Voice of the DBA

Tag: deployment

  • No State Based Deployments

    I’ve been studying deployment ideas and technologies for quite some time now, but it’s been a focus for the last year or two as my employer is interested in this area. I’ve been reading books and talking to people, comparing their experiences with my own. I’ve found that deployments for databases are much more complex than those for other software, but essentially there are two approaches that people choose: migrations and state based migrations based deployments.

    Migrations based deployments are based on tracking each change to a database in a separate script and then executing each script, in order, on your production database. This is hard to do, requires discipline, and is what most people find difficult to do. Scripts get quite numerous, ordering is hard, and deployments become scary and largely unsuccessful without a lot of care. The scripts require tweaks and fixes to work efficiently, and you need smart DBAs and developers.

    State based deployments are what many people try to use. They look at the state of production, the state of development, run some tool like SQL Compare, or a process, and determine what changes need to be made to production to get it to match development. There are variations, but this is the basis for how many people try to create a deployment process. It seems easier.

    The problem is that at some point a state based deployment process won’t work. It’s actually impossible for a state based approach to work in all situations. Simple actions, like renames, can’t be handled by examining two states. In order to properly manage changes, and keep your data intact and safe, you need to understand “what” happened during the process, not just the end result.

    In many situations, state based deployments can work, and they can be used successfully, but there is always the possibility that you’ll need to customize your state based approach to include some migrations scripting. Keep that in mind, and make sure that any process or tool you choose has that flexibility. If you understand you need to review and modify scripts at times, you should be successful.

    Steve Jones

    The Voice of the DBA Podcast

    Listen to the MP3 Audio ( 2.5MB) podcast or subscribe to the feed at iTunes and LibSyn.

  • Selling Automation to Ops

    The DevOps movement isn’t new in some companies. It’s the same coordination and teamwork that has existed for a long time between the development and operations staffs. Developers take advantage of the skills in Operations to get standardized environments for their work, and let the Ops people manage (and track) changes. Operations people talk to developers about the challenges and issues faced in production, and the let the developers build applications that can easily be deployed. The sharing of information ensures each group knows what the other faces, and the regular contact builds bonds and respect between employees. Neither wants to let the other down or make someone else’s job any harder than it needs to be.

    However that’s not the case in many companies where developers view Operational staff as complainers that slow the process down. Operations staff see developers as wild and irresponsible, tossing code into production that they don’t need to support and haven’t tested. Both of these views are correct in that each side sees a reality in the process that makes their job more difficult.

    Ultimately I believe it’s up to developers to change things. Those of us that build the software need to respect the problems that instability causes and learn to help ensure that our changes can be deployed smoothly. The development side of an organization has more skill in tracking changes in version control, in managing the movement of those changes among environments, and in programming systems. We should be working to help push that knowledge through to Operations personnel that become responsible for our applications.

    That means we need to build scripts and tools to migrate our changes and give them to Operations. I’d recommend that we learn how to automate the configuration of our development systems, as well as script our changes. Most modern platforms allow us to programmatically make changes, so let’s do that. Then let’s take a few hours and show Operations people how to use these scripts, and let them setup and change our development environments. It will be slow at first, but they’ll learn to make changes faster, but also bring stability to every environment from development to QA to production, and can ensure we have the same configurations everywhere. We’ll have one less thing to manage, and our changes will get deployed faster, but also more consistently.

    Ultimately we all want the same thing. Better software delivered to customers faster. We want to Ship Safe, and Ship Often.

    Steve Jones

    The Voice of the DBA Podcast

    Listen to the MP3 Audio ( 2.5MB) podcast or subscribe to the feed at iTunes and LibSyn.

  • Recovering from Bad Deployments

    Someone asked me recently if I stored backups of my database in version control. This person wanted to recover from a bad deployment and use a backup to do so. They felt that keeping a copy of the backup in a VCS, alongside the code being deployed, would be important. It might be, but I said that recovering from a bad deployment isn’t something I want to do with a restore if I can avoid it. Then I was asked how to recover from a bad deployment if you have a busy, 24×7 environment.

    I, of course, answered, “it depends.”

    It does depend on the deployment, but it also depends on your preparation. There are ways in which you can work to minimize the problems that might occur during a deployment. Obviously testing your scripts and deployment process is important, but it’s also good for you to understand how your scripts work and what techniques you can use to rollback problematic deployments.

    There’s a switch in SQL Compare that lets you build a deployment script, and then immediately generate a rollback script. It’s handy, but it’s also not going to always work. If you’ve added a column during deployment, you might not want to just remove it on rollback. 

    However you can prepare for issues, like having a script that might save data in the new column before you remove it. You might choose to copy the table as part of a pre-deployment process (or during deployment), having this copy of the table used in a rollback scenario. You might even bring up a warm copy of your database and prepare to swap entire databases if problems arise. This would allow you time to save and move data that was changed after your deployment, but before your rollback.

    There are lots of possibilities in how you might recover from a failed deployment, but as with many of the solutions that we build in technology, a well thought out plan makes everything run smoother.

    Steve Jones

    The Voice of the DBA Podcast

    Listen to the MP3 Audio ( 2.1MB) podcast or subscribe to the feed at iTunes and LibSyn.