Synchronizing App_Data with git

When you have a web application that stores data in it's App_Data folder, at some point you'll want to synchronize the development environment(s) with the live environment. This way your development machine will have the latest live data. This is a task that can be achieved with git.

Requirements

I have actually made the following work, but with the following adjustments:

  • All involved Windows machines (development and live boxes) have Cygwin installed
  • Windows live machine has Cygwin’s open ssh server running

Step 1 – Convert live App_Data into a git repository

The first step is to convert the App_Data on the live server into a git repository. This will be the source from which all other machines will fetch data from.

To do that, a simple way is to create a new empty git repository, add all existing files, commit and replace the old App_Data folder with the new git repository. Assuming you have a unix (Cygwin) shell, you can write something like:

cd /c/web/myapp
mkdir NewAppData
cd NewAppData
git init .
cp -R ../App_Data/* .
git add *
git commit -m "Copying existing App_Data"
cd ..
mv App_Data App_Data.backup
mv NewAppData App_Data

At this point the App_Data folder is now a git repository containing all the existing data.

Step 2 – Convert development App_Data into a git repository

The second step is to go to the development environment and fetch the App_Data from the production server. Something like:

cd /c/projects/myapp
mv App_Data App_Data.backup
git clone ssh://my.live.server/c/web/myapp/App_Data App_Data

The local development machine now has a folder C:projectsmyappApp_Data that contains all the live data.

Step 3 – Keeping up to date

As the live server and the development machine continue to work on their App_Data folders, they will reach again a point that they need synchronization. To do that, first you need to commit the changes on the server:

git add *
git commit -m "Blindly adding all changes"

Next, on the development machine you fetch the changes:

git pull

That's it.

A bonus step can be automation. By creating scheduled tasks for the commit and pull steps you can stay in sync automatically.

Hope this helps.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s