# Tekton v1.10.0

2026-02-27 · [Original release notes](https://github.com/tektoncd/pipeline/releases/tag/v1.10.0) · [Full summary](https://ratatosk.io/en/releases/tekton/v1.10.0)

Breaking changes 4 · Deprecations 1 · Important fixes 3 · Enhancements 1 · Dependency updates 23 · Other changes 37

## Check before upgrading

### Prerequisites

- **Observability metrics destination setting updated**: The `config-observability` ConfigMap should use `metrics-protocol` set to `prometheus`, `grpc`, or `http` instead of `metrics.backend-destination`. No change is needed if Prometheus was already in use.
  >   Upgrade actions:
  >   1. **Update Config**: Ensure your `config-observability` ConfigMap uses `metrics-protocol: prometheus` (or `grpc`/`http`) instead of the old `metrics.backend-destination`. If prometheus was already being used, no changes are needed.

### Upgrade guide

- **Full metrics migration table**
  >   See the [full migration table in PR #9043](https://github.com/tektoncd/pipeline/pull/9043) for complete details.

## Breaking changes (4)

### Tekton infrastructure metrics renamed

- Affects: setups that collect or query Tekton infrastructure metrics, including Go runtime, Workqueue, or Kubernetes client metrics
- Action required: Update `config-observability` to use `metrics-protocol: prometheus` (or `grpc`/`http`) instead of `metrics.backend-destination`, unless Prometheus was already in use; update dashboard queries to the new metric names and check `pipelinerun_duration_seconds` aggregations for the added `reason` label.

These metrics previously used the `tekton_pipelines_controller_` prefix, such as `tekton_pipelines_controller_workqueue_adds_total`; they are now emitted under names such as `kn_workqueue_adds_total`, `http_client_request_duration_seconds`, and `go_*`.

> * :rotating_light: **Metrics migration from OpenCensus to OpenTelemetry** (#9043)
>   **ACTION REQUIRED**: Infrastructure metrics (Go runtime, Workqueue, K8s Client) have been renamed from the `tekton_pipelines_controller_` prefix to standard OpenTelemetry/Knative namespaces. The `reason` label has been added to duration metrics (`pipelinerun_duration_seconds`, `taskrun_duration_seconds`). The `reconcile_count` and `reconcile_latency` metrics have been removed.

### Tekton controller reconcile metrics removed

- Affects: controller-performance monitoring using either custom reconcile metric

The pull request points to `kn_workqueue_process_duration_seconds` and `kn_workqueue_adds_total` for controller-performance monitoring.

Evidence: “3. Removed Metrics * `tekton_pipelines_controller_reconcile_count` and `tekton_pipelines_controller_reconcile_latency`: These custom reconcile metrics are no longer emitted. Users should rely on kn_workqueue_process_duration_seconds and kn_workqueue_adds_total to monitor controller performance.” https://github.com/tektoncd/pipeline/pull/9043

>   **ACTION REQUIRED**: Infrastructure metrics (Go runtime, Workqueue, K8s Client) have been renamed from the `tekton_pipelines_controller_` prefix to standard OpenTelemetry/Knative namespaces. The `reason` label has been added to duration metrics (`pipelinerun_duration_seconds`, `taskrun_duration_seconds`). The `reconcile_count` and `reconcile_latency` metrics have been removed.

### PipelineRun and TaskRun metrics migrate to OpenTelemetry

- Affects: setups collecting or querying PipelineRun, TaskRun, or infrastructure metrics (Inferred)
- Action required: The `config-observability` ConfigMap uses `metrics-protocol: prometheus` (or `grpc/http`) instead of `metrics.backend-destination`, unless Prometheus was already in use; dashboards use `kn_workqueue_*` and `go_*` in place of the old names, and duration aggregations account for the `reason` label.

Previously, infrastructure metrics used the `tekton_pipelines_controller_` prefix, duration metrics lacked the `reason` label, and the two custom reconcile metrics were emitted. Infrastructure metrics now use standard OpenTelemetry and Knative names, including `kn_workqueue_*` and `go_*`; duration metrics gain the `reason` label, and the custom reconcile metrics are no longer emitted.

