Monitor Details & Runs
Click a monitor in the list to open its details: its state, uptime, response times, every run and every outage.
Header and Actions
The header shows the monitor's name, state, type, service label and version, then its target, interval and timeout. A CI chip shows when the monitor is non-blocking or skipped in CI, and a clock badge when it is outside its active hours.
| Action | Description |
|---|---|
| Run now | Runs the monitor once and shows the result inline: OK or the error, the duration and the HTTP status |
| Pause / Resume | Stops or restarts scheduled runs and alerts |
| Open in Explore | Opens the monitor's results in the Data Explorer |
| Access | Who else can view or edit the monitor (editors only) |
| Edit | Opens the monitor editor |
| ⋯ | Duplicate, Download Kubernetes YAML, Download JSON, Copy Kubernetes YAML, Delete |
When the monitor's alert is firing, an alert block at the top links to it.
The time range in the top bar applies to every tab.
Overview Tab
| Tile | Description |
|---|---|
| Uptime | Share of successful runs in the time range |
| Avg latency, p50, p95, p99, Min / max | Response times |
| Checks | Successful and total runs, with the failures by error type |
- Uptime timeline — uptime per time bucket: ≥ 99.9%, 95–99.9%, < 95% or no data.
- Latency — p50 / p95 / p99 over time, or p95 per location.
- Failures by step — for multistep and browser tests, which steps fail and how often.
- Probes — one row per agent: location, uptime, run count, p95 and last state.
Results Tab
Every run of the monitor in the time range: Time, Status (OK / FAIL), Latency, HTTP status, Probe (agent), Error and, for SSL monitors, SSL expiry. A CI chip marks runs from CI batches (they never alert or count toward uptime), and N attempts marks runs that were retried.
Filter with the search bar or the facets (status, location, run type, labels and more). Click a run to open its details; the up and down arrow keys move through the list.
Run Details
| Tab | Shows |
|---|---|
| Result | The type's result: the response, DNS answer, gRPC reply, TCP connection, received message, MCP server reply or network path hops, and the TLS certificate (subject, issuer, SANs, validity, fingerprint, protocol, cipher) |
| Timing | The waterfall: DNS, TCP, TLS, time to first byte and download |
| Assertions | Every assertion, with its condition, expected and actual value, and result |
| Snapshot | Response headers and body |
| Steps | Multistep tests: each step's assertions, extracted values, response and timings. Secure values are masked |
| Journey | Browser tests: each step's screenshot, selector, URL, HTTP status, duration and Web Vitals, plus the Console, Network and Resources tabs and the page's HTML when a step failed |
| Explain | Failed runs: an AI explanation with evidence, likely causes and next steps (needs access to AI chat) |
| Metadata | Result id, agent, attempts, run type, location, response size, and links to the run's Trace and RUM session |
Compare with last success shows this run next to the last successful one.
The Trace link opens the backend trace the request started, when the service is traced by KubeSense. For browser tests whose pages load the KubeSense RUM SDK, the RUM link opens the synthetic session (see Settings → Browser test integrations).
Outages Tab
Every period the monitor was down: Status (Ongoing, Resolved), Severity, Root cause, Failing probes, Started, Duration, Ack and Assignee. Click an outage to acknowledge it, assign it, change its severity, add notes, and see its failed runs and timeline.
Activity Tab
The monitor's alert events — down, recovery, degraded, no data, re-notifications, SSL expiry and flapping — with a filter by event type.
Properties Tab
The monitor's full configuration, read-only: check settings, targeting (locations, agents, agent labels), retries and schedule, the alert policy, maintenance windows in effect, monitors that use it as a subtest, its access and its metadata.
History Tab
Every saved version of the monitor: Version, Kind, When, Author, Change and Changed fields. Open a version to see its definition, compare two versions field by field, or Restore an earlier version (not for GitOps-managed monitors).