Sergey Galushko - Fotolia
One of the principal benefits of containers is deployment flexibility. The idea is that a container is assembled from a series of layers, which allow the container to package all of the dependencies that are involved in the application -- thus enabling the container to run almost anywhere regardless of the platform. But the OS itself can pose a serious wrinkle in container compatibility that's often overlooked.
Even though containers share the underlying OS kernel, each container itself will include some OS components -- an OS layer. The problem is that the OS components included with the container must match the OS version running in the host kernel. If the OS versions don't match, the container might not function properly -- if at all.
Containers are blocked if the build numbers are different. For example, Microsoft uses a four-level version notation system that designates major, minor, build and revision -- such as 10.0.14393.0. Ideally, all four levels of the version system should match before a Windows or Hyper-V container will run on a Windows Server host. In actual practice, a container will start if there's a difference in the revision -- lowest -- designation level, but there's no guarantee that the container will work properly. However, no differences are allowed in the major, minor and build levels.
This can become a serious problem for DevOps teams. Consider an environment where an IT staff updates the OS on production hosts, only to find that hundreds, thousands or even tens of thousands of containers no longer function properly because of an OS kernel mismatch. Operations staff will need to coordinate OS patches and updates with developers so containers can be updated and redeployed with the equivalent OS layer in concert with production environment changes. The challenge is a bit less pressing for Hyper-V container deployments where there might be more latitude in delaying and coordinating VM OS upgrades.
Differences between Windows Server containers and Hyper-V containers
Evaluate Microsoft Azure Container Service
Understand the relationship between containers and microservices
Dig Deeper on Application virtualization
Related Q&A from Stephen J. Bigelow
Navigating data center malfunctions when hardware is off premises can be tricky. Organizations must have strong SLAs with their colo provider to ... Continue Reading
Regression tests and UAT ensure software quality and both require a sizeable investment. Learn when and how to perform each one, and some tips to get... Continue Reading
Learn the meaning of functional vs. nonfunctional requirements in software engineering, with helpful examples. Then, see how to write both and build ... Continue Reading