UpHeld
$100.00
Monitoring

Uptime and incidents

Health checks, uptime calculations and outage records for every service.

Services monitored
3
Incidents recorded
0
0 automatic · 0 manual
SLA breaches
0
Downtime this window
0m

Watchdog

Checks every service on a fixed interval. After 3 failed checks in a row, it records the outage. If the SLA is breached, subscriber refunds become claimable.

Stopped · 3 services

0 incidents recorded automatically. Start the watchdog to resume checks.

Every 5s
Each service is checked once.
1 failed check = 5 min
Demo timing. Production checks run every 5 minutes.
3 in a row → record
Records 15 min, then again after every 3 failed checks.
Recovery → close
Records any remaining failed checks and closes the incident.
  • Streamly
    Simulated host · api.streamly.demo/health
    Off0 failed in a row

    Simulated host. Use Simulate outage to fail its checks.

  • CloudBox
    Simulated host · status.cloudbox.demo/health
    Off0 failed in a row

    Simulated host. Use Simulate outage to fail its checks.

  • HyperNode RPC
    Real HTTP check · sepolia.base.org
    Off0 failed in a row

    Checked over HTTP while the watchdog runs.

    Real endpoint
The watchdog runs in your browser while /monitor is open and records outages in the sandbox ledger. Nothing is sent to Base Sepolia. Failed checks that were not yet recorded are dropped when you stop it or leave the page. In production, these records would be signed by an independent probe network. To test a real failure, register a service on /provider with the health endpoint https://httpbin.org/status/503.

Service availability

Check history and current uptime against each SLA target.

  • Streamly
    api.streamly.demo/health · simulated
    30d agoNow
    Operational100.00% / 99.90%
  • CloudBox
    status.cloudbox.demo/health · simulated
    30d agoNow
    Operational100.00% / 99.95%
  • HyperNode RPC
    sepolia.base.org · live
    30d agoNow
    Operational100.00% / 99.99%

Night shift report

Incidents from the last 12 hours, with overnight hours (22:00 to 06:00 UTC) highlighted.

Loading report…
Incidents
0
0 auto · 0 manual
Downtime
0m
0 SLA breaches
Refunds unlocked
$0.00
Claimable by subscribers
Most affected
None
No downtime
Overnight · 22:00 to 06:00 UTCNo incidents overnight.
…Now
Times are in UTC.

Endpoint probes

Send an HTTP request to one endpoint, or check all live endpoints every 15 seconds.

GET /api/monitor/ping
Probe an endpoint

Choose a preset or enter an https URL.

Presets
Streamly and CloudBox run on sample .demo hosts, so their checks are simulated. HyperNode RPC and any public https endpoint registered by a provider receive real requests.
Probe log

Status and latency for every probe in this session.

0
  1. No probes yet. Select Probe now to send a request.

SLA breach calculator

Converts recorded downtime into uptime in basis points, then into a refund tier and a USDC amount.

Calculation for CloudBox

Based on downtime recorded in the current 30-day billing window.

Within SLA
Formula
uptimeBps = floor((43200 − downtime) / 43200 × 10000)
floor((43200 − 0) / 43200 × 10000) = floor(10000.0000) = 10000 bps (100.00%)
SLA check
10000 ≥ 9995 → OK
Tier
Within SLA · 0 bps
Refund per subscriber at current uptime
(10,000,000 × 0) / 10000 = 0 minor units = $0.00 USDC
Tier thresholds: Tier 1 below 99.95% · Tier 2 below 99.00% · Tier 3 below 98.00%. Downtime budget: 21.6 min.

Incident log

Each incident is hashed as keccak256(serviceId, incidentId, downtimeMinutes, resultingUptimeBps). Select Verify to recompute the hash in your browser. Auto incidents come from the watchdog and Manual incidents from the Chaos Theater.

#Service · incidentDowntimeUptimeTierPayoutAttestation hash
No incidents yet. Start the watchdog, or record an outage in the Chaos Theater.