Evidence: “* `tekton_pipelines_controller_reconcile_count` and `tekton_pipelines_controller_reconcile_latency`: These custom reconcile metrics are no longer emitted. Users should rely on” https://github.com/tektoncd/pipeline/pull/9043

> * :sparkles: feat(metrics): Migrate from OpenCensus to OpenTelemetry (#9043)
>   Migrated PipelineRun and TaskRun metrics to OpenTelemetry instruments (histograms, counters, gauges). Updated Knative to 1.19. See [Upgrade Notices](#upgrade-notices) for breaking changes and required actions.

### Metrics configuration switches to `metrics-protocol`

- Affects: metrics settings in the `config-observability` ConfigMap

The prior `metrics.backend-destination` setting is replaced by `metrics-protocol` with `prometheus`, `grpc`, or `http`; installations already using Prometheus need no configuration change. Infrastructure metrics now use standard OpenTelemetry/Knative namespaces instead of the `tekton_pipelines_controller_` prefix, and `reconcile_count` and `reconcile_latency` are removed.

Evidence: “1. **Update Config**: Ensure your `config-observability` ConfigMap uses `metrics-protocol: prometheus` (or `grpc`/`http`) instead of the old `metrics.backend-destination`. If prometheus was already being used, no changes are needed.” https://github.com/tektoncd/pipeline/releases/tag/v1.10.0

>   Migrated PipelineRun and TaskRun metrics to OpenTelemetry instruments (histograms, counters, gauges). Updated Knative to 1.19. See [Upgrade Notices](#upgrade-notices) for breaking changes and required actions.

## Deprecations (1)

### Unused `featureFlags.EnableTektonOCIBundles` field accepted for compatibility

- Affects: PipelineRun and TaskRun resources containing `featureFlags.EnableTektonOCIBundles` (Inferred)
- Removed in: not stated

Previously, `EnableTektonOCIBundles` was treated as an unknown field. It is now accepted but has no effect, and is retained to allow deletion of PipelineRuns created before v0.62.x.

> * :bug: accept featureFlags.EnableTektonOCIBundles to fix unknown field error (#8996)

## Important fixes (3)

### Redundant `ResolutionRequest` short names removed

- Affects: installations using the `ResolutionRequest` CRD on Kubernetes 1.33+

On Kubernetes 1.33+, the redundant short names caused a `ShortNamesConflict` condition on the CRD.

> * :bug: fix: Remove redundant shortNames from ResolutionRequest CRD (#9398)
>   Remove redundant shortNames from ResolutionRequest CRD that caused ShortNamesConflict on Kubernetes 1.33+

### PipelineRun validation accepts non-parameter variables in defaults

- Affects: PipelineRuns with non-parameter variables in pipeline parameter defaults

Previously, validation failed when a pipeline parameter default referenced a value such as `$(context.pipelineRun.name)`. Such defaults now validate without that failure.

> * :bug: fix(pipelines): allow pipeline param defaults to use non-param variables (#9386)
>   Fixed a bug which caused PipelineRun validation to fail when a pipeline parameter's default value referenced a non-parameter variable (e.g. `$(context.pipelineRun.name)`)

### Pipeline-level results include outputs from unsuccessful tasks

- Affects: PipelineRuns with pipeline-level results referencing outputs from failed, cancelled, or timed-out tasks

Previously, these references could remain unresolved variable strings. Results from completed failed, cancelled, or timed-out tasks are now included in pipeline-level results.

> * :bug: fix: pipeline-level results not recorded from failed tasks (#9367)
>   Pipeline-level results now include results from failed, cancelled, and timed-out tasks, fixing cases where results referencing non-successful task outputs were left as unresolved variable strings.

