With more than 1,700 employees, over 80 stores and 100,000 customers, Foussier relies on a strongly multichannel sales model spanning its website, field sales representatives, stores, and telephone. The availability of its information system is therefore directly linked to the continuity of its business.
A few years ago, Foussier had migrated its applications to Kubernetes, deploying on a managed Azure AKS cluster and a second cluster on its own infrastructure. The development teams fully mastered their core business, but operating the clusters required specialized expertise with significant operational stakes around Kubernetes and the numerous middleware in use. Cloud costs, for their part, remained difficult to forecast.
The group therefore sought a partner capable of taking over the complete operation of its platforms with 24/7 support. This is what Enix offers with its Secure Managed Kubernetes (SMK) solution: a Kubernetes environment that is not only managed but also fully administered, with close, hands-on support.
Take back the existing setup before transforming it
The collaboration does not begin with a migration but with a practical takeover of what already exists. Enix first audits the current environment and takes charge of operating the two Kubernetes clusters already in production.
In two weeks, the teams put the environments under supervision, secure remote access, and resume outsourced management as-is. Foussier thus quickly benefits from outsourced, responsive operations, without waiting for the future architecture to be built.
Only then does the transformation project begin. To replace the AKS cluster on Azure and to modernize the vanilla Kubernetes installed at Foussier, Enix proposes to stay with a hybrid architecture but rebuild the two environments around its Secure Managed Kubernetes service, while leveraging the existing infrastructure.
SMK, the same Kubernetes solution on two infrastructures
The first SMK platform is hosted by Enix on a Proxmox VE virtualization stack deployed in high availability, distributed across three data centers in the Île-de-France region (covering both the Kubernetes control plane and the worker nodes). This platform largely hosts the workloads that had been on the public Azure cloud.
The second SMK platform is deployed directly at Foussier, on the VMware infrastructure the company already owns. It hosts the same IT applications as the initial platform but in a more industrialized and secure manner. The two environments are linked via a secure site-to-site VPN tunnel.
This is one of SMK’s core principles: Enix operates the same type of Kubernetes cluster regardless of the underlying infrastructure. At Foussier, it involves Proxmox on one side and VMware on the other. The company thus benefits from a uniform level of service, operations, and security across both platforms while leveraging its existing infrastructure without having to move it wholly to the service provider.
This approach also helps limit dependence on a single technology or cloud provider. Kubernetes and the components used are based on open standards and technologies, with an architecture that Foussier could evolve or migrate to another environment if its needs were to change.
Talos, security realized with restraint
Security, the “S” in SMK, rests on sound security practices and the use of specialized components, notably a Talos-based foundation—an immutable operating system designed specifically for Kubernetes environments.
Updates are performed by automatically reinstalling the system, ensuring a uniform configuration across all servers. Talos carries only the components strictly necessary: fewer software packages mean fewer potential vulnerabilities and a reduced attack surface. For example, SSH access—the remote connection usually used by administrators to intervene on a server—is removed.
Services are also isolated so that an incident in one does not spread to others. This is a crucial requirement to avoid pooling operational risks across the different sales channels. The new architecture relies on standard middleware controlled by Enix, such as PostgreSQL, Redis, and RabbitMQ. These are deployed in the Kubernetes cluster via operators and are also monitored with a cloud-native monitoring stack.
Two new Kubernetes platforms operational in 2.5 months
Following the takeover of the existing setup, Enix spent about a month building the two new SMK platforms on the two infrastructures (including its Proxmox private cloud for the platform deployed in Enix’s data centers). An additional month was needed to migrate the applications and their data to these new environments. The Helm charts and application packaging were taken over by their teams to adapt them to cloud-native best practices.
In total, 2.5 months after project kickoff, the new Kubernetes platforms were in production, hosting all workloads and fully managed by Enix.
The migration had been prepared and tested in advance to minimize its impact on operations. The switch-over was performed in the evening, within a low-traffic window, with service downtime limited to about one hour for users.
Controlled costs, greater maneuvering room
At the start of the project, price was not Foussier’s primary criterion. Yet, the combined tailored hosting and managed service proved more economical than the former Azure public cloud platform.
That saving primarily enabled Foussier to scale up without increasing its budget. The company moved from three to four clusters and now has a dedicated development platform for each environment, with more powerful servers and Kubernetes nodes that have greater resources.
Another advantage of deploying on dedicated, pre-provisioned infrastructures (instead of public cloud like Azure or equivalent) is that costs are predictable and scoped from the outset of the collaboration.
A relationship designed for the long term
At Enix, a Run-focused specialist, a project does not end with the construction of platforms and migration. It is always framed as a long-term partnership, led by a tight-knit team of experienced engineers who manage operations and later support the platform’s evolution.
The promise of proximity and responsiveness in operations that Foussier sought is demonstrated by using a dedicated messaging channel between the teams and the senior engineers responsible for the platform, smoothing communications and the tracking of operations.
As the weeks passed, trust grew on both sides, eventually opening to Enix’s teams the doors of Foussier’s “factory” for a fascinating tour of the company’s know-how.
A new phase is now underway: migrating storage services still present on Azure to Ceph-based object storage managed by Enix. Foussier thus continues the movement begun with SMK: reducing dependence on public cloud while gaining greater control, predictability, and freedom of choice over its infrastructure.
