Service status

Measured availability of the Isto Intents forwarder and of every server listed in the registry, over the last 30 and 90 days. These are measurements, not a promise: Isto Intents offers no SLA.

Day by day#

every check passed · some failed · all failed · not measured

Incidents, last 90 days#

• Loading…

How it is measured#

• Forwarder (https://istointents.com/mcp): every 5 minutes the registry calls tools/list. A check passes when the answer is HTTP 200 within 10 seconds and lists the four tools of the consent-link profile. The response time is that call's.
• Listed servers: the registry's own automated check (how it works), every 6 hours per host. A check passes when its outcome is ok. The response time covers the whole check (initialize, notifications/initialized, tools/list).
• Availability is the share of passed checks, rounded down to a tenth of a per cent. A run of failed checks is one incident, from the first failure to the next passed check.
• Both are measured by the registry Worker on Cloudflare. An outage of Cloudflare itself is not visible on this page, because it takes the measuring Worker down too. A separate check outside Cloudflare (GitHub Actions) calls the forwarder twice a day and alerts the operator.
• Only counts are stored — per day and service: checks, passed checks, response times — and a row per incident. Nothing about who uses the services.

The same data as JSON: https://istointents.org/status.json (refreshed every 5 minutes).