Skip to content
Atlas Hoster
Service status

Your infrastructure,
held to 99.9 %.
In writing.

NVMe VPS, dedicated servers and high-availability architectures, run by a team of engineers in Morocco since 2007. Response and recovery times written into the contract, and a client area where everything can be checked.

Priced proposal within 24 business hours · migration handled for you

  • Response 15 min · recovery 2 h (Critical)
  • On-call engineers 24/7
Moroccan hosting and managed services provider since
2007
Monthly availability committed, VPS and dedicated servers, credits in the contract
99.9%
Response and recovery, Business tier, 24 / 7
30min · 4h
Response and recovery, Critical tier, around the clock
15min · 2h
Client area

What we promise, you can check.

Every server and its state, refreshed every 30 seconds, the month's availability against the SLA, backups, tickets and invoices in dirhams: the same portal for management and for the technical team.

Screen shownClient area preview — fictional data
Atlas Hoster client area, Services page: five fictional servers (web-prod-01, web-prod-02, db-primary-01, db-replica-01, cache-01) operational, availability this month 99.98 %.

Captures of the real Atlas client portal, on a demonstration account (in French): company, contact, servers and IP addresses (documentation range 203.0.113.0/24) are fictional.

01 · management

The contract, measured

The month's availability is measured by our monitoring and compared with the 99.9 % commitment. A credit is asked on figures, not on an impression.

02 · technical team

Every machine in detail

Checks, metrics, HTTP probes, nginx logs, DNS and Cloudflare security rules, on your services and nothing else.

03 · accounting

Invoices and due dates

Invoices in dirhams, renewal calendar, monthly reports and documents, with role-based access for each colleague.

Service commitments

Business, the first tier truly in writing.

A response time and a recovery time, both in the contract, and a credit on the monthly fee when we do not meet them.

Service level agreement (SLA) · updated on October 11, 2026Read the full SLA
Recommended · entry point

Business

24 / 7
30 min
Response
4 h
Recovery
4 business h
Restore
  • 99.9 % infrastructure availability
  • Daily backups, 30 days retention

Critical

24 / 7
15 min
Response
2 h
Recovery
2 h
Restore
  • On-call engineers, nights and weekends
  • Daily backups, 30 days retention

Essential remains available for staging and showcase sites: response within 4 business h, no guaranteed recovery time.

Credits, on figures

Asked within 30 days, applied to the next invoice, up to 50 % of the monthly fee. Measured by our monitoring, visible in your client area. Maintenances announced 48 h in advance excluded.

If, over the monthCredit on the monthly fee
Availability below 99.9 %10 %
Availability below 99.0 %25 %
Availability below 95 %50 %
Response time missed10 % per miss
Public status

The status is public. Incidents are explained.

The status page is fed by the monitoring that measures your SLA: what you read is what we measure.

Every incident announcedOn the status page and in your client area.
Written reportAfter any major incident (over 30 minutes or with client impact).
Maintenances announced 48 h aheadOutside your peak hours whenever possible.
See the status page
Under the hood

A stack we can explain to you.

For the technical team doing the architecture review: each layer, its role, and who runs it.

compute

Proxmox VE, KVM

Proxmox clusters, KVM virtual machines with root access.

storage

Ceph replicated on NVMe

Ceph storage replicated on NVMe drives, for performance and redundancy.

backup

Proxmox Backup Server, verified

Scheduled and verified; restore from the client area, with a delay committed in the contract.

network

Cloudflare anti-DDoS, web application firewall and CDN

Abusive traffic blocked, rules and logs readable in the client area.

operations

Monitoring and engineers 24/7

Monitoring, probes and alerts; a team that acts, not a queue.

Migration

Tell us what you host.

An audit of what you run, a priced proposal, then a switchover planned with you and followed in your client area.

Priced proposal within 24 business hours · migration handled for you

  1. Audit of what you runServers, databases, DNS, e-mail.
  2. Priced proposalWithin 24 business hours, in MAD excl. tax.
  3. Migration by our engineersPlanned outside your peak hours.
  4. Switchover and monitoringYour services show up in the client area.