Skip to content

Check status and report a problem

4 min readAll workspaces

Use the public System status page to check current service health, hourly history, and incident observations, then use Contact support if you need to report a specific problem. The page is read-only, uses UTC times, and can show unknown or stale measurements when current health cannot be confirmed.

Before you start

  • You can open the public status page from the System status link in support; no workspace sign-in, role, or subscription is required to read it.
  • If you want French, use the language selector on the status page.
  • If you plan to contact support, have the affected page, service, time, time zone, and any visible reference ready.

Check current status

  1. Open System status in support. This takes you to the public status website.System status
  2. Review the current banner and the service groups: Platform, Apps & portals, Integrations, and Private AI & analytics.Apps & portalsIntegrationsPlatformPrivate AI & analytics
  3. Use Find a service or integration to search for a service or integration by name or description.Find a service or integration
  4. Use All services or a group filter to narrow the list.All services
  5. If needed, choose History to view 24 hours, 7 days, or 14 days of UTC hourly history.History
  6. Select Refresh status to request a new read of the page.Refresh status

Expected result: you see the latest observed status, the last probe cycle time, and the matching service list for your filter.

Read the hourly report

Open Hourly report on a service to view every UTC bucket for the selected window. The columns are Availability, Coverage, Samples, and Mean probe.

  • Availability is the share of recorded samples that succeeded.
  • Coverage is the share of expected observation opportunities that were actually recorded.
  • No data is different from zero. A missing bucket is unknown, not healthy or failed.
  • The current hour shows elapsed-time coverage only. Earlier unobserved hours stay unknown and are not backfilled.
  • The latest probe latency appears only when available; it is not a promise about customer request latency.

Expected result: you can compare availability and coverage instead of treating a single percentage as full monitoring.

Check incident observations

The Incidents in the last 24 hours section lists automatically detected observations with one-minute resolution.

  • Failure observed means the automated checks saw a failure.
  • Recovery observed means a later successful observation was seen.
  • Monitoring interrupted means the observation stream stopped before recovery was confirmed.

These observations do not establish root cause or exact outage boundaries. No recorded failures during monitoring gaps does not prove availability.

Download the report

  1. Set the History window you want to review.History
  2. Select Download CSV.Download CSV

Expected result: you download a CSV file with the complete loaded hourly measurements for the report, including unknown buckets.

Report a problem to support

  1. On the status page, select Contact support.Contact support
  2. Describe the affected page, tool, or service.
  3. Include the time, time zone, what you expected, what you observed, and any visible reference.
  4. Do not include passwords, access keys, or full bank or card details.

Expected result: the support form opens so you can submit a report about the issue you saw.

If the status page will not confirm health

  • If the page shows current health cannot be confirmed, wait a moment and select Refresh status.
  • If the page shows missing or stale measurements, treat the current result as unknown rather than healthy.
  • If the page itself is unavailable, a previously loaded history may still appear, but current health remains unknown.

If you are dealing with a financial write, do not rerun an uncertain transaction just because the status page looks green. Confirm the workflow result in the product first or contact support with the details above.

Updated

Related guides

We can help

Still need a hand?

Tell us which workspace and page you were using, what you expected, and what happened. Include the time and any request reference shown on screen.

Keep passwords, access tokens and private bank details out of your message. Review screenshots before sharing them.