⚡ TL;DR — The Promises That Matter
- 99.9% uptime guarantee on all paid VPS plans — measured monthly
- Free tier has no SLA — we target 99.9% but don't guarantee or compensate for downtime
- Credit compensation up to 100% of monthly fee when we miss the target
- Scheduled maintenance doesn't count as downtime (announced 48h+ in advance)
- We measure from our network edge — not from your ISP or application layer
- You must claim credits within 7 days of the incident — we don't auto-apply them
- Network SLA: 99.99% — separate from compute, with its own compensation schedule
1. Scope & Applicability
This Service Level Agreement ("SLA") applies to all paid virtual private server plans purchased through FreeVPS.it ("we," "our," "the Service"). It forms part of the agreement between you and FreeVPS.it, alongside our Terms of Service, Acceptable Use Policy, and Privacy Policy.
1.1 What This SLA Covers
- Compute availability — your VPS instance being reachable and operational
- Network availability — connectivity between your VPS and the public internet
- Storage availability — ability to read/write to your VPS disk
- Control panel availability — ability to manage your VPS via our web interface
1.2 What This SLA Does Not Cover
- Free-tier (Starter Tier) VPS instances — see Section 10
- Application-layer issues within your VPS (your software crashing is not our downtime)
- DNS resolution for your domains
- Third-party services integrated with your VPS
- Issues caused by your violation of our Acceptable Use Policy
2. Uptime Commitment
We commit to the following monthly uptime percentages, measured from the first day to the last day of each calendar month:
"Uptime" means your VPS instance is powered on, the hypervisor is responsive, network connectivity is available, and disk I/O is functional. We measure this from our infrastructure monitoring — not from external probes that might be affected by internet routing issues unrelated to our platform.
3. Downtime Allowance — The Math
Here's what the uptime percentages translate to in actual allowed downtime:
These numbers are based on a 30-day month (43,200 minutes). Actual monthly minutes vary slightly.
4. How We Measure Uptime
4.1 Monitoring Infrastructure
We run internal monitoring agents on every host node that check the following every 60 seconds:
- Hypervisor health: KVM/QEMU process status, libvirt daemon responsiveness
- VM state: Whether each VPS instance is in "running" state at the hypervisor level
- Network reachability: ICMP and TCP probes to the VPS's assigned IPv4 gateway
- Storage I/O: Read/write latency on the backing NVMe array
- Control panel API: HTTP health checks against the management API endpoint
4.2 Downtime Definition
A downtime event is recorded when any of the following conditions persist for more than 5 consecutive minutes:
- Your VPS is reported as "stopped" or "crashed" at the hypervisor level without being initiated by you
- Network packets to/from your VPS's gateway experience 100% packet loss
- Disk I/O latency exceeds 10 seconds for read or write operations
- The control panel is unreachable, preventing you from managing your instance
4.3 What Doesn't Count as Downtime
- Outages lasting less than 5 consecutive minutes
- Scheduled maintenance announced 48+ hours in advance
- Issues caused by your actions (bad firewall rules, kernel panics from your software, resource exhaustion)
- DDoS attacks directed at your IP (we mitigate, but can't guarantee zero packet loss during active attacks)
- Force majeure events (natural disasters, war, government orders)
5. Credit Compensation
When we fail to meet the uptime commitment for a given calendar month, you are entitled to service credits applied to your next billing cycle. Credits are calculated as a percentage of your monthly recurring fee for the affected service.
5.1 Compute SLA Credits
| Monthly Uptime | Downtime (approx.) | Credit — Standard/Pro | Credit — Dedicated |
|---|---|---|---|
| 99.9% – 99.0% | 44 min – 7.2 hrs | 10% | 10% |
| 99.0% – 98.0% | 7.2 hrs – 14.4 hrs | 25% | 25% |
| 98.0% – 95.0% | 14.4 hrs – 36 hrs | 50% | 50% |
| Below 95.0% | > 36 hrs | 100% | 100% |
5.2 Credit Limits
- Maximum credit per incident: 100% of the affected service's monthly fee
- Maximum cumulative credit per calendar month: 100% of the affected service's monthly fee
- Credits are applied as billing credits — not cash refunds
- Credits cannot be transferred between services or accounts
- Credits expire if not used within 12 months of issuance
5.3 Credit Calculation Example
Monthly uptime: (43,200 - 240) / 43,200 = 99.44%
SLA tier: 99.0% – 99.9% → 10% credit
Credit amount: $20 × 10% = $2.00 credit applied to April invoice
6. How to Claim Credits
SLA credits are not automatically applied. You must submit a claim. Here's the process:
- Identify the incident: Note the date, approximate start time, duration, and affected VPS instance(s)
- Submit a claim: Email support@freevps.it.com with subject line
SLA Credit Request — [Your Account ID] - Include details: Instance ID, datacenter region, description of the issue, and any monitoring data you have
- Deadline: Claims must be submitted within 7 calendar days of the incident. Late claims are not eligible.
