FAQ — Technical Aspects
Uptime monitoring without compromises
Frequently Asked Questions
Detailed answers to technical questions about our monitoring infrastructure, alerting logic, and performance metrics.
How does multi-region checking work?
StatusPul deploys lightweight monitoring agents across 14 strategic locations, including Frankfurt (DE), Tokyo (JP), São Paulo (BR), and Seattle (US). Each agent performs independent TCP/HTTP(S) probes every 60 seconds using dedicated /24 subnets. Responses are aggregated in our central time-series database, and latency percentiles (p50, p95, p99) are calculated per region. If a single location reports a timeout, it triggers a regional alert but does not mark the target as globally down until at least three geographically distributed nodes confirm the failure within a 180-second window.
What counts as downtime?
Downtime is recorded when a monitored endpoint fails to return a successful HTTP 2xx/3xx response or a valid TCP handshake within the configured timeout threshold (default: 10 seconds). SSL certificate validation errors, DNS resolution failures, and connection resets (RST) are also logged as incidents. Planned maintenance windows defined in your dashboard suspend downtime tracking automatically. We calculate monthly uptime percentages using the formula: (Total Minutes − Incident Minutes) / Total Minutes × 100, rounded to two decimal places.
How are alerts routed and deduplicated?
Alerting uses a state machine that evaluates consecutive failed probes before triggering notifications. By default, three consecutive failures generate an incident. Deduplication prevents alert storms by grouping related endpoints under the same root cause ID. Notifications are dispatched via webhook, email, or PagerDuty integration with a 15-minute cooldown period. You can customize escalation policies in the Alerts tab, including business-hour routing and on-call handoff schedules.
What data retention policies apply to historical metrics?
Raw probe data is stored for 90 days in high-resolution 1-minute intervals. Aggregated daily and monthly rollups are retained indefinitely for SLA reporting and audit compliance. If you require extended retention for regulatory purposes, enterprise plans support custom cold-storage archives via S3-compatible endpoints. Data export is available in JSON and CSV formats through the REST API or direct dashboard download.
Need deeper technical details?
Our engineering team maintains comprehensive runbooks and architectural diagrams for all monitoring workflows.
Access implementation guides, webhook payload schemas, and latency benchmark reports. Our support SLA guarantees a technical response within 4 hours for business-critical incidents.
Contact Engineering View API Docs