René's URL Explorer Experiment

{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"EventFilterWindow suppresses mapper events when controller writes through shared InformerEventSource","articleBody":"## Description\n\nWhen a controller patches a secondary resource's status through its own `InformerEventSource` (via `eventFilteringUpdateAndCacheResource`), the `EventFilterWindow` correctly identifies the resulting informer event as an \"own update\" and suppresses it. However, this suppression also prevents the `SecondaryToPrimaryMapper` from ever seeing the updated resource state, which can leave the controller in a dead state.\n\n## Scenario\n\nA controller reconciles a primary resource and has an `InformerEventSource` for a secondary resource type. The mapper uses a generation/observedGeneration guard:\n\n- `generation != observedGeneration` → return empty set (secondary not ready, skip)\n- `generation == observedGeneration` → return primary ResourceID (secondary ready, trigger)\n\nDuring reconciliation the controller patches the secondary's `status.observedGeneration` to match `metadata.generation` through its `InformerEventSource`. The `EventFilterWindow` suppresses the resulting event as an own update. The mapper never sees `gen == obsGen`, so the controller is never re-triggered.\n\n## Expected behaviour\n\nThe status-change event should reach the mapper so it can evaluate the new state and trigger reconciliation of the primary.\n\n## Actual behaviour\n\nThe `EventFilterWindow` suppresses the event because its resource version matches `ownUpdateVersions`. The mapper never fires. The controller enters a dead state for that secondary resource.\n\n## Root cause\n\n`ManagedInformerEventSource.eventFilteringUpdateAndCacheResource` (line 96) opens a filter window, performs the update, records the response resource version as an \"own\" version, then closes the window. When the informer event arrives with that resource version, `EventFilterWindow.eventForRangeAndClear` (line 183) sees `events.keySet().equals(ownResourceVersions)` and returns `Optional.empty()` — the event is swallowed before it reaches `propagateEvent` and the mapper.\n\n## Impact\n\nThis affects any controller that:\n1. Writes to a secondary resource through its own `InformerEventSource` (directly or via a `KubernetesDependentResource` sharing the event source with `useEventSourceWithName`)\n2. Has a mapper that depends on seeing the result of that write\n\nThe pattern is common in controllers that manage secondary resources and use generation/observedGeneration guards to sequence reconciliation.\n\n## Reproducer\n\nBranch: `reproducer/status-event-filtering` — see `StatusEventFilteringIT` in `operator-framework/src/test/java/.../baseapi/statuseventfiltering/`\n\n## Downstream issue\n\nhttps://github.com/kroxylicious/kroxylicious/issues/4241\n\nHappy to look at a fix if the maintainers agree with the analysis.","author":{"url":"https://github.com/SamBarker","@type":"Person","name":"SamBarker"},"datePublished":"2026-06-29T05:35:01.000Z","interactionStatistic":{"@type":"InteractionCounter","interactionType":"https://schema.org/CommentAction","userInteractionCount":3},"url":"https://github.com/3445/java-operator-sdk/issues/3445"}