One of the more problematic issues that containers raise from an IT management perspective is how to monitor what’s occurring inside the container. As far as traditional IT monitoring tools are concerned every container is essentially a black box that hides the application.
At the recent DockerCon 2016 conference Sysdig moved to address that issue in highly regulated industries by making an on-premises instance of its container monitoring service available. Previously, Sysdig’s container monitoring service was only available as a service.
As a phenomenon primarily led by developers, most containers today are invoked using a cloud service. But Apurva Dave, vice president of Marketing for Sysdig, notes that as internal IT organizations become more familiar with containers, they are deploying them either on a local instance of a platform-as-a-service (PaaS) environment or using containers to construct their own lighter-weight container-as-a-service (CaaS) environment.
In either case they need a container monitoring tool that allows them to see what’s occurring inside the container. To achieve that goal, Dave says, Sysdig developed ContainerVision to instrument the kernel on which the containers run. Every system calling the application residing inside a container becomes visible to the IT operations team.
In general, there’s some debate concerning who will be responsible for monitoring containers. There is one school of DevOps thought that contends whoever wrote the code is responsible for managing it. But there is a dedicated IT operations team in large organizations. The working theory is that every minute developers spend maintaining an application is time that would be better spent writing new applications.
Regardless to the degree of DevOps employed, Dave says developers and IT operations alike need a common framework through which they can collaborate. In addition, developers can use Sysdig to manage not only applications they wrote but also ones they inherit from other developers. Given the turnover rate in the application developer community, IT organizations need to have tools they can use throughout the life cycle of a container application.
Just as importantly, IT organizations can use Sysdig to holistically monitor all the containers that make up an application, Dave says. Rather than having to sort through reports on each container, Sysdig allows an IT organization to more holistically manage container applications. Like it or not, managing containers at scale is more challenging for IT organizations. The trade-off is that containers make developers more agile, which results in applications being developed faster than ever. The downside of all that agility is that containers tend to be ephemeral, which means IT operations team can have a hard time keeping track of where any given container might be running at any moment.
In the fullness of time, Dave says IT organizations will find themselves managing containers running on-premises and in the cloud. Regardless of whether those container are running on a virtual machine, in a PaaS or on a bare-metal server, Dave says Sysdig provides a common pane of management glass to monitor those containers. But as is often the case, something has to be seen before it can be managed.