UPTIME target
99.95% monthly
Monthly availability commitment for End Customer VMs. Measured monthly; planned outages excluded.
SLA OVERVIEW
Availability, incident handling, and service credits
Separation of control, compute and storage keeps the platform reliable, easy to grow, and simpler to operate over time.
Standard support hours: 09:00–17:00 EST (additional coverage by agreement).
typical Response
Under 30 minutes
Typical first reply in the shared Slack channel (working hours). SLA first response targets: 30 min–4 h (by priority).
CREDITS
Up to 100%
Service credits applied to future payments (no refunds). Issued when monthly credit exceeds €100.
OPERATIONAL MODEL
How support works in practice
Clear channel roles, disciplined tracking, and proactive signals help partners resolve issues faster.
Dedicated Slack channel
Each partner typically operates with a shared channel for day-to-day coordination, questions, and incident updates.
Ticketing for accountability
Production issues should be registered as tickets to ensure ownership, tracking, and closure.
Proactive notifications
Where configured, automated health signals can surface potential bottlenecks early in the shared channel.
Escalation path
Escalations follow the L1–L3 model below, with ownership split between provider and platform.
SUPPORT MODEL
Responsibilities by support level
Warren supports the Partner on Platform-related topics, while the Partner remains the primary interface for End Customers. Escalations follow the levels below.
Level | Description | End-user requests | Admin requests |
|---|---|---|---|
LEVEL 1 | Triage & guidance
| Provider | Warren |
LEVEL 2 | Investigation & coordination
| Provider * | Warren |
LEVEL 3 | Engineering & remediation
| Warren (Platform)
| Warren (Platform)
|
* Level 2 is typically handled by the Provider, but can be outsourced to Warren (additional fees agreed case-by-case).
AVAILABILITY
Uptime commitment
Warren uses commercially reasonable efforts to make End Customer resources (virtual machines) available 99.95% of the time, measured monthly.
1
Single VM connectivity loss
A single virtual machine has no external connectivity.
2
Multi-VM connectivity loss
Workloads across two or more virtual machines (same region) concurrently have no external connectivity.
3
Multi-VM volume IO = 0
Attached volumes across two or more virtual machines perform zero read/write IO while IO is pending in queue.
Planned outages
Installation of upgrades, routine server/network configuration changes, and other reasonable maintenance activities.
Exclusions
Improper Partner configuration, unsupported hardware/software, or insufficient hardware redundancy.
Resilience and proactive operations
- The platform is designed to recover from routine infrastructure failures through automation where possible.
- Where configured, health signals can trigger automated notifications in the shared channel to surface early warning signs.
- When manual action is required, incidents follow the priority targets listed below.
SERVICE CREDITS
Transparent credit policy
If the monthly uptime target is not met, Outages are compensated with service credits according to the table below.
Monthly uptime percentage | Service credit |
|---|---|
Less than 99.95% but equal to or greater than 99.0% | 10% |
Less than 99.0% but equal to or greater than 95.0% | 30% |
Less than 95.0% | 100% |
How credits are applied
Applied only against future payments (no refunds).
May be issued on the next invoice for the billing cycle in which Outages occurred.
Non-transferable; cannot be applied to another account.
Private & Enterprise Clouds
Calculated based on the percentage of End Customers affected.
Applied relative to revenue generated via the Platform as Warren subscription fee.
Issued only if the monthly credit amount exceeds €100.
Requesting credits
Submit a request with relevant timestamps, affected resources, and region. Warren confirms eligibility and applies credits where applicable.
INCIDENT HANDLING
Priorities and targets
Problems are handled according to the priorities, first response targets, and resolution targets below. If a notification is sent outside standard support hours, the clock starts when standard support becomes available.
Priority | Description | First response | Resolution | Goal |
|---|---|---|---|---|
Critical | Outage or unavailability of over 30% of End User resources (virtual machines) | 30 min | 4h | 95% |
High | Outage or unavailability of End User resources (virtual machines) | 1h | 8h | 90% |
Normal | Outage or unavailability of Warren platform UI & API | 2h | 16h | 85% |
Low | Unavailability of supporting functions or Add-on Products | 4h | 32h | 80% |
Communication during incidents
- Updates are shared in the partner channel and the ticket thread.
- For major incidents, a brief post-incident summary can be provided after resolution.
Support
Submitting requests
Problems should be reported promptly. For production-impacting issues, open a ticket for tracking and closure; Slack can be used alongside for coordination.
SLACK (collaboration)
Direct partner channel
Day-to-day coordination, questions, and incident updates with the Warren team.
Ticketing (system of record)
Issue tracking & closure
Register incidents and tasks for ownership, tracking, SLA timing, and formal closure.
Email (fallback)
Use when Slack is unavailable or for formal notices. Monitored during standard hours.
Expenses and out-of-scope
If a request is outside Warren’s responsibility (for example: third-party software not part of the Platform, underlying hardware or provider-managed configurations, or end-user application-specific issues), the Partner compensates Warren for expenses incurred to process the request.
Documentation and resources
Technical resources
Downloadable architecture and reference materials for deeper technical insight.
