Server Migration

Migrating a server means moving a running machine: the data, the services, the dependencies and the users relying on them. LaMeDuSe prepares and executes your server migrations with rehearsal, rollback plan and minimal downtime.

The problem you're facing

An improvised server migration shows in its aftermath: services that won't restart, data lost in a forgotten disk corner, DNS still pointing at the old server three days later, and users discovering problems before you.

Our answer

We treat migration as a short, scoped project: inventory of services and data, rehearsed test migration, data synchronisation before cutover, planned intervention window, post-cutover checks and the old server kept as a safety net.

What LaMeDuSe takes care of

  • Complete inventory: services, data, dependencies, DNS
  • Migration plan with window and rollback
  • Full rehearsal on a test server
  • Data synchronisation before cutover (minimal downtime)
  • Planned cutover with systematic checks
  • Old server kept as fallback during stabilisation
  • Documentation of the target environment

Our methodology

  1. 1

    Inventory & mapping

    Applications, data, dependencies and flows: you only migrate well what you have mapped.

  2. 2

    Migration plan

    Migration order, intervention windows, rollback plan and success criteria defined per application.

  3. 3

    Rehearsal

    Every migration is rehearsed in a test environment before the real switch: no first time in production.

  4. 4

    Cutover & stabilisation

    Execution on D-day, verification of monitoring, backups and performance, then handover.

Relevant technologies

Kubernetes, HelmTerraform, AnsibleRestic, Borg, rsyncVeleroPostgreSQL logical replication

Security

Encrypted transfers, segmentation of source and target environments, and data integrity checks after every step.

Hosting & infrastructure

The target can be our European cloud or your own infrastructure: we migrate to what serves your constraints, not the other way round.

Maintenance

After the cutover, operations can stay with us (managed services) or return to your teams, with full documentation.

Why LaMeDuSe

We've migrated production servers for years, ours and our clients', with the same rule: never a first time on D-day. Every migration is rehearsed before it's executed.

Frequently asked questions

From a few minutes (simple service with prior synchronisation) to a few hours (large database). Duration is measured during the rehearsal, so it's known before D-day : not finger-in-the-air estimated.

Yes: physical-to-virtual conversion (P2V) when hardware tires, data and service takeover, and modernisation of operations (monitoring, backups) along the way. A server no longer maintained by its vendor is a risk : migrating is often the right decision.

The rollback plan applies: the old server isn't decommissioned before the end of the stabilisation period, and abort criteria are defined before starting. You go back in hours, not days.

Yes: OVH, Scaleway, AWS, on-premise servers : to our European infrastructure or to another host if that serves your project. Migration is our craft; the destination is your decision.