Support & operations
Support and day-to-day operations
After a VMware exit, the question is no longer the import: it is who watches the platform at 3am on a Sunday, and who does what when a VM stops answering. The commitments below are the contract’s, not a sales pitch; the binding text is on the general terms.
- 24×7 infrastructure on-call
- P1 critical: response within 4 h
- Engineers in Geneva, FR / EN
- Portal and email, 08:00–18:00 CET
- ISO 27001
- 3 DC
- SLA 99.99%
Updated 2026-08-23
Facts
Two regimes, not one
Support handles your requests on business days 08:00–18:00 CET. Infrastructure monitoring and on-call run 24×7: a platform incident does not wait for you.
Named engineers, not an offshore centre
Support runs from Geneva, in French and English. No scripted tier 1 before you reach someone who knows the platform.
Response ≠ repair
The published times are maximum response times. Repair times are not guaranteed and no MTTR is published; a committed deadline is contracted case by case.
Shared responsibility
Hidora operates and secures the Environment: hypervisor, three DCs, VPC, block replication, platform backup. You administer the OS, middleware, Active Directory and your applications.
GPU passthrough exception
On passthrough GPUs, SLA and MTTR commitments do not apply the same way: the contract says so explicitly. To be framed before GPU production.
Integration projects under SOW
Inventory, network workshop, VM wave imports, cutover: project work is ordered through a SOW, billed by time spent or by milestones, at the current rate card. It is not included in PAYG.
Maximum response times per priority, per the contract. Requests via portal and email. Repair times are not guaranteed and no MTTR is published.
| Priority | Description | Maximum response time |
|---|---|---|
| P1 – Critical | Complete system crash or hang, data corruption or loss, critical functions unavailable with no workaround | 4 h |
| P2 – Major | Restart or recovery needed after a failure, performance severely blocked, system running in restricted mode | 8 h |
| P3 – Minor | Error with a workaround available, minor performance degradation, minimal impact | 24 h |
| P4 – Low | Platform configuration or usage questions, improvement suggestions | 72 h |
In detail
You lose the vSphere console, and you also lose hardware refresh, per-core licences and the machine room. Day to day: the Hikube API and console to create, resize and destroy; quotas and the CHF estimate visible in the tenant; status.hikube.cloud for platform state; kubectl, Helm and S3 clients unchanged. Terraform and Cluster API are announced, not yet shipped: until then, the API and console are the path.
If the cause is the platform (host, storage, network, DC), Hidora’s 24×7 monitoring and on-call are already on it, independently of your ticket. If the cause is in your OS, application or Active Directory, it is your runbook: support picks the thread up on business days, and you open a P1 ticket for the trail. That boundary is the contract’s shared responsibility model, not a helpdesk policy.
The same checklist we hand you for a VMware exit has to apply to us. The contract governs reversibility at the end of the relationship: the return procedure, the retention period and any fees are set there, and the contract is what binds, the general terms. Technically nothing is proprietary: volumes exportable as disk images, S3-compatible storage, kubeconfig, standard kubectl and Helm. There is no SDN fabric or in-house backup format to buy back on the way out.
Three setups, chosen by the size of your team. A Hidora engineer frames the inventory, the network workshop and the 10–20 VM pilot: that is the starting point in every case. Then either your team runs the factory with the pilot’s runbook, or Hidora runs it as project work under a SOW (time spent or milestones, current rate card), or a partner integrator owns the project and keeps the relationship. None of the three is billed under PAYG: ask for the framing before you book a cutover window.
Frequently asked questions
Infrastructure monitoring and on-call are 24×7. The support desk handles requests on business days 08:00–18:00 CET, via portal and email. In other words: the platform is watched continuously, but an application-level ticket filed on a Saturday evening is picked up on Monday. If your production needs a wider commitment, that is written into the order or the SOW, not on this page.
Not to date: support runs in French and English from Geneva. The site and the contracts exist in German, the support language does not. For a German-speaking SME, settle that point with an engineer before the pilot rather than discovering it at the first incident.
Hidora monitors the platform: hosts, storage, network, DCs. You monitor your VMs from the inside (OS, services, applications) and you patch your OS and applications, exactly as you did under vSphere. The platform monitoring visible in the tenant does not replace your application monitoring.
Framing the pilot with an engineer is pre-sales. Running the factory (importing the following waves, cutover, runbook handover) is project work: SOW, time spent or milestones, current rate card, billed on top of PAYG. A partner integrator can own the project instead. Ask for the quote with the inventory in hand, not before.
The contract governs reversibility at the end of the relationship: the return procedure, the retention period and any fees are set there, and the contract is what binds, the general terms. Technically, volumes export as disk images, object storage is S3-compatible, clusters stay driveable with kubectl and Helm.
They are in Hidora SA’s contract documents, which are what binds: the general terms and the Hikube service agreement. The scope and the exceptions, including passthrough GPUs, are set there. Current platform state is public on status.hikube.cloud.
Ready to run on 100% Swiss infrastructure?
14-day trial, no credit card. GPUs included.