For the past year or so, nearly every 45royale project has taken advantage of the many benefits of Subversion. Using Subversion (or SVN) gives us the ability to track revisions of our code for every single file on a project. That means we have record of the CSS or XHTML changes we make and if we accidentally delete an important line of code, we can easily revert to an earlier version of the file. Having version control is even more important on large projects where we have multiple developers working on the same files. Subversion allows us to track who is making updates and enables everyone to sync up with the latest revision of a file. However, depending on the number of people in your company and on your project, it may be difficult and time consuming at times to manage user permissions for every subversion repository. So we asked ourselves, is there a better way to manage our repositories?

After doing some initial research I came across a company, Wildbit, who is trying to bring Subversion to the masses. Their new web application, Beanstalk, is a hosted Subversion solution that enables you to setup online repositories and easily manage and track user access. Since they had a free plan that allows you to create one repository, I figured it was worth a shot to see if it could be the solution we were looking for.

In the beginning

Sign up for the service was pretty quick and painless and I was immediately given my own project site. The project site reminds me a lot of Basecamp so it felt familiar right away. The project site has a Dashboard tab where I can keep up to date with the latest activity, a Repository tab where I can browse my repo online, and a Users tab that allows me to easily add users and update their permissions. In addition to the ease of user management, probably the coolest thing about Beanstalk is that not only can I create a new repo from scratch using the online interface, but I can also import a repository from another server or even import a zip of a working copy from my local machine. How cool is that? This feature means it would be very easy to migrate over all of our existing repositories if the service turned out to work as expected. Since I already had a working copy on my local machine I decided to zip it up and use it to build a new Beanstalk repository.

Problems Arise

Despite my initial good vibes about Beanstalk, the moment I started to truly use the application I began having issues. My initial working copy imports all failed miserably, each one becoming stuck during the import process. At one point I had three imports visible on my dashboard which each one only 20% imported. In my mind I’m thinking, “I’m on the free plan now with only one repo, what will happen if all these attempts finally finish importing?” I never got to that point though because after several hours of waiting for the import process to complete, it became clear to me that they weren’t ever going to finish. Seriously, how long does it take to import a zip file that is under 1mb?

Finally I decided to delete each one and start fresh by creating a repo from scratch. While I was able to create the empty repository with ease, my initial attempts to connect to the repo via my SVN client failed. At this point I was extremely frustrated and I was very close to abandoning Beanstalk entirely, but something made me send an email to Beanstalk Support before I shut off my computer for the night and I’m very glad I did.

Beanstalk Support

To my surprise someone from Beanstalk had responded to my email later that night, apologizing for the problems I had experienced and asked me to login to their Campfire chat room the next morning so they could troubleshoot the issue with me. As luck would have it, I ended up being the only customer in the chat room that morning, so I was able to get quick one on one support. The Support technician was extremely patient and helpful, and we walked through several steps and were able to resolve all of my issues in under a half an hour. It turned out that I wasn’t doing anything wrong when I was trying to import my working copy, they had a experienced critical issue with their import service the previous day and they had elected to disable it to prevent further issues.

Lasting Impressions

Since my initial problems with the application, things have been going pretty smoothly. It’s been several weeks now and I’ve been able to access and make updates to my repo via the command line, my SVN client, and their online interface without incident. However, after my conversation with the Support Tech you can tell that the folks at Wildbit are still working out a lot of the kinks with Beanstalk. I’ve seen quite a few people report problems with the import service and I’ve heard rumors that the FTP tools for the paid plans are extremely slow at times. My guess is that Beanstalk started taking off and Wildbit is now scrambling to update their infrastructure to support the new influx of users. Until they stabilize the infrastructure, sporadic import problems and slow file transfers are bound to occur.

All in all though, I think the service is pretty sweet and it’s exactly the type of thing we’ve been looking for here at the shop. I’ve been sold on the benefits of using Subversion for our development projects for a while now, and a service like Beanstalk makes managing our repositories even easier. Despite my initial problems with app, their Support staff provided a great customer service experience and in the end I think that’s the main reason I decided to stick it out with them. I’m going to give Wildbit a little longer to work out some of the kinks before I upgrade to a paid plan, but I think they’ll eventually stabilize Beanstalk enough to where we can confidentially use it as our hosted subversion solution. If you’ve experienced issues with Beanstalk too and have since resolved them, let us know about it in the comments!