Status
Loreflow publishes uptime for every component of the platform. Historical incidents stay online permanently — including the postmortems. The figures on this page are sample data showing how it will read; the production monitoring that populates them is not live yet.
Each component is probed every 30 seconds from three regions. A component is only marked healthy when all three probes pass. Sample readings:
We post an initial acknowledgement within 15 minutes and a full postmortem within five business days for every Sev-1 and Sev-2 incident. Example entries, showing the detail every incident will carry:
Planned maintenance is announced at least 72 hours ahead and scheduled for Saturdays 02:00–04:00 UTC, when workspace activity is lowest.
Right now
Once monitoring is live, this panel updates every minute from the same monitors that page our on-call engineer. The values below are sample data.
0.00%
Uptime, 90 days
Weighted across API, sync and search.
0 ms
Median API latency
Measured at the edge, not in our own network.
0 min
Sync interval
Per source, with nightly full re-index.
0
Open incidents
Last incident resolved 63 days ago.
We count degradations against ourselves the same way outages are counted, and report them at this level of detail. The sample window below shows two brief degradations in the sync workers, resolved before most workspaces would notice, with search and API up throughout.
API latency
Sample series — median response time, by week.
Incident density
Sample — incident density by week.
Components
Uptime by component
Sample — trailing 90 days.
Where the load is
Sample — request volume by surface.
When something breaks
Detect
Automated monitors page on-call within 60 seconds of an error-rate or latency threshold being crossed. We rarely learn about incidents from customers.
< 1 minute
Publish
The status page updates before we start debugging. You should never have to ask whether it is us or you.
< 5 minutes
Resolve and follow up
A written post-mortem within five working days for anything customer-visible, including what we changed so it does not repeat.
Post-mortem in 5 days
Subscribe to incident notifications and you will hear from us the moment a monitor trips.
The app, the sync pipeline and each connector family, checked continuously with per-component history. The figures shown on the status page today are sample data.
No. Syncs resume from the last confirmed checkpoint and backfill whatever happened during the outage.
Posted here as they develop, with an in-app banner for anything affecting sync freshness, and a written follow-up afterwards.