Cloud, Linux server and database management
Production servers with an owner, a runbook and a tested recovery
We administer the Linux servers and databases your business runs on. Patching, monitoring, deployment, DNS and certificates are handled under a written scope, and backups are restored on a schedule to prove they work.
Who this is for
- Businesses running their own application, store or portal on cloud or dedicated Linux servers.
- Development teams that want operations handled so they can ship features.
- Companies whose servers were set up by a contractor who is no longer around.
- Organisations preparing for growth, a migration or a customer security questionnaire.
Problems this service solves
- Alerts that nobody receives
- Monitoring is configured to page the people contracted to respond, with thresholds tuned to your workload instead of defaults.
- Backups nobody has restored
- We track two separate conditions: backup configured and recovery tested. Only a successful restoration test moves a system to “recovery tested”.
- Expired certificates and DNS surprises
- Domains, DNS records and TLS certificates are inventoried with renewal dates and automated wherever possible.
- Slow pages and mysterious load
- Performance diagnostics across Nginx, PHP-FPM and the database identify the real bottleneck before anyone buys a bigger server.
What is included
Coverage depends on the plan and the number and role of servers in scope.
Administration
- AlmaLinux and Ubuntu administration
- Nginx and PHP-FPM configuration
- MariaDB/MySQL administration and query review
- Operating-system and application patching
Reliability
- Monitoring and alerting
- Backup configuration and scheduled recovery testing
- Capacity and availability planning
- Incident triage and post-incident notes
Change and delivery
- Secure application deployment procedures
- DNS and SSL/TLS management
- Server and database migrations (scoped separately when large)
- Performance diagnostics
How onboarding works
-
Server inventory
Hosts, roles, versions, exposed services, DNS, certificates and current backup arrangements.
-
Baseline review
Patch level, access accounts, firewall rules, logging and backup status, recorded in a risk list.
-
Access and monitoring
Named SSH keys and accounts, monitoring agents, and alert routing agreed with you.
-
Runbook
Start, stop, deploy, restore and escalation procedures, written down and shared with you.
-
First restoration test
We restore a backup to an isolated environment. Until this succeeds, the system stays marked “backup configured”, not “recovery tested”.
Service windows
Server work is normally planned in maintenance windows agreed with you. Production plans can add Critical Monitoring with a defined P1 on-call response.
- Standard 8×5
- Eight contracted hours per business day, five days per week, in a window agreed for your timezone.
- Extended 16×5
- Two-shift weekday coverage for teams spread across regions or working long days.
- Critical Monitoring
- Continuous automated monitoring of agreed systems, with a defined Priority-1 on-call response for contracted services.
Your proposal states every window in your own timezone. Outside contracted hours, non-critical requests wait for the next window.
| Priority | Description | Example |
|---|---|---|
| P1 Critical | A supported business service is unavailable, or a severe security event is active, with no workaround. | Production platform down |
| P2 High | Major impairment affecting several users; workaround limited. | Email unavailable for a department |
| P3 Normal | An individual issue or non-critical degradation. | One user’s application problem |
| P4 Request | A planned change, advice or a minor inconvenience. | New user setup |
A response target is how quickly a qualified person starts work on your ticket. A resolution objective is what we aim for. It depends on the cause, on vendors and on access, so it is never a guarantee.
Security and data handling
- Access is by individual SSH key or named account, never a shared root password, and it is removed when an engineer leaves the engagement.
- Production data is not copied out of your environment except into backups you have approved. Restoration tests run in isolated environments.
- Changes follow an agreed change procedure, and emergency changes are documented afterwards.
- Backups can be encrypted, and their location and retention are recorded in your backup responsibility schedule.
Never send passwords, one-time codes, private keys, payment-card details or production secrets through this website or by email. During onboarding we agree a secure method for sharing access, and we use named, revocable accounts wherever the system supports them.
Pricing basis
Priced per server environment per month. The role of the server, its availability needs and the monitoring window determine the plan.
- Single Server
- Starting from USD 299–499 per month
- Production Server
- Starting from USD 599–999 per month
- Business Infrastructure
- Starting from USD 1,500+ per month
Starting-from prices are indicative and non-binding. A written proposal follows discovery. Unless the proposal says otherwise, prices exclude taxes, third-party licences, cloud consumption, hardware, travel and out-of-scope work.
Evidence
We publish client results only with measured baselines and written client approval. References for this service are available on request during discovery.
Ask for a referenceFrequently asked questions
Is my backup guaranteed to work?
Not until it has been restored. We report each system as either “backup configured” or “recovery tested”, and only a successful restoration test changes that status. Tests are repeated on the schedule in your agreement.
Which cloud providers do you work with?
We administer Linux servers on mainstream cloud and dedicated hosting providers where we can have appropriate administrative access. We are not a reseller or an accredited partner of any provider unless we state so on this site.
Do you guarantee uptime?
No. Uptime depends on your hosting provider, your application and your architecture. We monitor availability, respond within the targets in your SLA, and recommend changes that reduce the risk of downtime.
Can you migrate us to a new server?
Yes. Migrations are scoped as a project with a rollback plan, a test window and a cut-over checklist.
Do you support Windows servers?
Our core focus is Linux. Other platforms may be supported by agreement, and your proposal will list exactly which systems are in scope.
Know the real state of your servers
Start with a server and backup review: inventory, risks and a recovery-test plan.