- Review: We'll verify the incident against our monitoring data and respond within 5 business days
- Credit application: Approved credits appear on your next invoice within one billing cycle
7. Exclusions
The following events are excluded from uptime calculations and do not qualify for SLA credits:
7.1 Scheduled Maintenance
Planned maintenance announced at least 48 hours in advance via email and status page. See Section 8 for maintenance window policies.
7.2 Customer-Caused Issues
- Misconfigured firewall rules blocking all traffic (iptables/nftables/UFW lockouts)
- Kernel panic caused by your software or custom kernel module
- Resource exhaustion from your workload (OOM kills, full disk, fork bombs)
- Accidental shutdown or reboot initiated from within the VPS
- Failed OS update that renders the instance unbootable
7.3 External Factors
- DDoS attacks: Active attacks targeting your IP that overwhelm mitigation capacity. We filter L3/L4 attacks by default but cannot guarantee zero packet loss during sustained volumetric attacks
- Upstream provider outages: Failures at transit providers or IX peering points beyond our direct control
- Internet routing anomalies: BGP hijacks, route leaks, or other global routing events
- Force majeure: Natural disasters, war, government-mandated shutdowns, pandemics, acts of terrorism
7.4 Policy Violations
Downtime resulting from enforcement actions under our Acceptable Use Policy — including suspension for crypto mining, spam, DDoS origination, or other AUP violations — is excluded from SLA calculations.
7.5 Account Status
SLA credits are only available to accounts in good standing. Accounts with overdue payments, active AUP investigations, or pending termination are not eligible for credit claims.
8. Maintenance Windows
8.1 Scheduled Maintenance
Routine maintenance that may impact service availability is performed during designated windows:
| Maintenance Type | Preferred Window | Advance Notice | Expected Impact |
|---|---|---|---|
| Host node patching | Tuesday/Wednesday 02:00–06:00 UTC | 48 hours minimum | Brief reboot (1–5 min per VM) |
| Network upgrades | Sunday 01:00–05:00 UTC | 72 hours minimum | Possible brief connectivity gaps |
| Storage firmware | Wednesday 03:00–05:00 UTC | 48 hours minimum | Temporary I/O latency increase |
| Major infrastructure | Scheduled per event | 7 days minimum | Varies — detailed in announcement |
8.2 Emergency Maintenance
When a critical security vulnerability or hardware failure requires immediate action, we may perform emergency maintenance without advance notice. In such cases:
- Users are notified as soon as the maintenance begins (or within 15 minutes)
- A post-incident explanation is published within 24 hours
- Emergency maintenance does count toward SLA calculations if it exceeds 30 minutes total duration
8.3 Maintenance Notifications
Scheduled maintenance is communicated through:
- Email: Sent to the account email for all affected VPS instances
- Status page: Published at freevps.it.com/status.php
- Control panel banner: Displayed to all logged-in users
9. Incident Response
When an unplanned incident occurs, here's what happens and when:
Detection
Automated monitoring detects the anomaly. On-call engineer receives alert via PagerDuty. If 3+ VPS instances on the same host are affected, it's auto-escalated to a platform-wide incident.
Acknowledgment
On-call engineer acknowledges the alert and begins triage. Status page updated to "Investigating" if the incident affects multiple users.
Initial Assessment
Root cause identified or narrowed down. Status page updated with scope, affected region, and estimated time to resolution. Email notification sent to affected paid-plan users.
Mitigation
Fix deployed or workaround implemented. If the fix requires a longer maintenance window, users are notified with a revised ETA. Status page updated to "Identified" with cause description.
Resolution
Service fully restored. Status page updated to "Resolved." All affected paid users receive email confirmation.
