Created analytics on a stick (markdown)

wiki/analytics-on-a-stick.md

@@ -0,0 +1,39 @@
+# Sailing Analytics on a stick
+
+This is a package to enable running the SAP Sailing Analytics from a stick.
+It is provided as-is without warranty. The most common usecase is to have a laptop 
+running in a loung that replicates data from a master server that runs in the cloud.
+
+## Create a stick
+
+- Copy the contents of $GIT/stick to the root of your USB stick
+- Create a directory named `server` and in there `logs` and `tmp`
+- Extract the file http://releases.sapsailing.com/stick-binaries-1.0.tar.gz to the root of your USB stick
+- Install a release by executing `scripts/refreshInstance.sh install-release <release>`
+
+## Configuration
+
+- (Windows) Copy templates/env.bat.template to the root of the USB stick and rename it to
+env.bat. Now open it with your favourite text editor and adapt the values to your needs.
+As Windows does not handle data writes on a stick well it is recommended that you also
+edit mongodb-windows.cfg and set the dbpath to an existing directory on your hard disk.
+
+- (Linux) Copy templates/env.sh.template to the root of your USB stick and rename it to
+env.sh. Fire up a vi and adapt the values. It can be wise to also have a look at the conf
+for the database that is located in mongodb.cfg. You can specify a separate directory
+on your harddisk or a ramdisk for better database performance.
+
+If you have configured the server to replicate then you absolutely need to make sure
+that the server version on the stick corresponds to the one that is running in the cloud.
+
+## Starting
+
+- (Windows) First you need to start the database by executing start-mongo.bat. Wait at least
+5 minutes until MongoDB has created all necessary files. Before the database hasn't been
+initialized you won't be able to start the Sailing Analytics server. After the database
+has started up you can start up the analytics server by running start.bat. Watch out for any 
+error messages that might come up. After the server has been started you should be able to
+reach it by using http://127.0.0.1:8888/.
+
+- (Linux) All you need is to start the whole thing by executing start.sh. This will take care
+of starting up the database and then starting the analytics server.