5 DevOps Management Faults

Sometimes the barriers to success with DevOps can't be blamed on poor tools or poor workmen, but on the managers who are in charge.

Andrew Froehlich, President & Lead Network Architect, West Gate Networks

May 4, 2017

4 Min Read
Image: Pixabay/Paulbr75

By all estimations, DevOps has been proven to be effective in accelerating the delivery of applications with a consistent level of code quality. In fact, DevOps has such a big impact on enterprise IT, that some call it a "key driver of business growth" in 2017. Despite the potential, management of DevOps has proven difficult in some organizations. The biggest problems seem to revolve around improper management practices that should bind operations and development teams together.

So where are DevOps managers falling short? We’re going to talk about five key issues today that span areas of communication, coordination and improper usage of DevOps tools.

1. Not requiring Ops to participate in quality analysis. One of the biggest management mistakes that can be made in any DevOps practice is not requiring operations personnel to participate in the code QA process. While it’s not always completely necessary, there are situations where infrastructure related problems can be identified and addressed prior to moving updated code into production. For managers who are new to DevOps, this is a vastly different approach from traditional siloed duties that keep operations separate from developer processes. Yet, this is precisely the point of DevOps in the first place. It’s up to the DevOps manager to understand the importance, create an atmosphere of cooperation, and follow through with in-depth test analysis from both sides of the house.

2. Managers consistently stall deployments. An issue that often is not discussed lies in the fact that many DevOps managers place too many roadblocks that ultimately delay deployments and hinder the effectiveness of a DevOps strategy. While some delays are completely legitimate from time to time, it’s important to make sure that DevOps managers have fully bought into a DevOps mindset for application deployments. In many cases where DevOps projects struggle, you’ll likely find they are being managed by someone who continues to feel more comfortable rolling out software in a traditional batch deployment format. If this is the case, your DevOps efforts will likely never pay dividends.

3. Neglecting the proper use of DevOps tools. It’s easy to blame DevOps tools for failures. But in many cases, the tools are perfectly adequate. Instead, management either lacks the overall understanding of how DevOps tools should be used – or the proper use of those tools is not enforced. Not everyone on the DevOps team is going to agree on the type of tools, or the best way to use them. That’s why you often see a split between how tools are used by Devs and operations admins. It’s critically important that these variations be curtailed to streamline processes and facilitate better coordination between the two teams.

4. Not providing an appropriate and up to date test environment. Possessing the proper DevOps tools – and enforcing their proper usage -- is one thing, but also providing the appropriate test environment is still another. It’s the responsibility of the manager to provide a QA infrastructure that closely matches production. Problems arise when major changes or upgrades occur to the production environment without making similar modifications to the test environment. Managers can’t simply assume that developers are going to complain when production and test environments are out of sync. Monitoring and ensuring test environments are up to spec is a duty the DevOps manager should be fulfilling.

5. Ignoring the benefits of value-stream maps. Value-stream maps are a way to gauge whether what you are doing from a DevOps perspective is adding value to the organization. As applications and underlying infrastructures grow in complexity, it’s easy to get caught up in simply going through the motions of a continuous integration (CI) strategy while forgetting why they’re doing it in the first place. A value-stream map baselines the value of an application today, so you can determine the added value over the course of incremental updates. This process helps to visualize where the true value of DevOps strategies can be made – and where the focus of innovation should be.

The practice DevOps has proven itself many times over to be useful in enterprise environments. If your organization is still struggling, don’t be so quick to assume the problem lies in commonly assumed shortcomings of staff skillsets or inappropriate tools. Instead, take a step back to see if the problem lies at the management level. Despite the greatest talent, training, tools and good intentions, without proper leadership, DevOps projects will continue to disappoint.

Read more about:


About the Author(s)

Andrew Froehlich

President & Lead Network Architect, West Gate Networks

Andrew has well over a decade of enterprise networking under his belt through his consulting practice, which specializes in enterprise network architectures and datacenter build-outs and prior experience at organizations such as State Farm Insurance, United Airlines and the University of Chicago Medical Center. Having lived and worked in South East Asia for nearly three years, Andrew possesses a unique international business and technology perspective. When he's not consulting, Andrew enjoys writing technical blogs and is the author of two Cisco certification study guides published by Sybex.

Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.

You May Also Like

More Insights