Post-Incident Report
For incidents exceeding 30 minutes: detailed post-mortem published on the status page. Includes root cause, timeline, impact scope, and corrective measures to prevent recurrence.
10. Incident Severity Classification
Incidents are classified by severity which determines response priority and communication cadence:
| Severity | Definition | Response Target | Update Cadence | Post-Mortem? |
|---|---|---|---|---|
| P1 — Critical | Complete service outage affecting multiple users or entire datacenter region | ≤ 15 minutes | Every 30 minutes | Yes (within 48h) |
| P2 — High | Partial outage: degraded performance, single host node failure, or control panel outage | ≤ 30 minutes | Every 60 minutes | Yes (within 72h) |
| P3 — Medium | Non-critical degradation: elevated latency, intermittent packet loss, slow panel | ≤ 2 hours | Every 4 hours | If > 4h duration |
| P4 — Low | Cosmetic or minor issues: status page reporting errors, non-functional UI elements | ≤ 24 hours | Daily | No |
11. Incident Communication Channels
During active incidents, we communicate through multiple channels to ensure you're informed regardless of which platform is affected:
- Status page: freevps.it.com/status.php — primary source of truth during incidents
- Email: Automated notifications to affected paid-plan account holders
- Control panel banner: In-app notification for all logged-in users
- Twitter/X: @freevpsit for brief status updates during major incidents
We do not withhold incident information or delay communication to protect reputation. Transparency during outages builds more trust than silence.
12. Free Tier (Starter Plan) — Expectations
This section sets honest expectations for free-tier users. No legalese to hide behind — just clarity.
What Free-Tier Users Get
- We target 99.9% uptime for free-tier instances, using the same infrastructure as paid plans
- Free-tier VPS runs on the same hardware, same network, same datacenters as paid plans
- You receive the same status page updates and incident notifications
- Your data is treated with the same security and privacy protections
What Free-Tier Users Don't Get
- No uptime guarantee — we don't contractually commit to any specific uptime percentage
- No SLA credits — there's no monthly fee to credit against
- No priority support — free-tier support is best-effort, behind paid-plan tickets in queue priority
- No post-mortem obligation — though we typically publish them for all incidents regardless
- Maintenance priority: During capacity-constrained maintenance, paid instances are migrated first
13. Network SLA
Network availability is tracked separately from compute availability because network issues have different characteristics and causes.
13.1 Network Uptime Commitment
We commit to 99.99% network availability for all paid plans. This is measured at the network edge — specifically, the availability of the upstream router connecting your VPS's VLAN to the public internet.
13.2 Network Credit Schedule
| Network Uptime | Monthly Downtime | Credit |
|---|---|---|
| 99.99% – 99.9% | 4 min – 43 min | 10% |
| 99.9% – 99.0% | 43 min – 7.2 hrs | 25% |
| Below 99.0% | > 7.2 hrs | 50% |
13.3 Network Exclusions
- Packet loss or latency caused by DDoS attacks targeting your IP
- Internet routing events outside our autonomous system (AS)
- Performance degradation caused by your bandwidth exceeding your plan's allocation
- IPv6-only reachability issues caused by incomplete IPv6 adoption at remote endpoints
14. Historical Performance
We publish our uptime metrics publicly because we believe accountability requires visibility.
14.1 Where to Check
- Real-time status: freevps.it.com/status.php
- 30-day history: Rolling uptime percentage displayed on the status page per datacenter region
- Incident archive: Complete history of all P1 and P2 incidents with post-mortem reports
14.2 Transparency Commitment
We do not hide incidents, silently remove downtime records, or manipulate uptime statistics. Our status page shows both automated monitoring data and manually reported incidents. If you experienced an outage that isn't reflected on the status page, report it to support@freevps.it.com and we'll investigate.
15. SLA Modifications
We may update this SLA to reflect infrastructure improvements, new service tiers, or changes in our operational capabilities.
15.1 Favorable Changes
Changes that improve the SLA (higher uptime commitments, larger credit percentages, shorter response times) take effect immediately upon publication.
15.2 Unfavorable Changes
Changes that reduce SLA commitments will be communicated at least 30 days in advance via email to all paid-plan customers. You may cancel your service without penalty during this notice period if you disagree with the changes.
15.3 Version History
All previous versions of this SLA are archived and available upon request to legal@freevps.it.com.
16. SLA Contact Information
For SLA credit claims, uptime disputes, or questions about this agreement: