Use Jupyter Server Proxy for TensorBoard

- https://github.com/jupyter/docker-stacks/pull/2126

LICENSE

@@ -26,6 +26,7 @@ their copyrights and license statements are listed below.
 
 --------------------------------------------------------------------------------
 base/conf/jupyter/etc/jupyter/jupyter_server_config.py
+base/scripts/usr/local/bin/before-notebook.d/71-tensorboard.sh
 base/scripts/usr/local/bin/run-hooks.sh
 base/scripts/usr/local/bin/start-notebook.sh
 base/scripts/usr/local/bin/start-singleuser.sh

NOTES.md

@@ -36,6 +36,8 @@ The following startup hooks are put in place:
   to create user's projects and workspaces folder.
 * [/usr/local/bin/before-notebook.d/30-code-server.sh](base/scripts/usr/local/bin/before-notebook.d/30-code-server.sh)
   to update code-server settings.
+* [/usr/local/bin/before-notebook.d/71-tensorboard.sh](base/scripts/usr/local/bin/before-notebook.d/71-tensorboard.sh)
+  to use Jupyter Server Proxy for TensorBoard.
 * [/usr/local/bin/before-notebook.d/90-limits.sh](base/scripts/usr/local/bin/before-notebook.d/90-limits.sh)
   * *soft* limit the *address space* based on the amount of *physical* and
     *virtual memory* of the host. (default: command `ulimit -v`)

base/scripts/usr/local/bin/before-notebook.d/71-tensorboard.sh

+#!/bin/bash
+# Copyright (c) Jupyter Development Team.
+# Distributed under the terms of the Modified BSD License.
+
+set -e
+
+# Initialize the TENSORBOARD_PROXY_URL with the appropriate path
+# to use jupyter-server-proxy.
+
+export TENSORBOARD_PROXY_URL="${JUPYTERHUB_SERVICE_PREFIX:-/}proxy/%PORT%/"