Events
Browse cluster events with filters by namespace, type, reason, and involved object.
The Events browser is a dedicated page for triaging Kubernetes events without scoping to a specific resource first.
What's on the page
The events list shows every event the API server is currently exposing for the connected cluster, sorted by last-seen time. Each row includes:
| Column | Description |
|---|---|
| Last seen | Relative time since the most recent occurrence |
| Type | Normal or Warning |
| Reason | The short reason code, e.g. FailedScheduling, BackOff, Pulled |
| Object | The involved object (kind/name) — click through to its detail page |
| Source | Component or node that emitted the event |
| Count | Number of times this event has fired |
| Message | Human-readable explanation |
Filters
Use the filter bar at the top to narrow the list:
- Namespace — scope to a single namespace or view across the whole cluster.
- Type — show only
Warningevents when triaging, orNormalto confirm a successful action. - Reason — pick from a dynamic list of reasons present in the current data.
- Object — search by kind or name to focus on one workload.
Filter state is local to the page; switching pages or clusters resets it.
Live updates
Events stream in as the API server emits them. New rows appear at the top and existing rows update their Count and Last seen in place, so you can leave the page open while reproducing an issue.
Detail panel
Click a row to open the detail panel with the full event payload — useful when you need the source Pod UID, the controller subobject path, or the raw firstTimestamp/lastTimestamp values.
Tips
- Filter to
Warningand the affected namespace right after a deployment to catch scheduler, image pull, or volume mount failures fast. - The cluster overview page also surfaces the most recent events; use this page when you need historical search across the whole cluster.
- The API server evicts events on its own schedule. If you do not see the event you expected, the cluster has already discarded it.