How can I prepare an SVN dump to import into Beanstalk?

We recently removed the import from server directory option in our import tools due to the number of problems it was causing. Instead, we will rely solely on the svnadmin dump option since it is much more reliable and produces a faster import. To make this change as easy as possible for everyone, here are some details on creating a proper dump file from your current Subversion server. Please note that you cannot do this on your local working copy.

First, verify the repository

The first thing you want to do is make sure the repository is not corrupt. This will help avoid any problems after you upload the import into Beanstalk. To verify the repository login to your Subversion server and run the command:

$ svnadmin verify /usr/local/svn/repos

Where /usr/local/svn/repos is the path to your repository on your server.

Next, create the dump file

To create a dump file, login to your server and run the command:

$ svnadmin dump /usr/local/svn/repos > reponame.dump

Where /usr/local/svn/repos is the path to your repository and reponame.dump is the name of the dump file. More options and instructions are available in the Subversion book.

Lastly, Import the dump file into Beanstalk

The last step is to import the dump file into Beanstalk. You can do this when creating a new repository. Just follow the instructions in our creating a repository help article and you will be on your way.