 of your notebook server's public IP address using https, e.g.,
+https://1.2.3.4:8089. Because your notebook server uses a self-signed certificate your browser will
+complain that it doesn't recognize the certificate. You can safely bypass those warnings and
+connect to your notebook server.
+
+Log into your notebook server using the password you created in [Task 2](#task-2-set-up-a-password-for-your-notebook-server). 
+Use any of the pre-installed notebooks to learn Julia or create your own.