SLA for Self-Hosted Cluster of Nodes
RPC Fast offers a self-hosted HA infrastructure setup along with optional tools such as Transaction Simulator, Mempool Data Stream, PredictKube, JSON RPC Caching Proxy, and others.
Service Level Agreement (SLA) for Self-Hosted Cluster of Nodes
The Setup Includes:
Initial configuration: servers setup, blockchain node integration.
Testing and validation: pre-launch testing, functional validation.
Monitoring and alerting deployment: real-time monitoring, alert configuration.
Performance optimization: node configuration for high throughput and low latency.
Additional tools/solutions setup and configuration: on request (TBD with the Client).
The Maintenance Includes(*):
Basic node monitoring: real-time monitoring, alert systems.
Node upgrades: software updates, hard forks.
Sync and network health: blockchain syncing, network performance tuning optimized for high network loads.
(*) What is not included: services not directly connected with the functioning of the above.
Scheduled Maintenance
In the event of scheduled maintenance, the Client is promptly notified as soon as the need arises. Notifications are sent via the designated communication channel, along with a proposed maintenance window, to ensure minimal disruption to the Client’s operations.
Support Schedule & SLA
We provide direct communication with our DevOps Engineers via a dedicated channel on Slack.
Service Level Agreement (SLA)
Priority Level 1: Critical issues response time is up to 4 hours (*). Update of server-side software in case of detecting critical vulnerability.
RPC Fast provides a fallback service (such as automated traffic rerouting to a healthy node during downtime or scheduled maintenance) with a load balancer.
Priority Level 2: Non-critical issues response time is up to 1 business day (*). Install, or update, or tune the server-side software, including required by Client’s application.
The support service is off during the public holidays in Ukraine:
Independence Day of Ukraine (August 24);
Christmas & New Year holidays (December 25 – January 5).
Maximum Initial Response Time
Requests from a Client’s contact person will be answered within a guaranteed time frame (see above). Support requests will be acknowledged within the maximum time listed, although actual response time may be faster. This does not include the total time required to resolve the request.
A shared dedicated channel between RPC Fast and the Client shall be used for reporting issues or bugs and for all written communication between the RPC Fast and Client teams. Critical issues must be reported through the dedicated channel. All business and financial communications shall be conducted via email.
Escalation Path: In the event that the primary point of contact is unavailable or unable to resolve a critical issue, please contact the Project Manager in the designated Slack channel or via email.
Important Context on SLA Design
As a dedicated infrastructure provider (not a SaaS provider for this matter), our SLA times are designed to reflect hands-on, system-level issue resolution – not just API-level service guarantees.
That said, our track record since 2018 shows:
99.99% node uptime.
Zero recorded incidents of actual service interruption.
All critical or potentially critical issues resolved immediately or within minimal time, often proactively before any impact occurred.
We treat infrastructure uptime as the core of our service promise and implement failover strategies even for single-node deployments to ensure uninterrupted availability.
Last updated