## Enhancements (1)

- Git resolver recognizes SHA-256 commit hashes

## Dependency updates (23)

23 more, none mentioning security

- `Migrated PipelineRun and TaskRun metrics to OpenTelemetry instruments (histograms, counters, gauges). Updated Knative to 1.19. See [Upgrade Notices](#upgrade-notices) for breaking changes and required actions.`
- `* :hammer: build(deps): bump golang.org/x/crypto from 0.36.0 to 0.45.0 in /test/resolver-with-timeout (#9426)`
- `* :hammer: build(deps): bump opentelemetry exporter packages to v1.39.0 (#9332)`
- `* :hammer: build(deps): bump github.com/google/go-containerregistry from 0.21.0 to 0.21.1 (#9433)`
- `* :hammer: build(deps): bump github.com/google/go-containerregistry from 0.20.7 to 0.21.0 (#9418)`
- `* :hammer: build(deps): bump google.golang.org/grpc from 1.79.0 to 1.79.1 (#9392)`
- `* :hammer: build(deps): bump github.com/jenkins-x/go-scm from 1.15.16 to 1.15.17 (#9391)`
- `* :hammer: build(deps): bump google.golang.org/grpc from 1.78.0 to 1.79.0 (#9389)`
- `* :hammer: build(deps): bump k8s.io/code-generator from 0.32.11 to 0.32.12 (#9388)`
- `* :hammer: build(deps): bump k8s.io/apiextensions-apiserver from 0.32.11 to 0.32.12 (#9385)`
- `* :hammer: build(deps): bump k8s.io/apimachinery from 0.33.7 to 0.33.8 (#9384)`
- `* :hammer: build(deps): bump k8s.io/client-go from 0.32.11 to 0.32.12 (#9383)`
- `* :hammer: build(deps): bump github/codeql-action from 4.32.1 to 4.32.2 (#9374)`
- `* :hammer: build(deps): bump the all group in /tekton with 4 updates (#9373)`
- `* :hammer: build(deps): bump step-security/harden-runner from 2.14.1 to 2.14.2 (#9372)`
- `* :hammer: build(deps): bump tj-actions/changed-files from 47.0.1 to 47.0.2 (#9371)`
- `* :hammer: build(deps): bump chainguard-dev/actions from 1.5.14 to 1.5.16 (#9370)`
- `* :hammer: build(deps): bump golang.org/x/crypto from 0.47.0 to 0.48.0 (#9369)`
- `* :hammer: build(deps): bump go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp from 1.39.0 to 1.40.0 (#9363)`
- `* :hammer: build(deps): bump go.opentelemetry.io/otel/exporters/otlp/otlptrace/otlptracehttp from 1.37.0 to 1.40.0 (#9345)`
- `* :hammer: build(deps): bump google.golang.org/grpc from 1.77.0 to 1.78.0 (#9337)`
- `* :hammer: build(deps): bump github.com/spiffe/spire-api-sdk from 1.14.0 to 1.14.1 (#9336)`
- `* :hammer: build(deps): bump sigstore/sigstore from 1.9.5 to 1.10.4 (#9331)`

## Other changes (37)

- PipelineRun and TaskRun duration metrics gain a `reason` label
- `/rebase` pull request workflow added
- Unified CI summary job replaces the e2e-only fan-in job
- Resolver cache adopts framework map-based configuration handling
- High-severity GoSec findings assessed with added `nosec` comments
- Cherry-pick reruns reset and force-push the existing PR branch
- Cherry-pick workflow fetches fork pull request commits
- `github/codeql-action` dependency updated to 4.32.4
- Updated the pinned `chainguard/crane` image in `/tekton`
- Pinned `chainguard/go` image updated
- Updated `tektoncd/plumbing/ko` pin in `/tekton`
- Updated `tektoncd/plumbing/koparse` pin in `/tekton`
- Updated `tj-actions/changed-files` dependency from 47.0.2 to 47.0.4
- `actions/dependency-review-action` updated from 4.8.2 to 4.8.3
- Updated `chainguard-dev/actions` from 1.6.1 to 1.6.4
- Sigstore dependency updated in the timeout resolver test
- `github.com/tektoncd/pipeline` updated to 1.9.1 in `/test/custom-task-ctrls/wait-task-beta`
- `chainguard/crane` image updated in `/tekton`
- `chainguard/go` image digest updated in `/tekton`
- Updated the `tektoncd/plumbing/ko` image pin
- Updated `tektoncd/plumbing/koparse` image pin in `/tekton`
- `github/codeql-action` updated to 4.32.3
- `chainguard-dev/actions` updated to 1.6.1
- `k8s.io/client-go` dependency updated in wait-task-beta tests
- `k8s.io/api` updated to 0.32.12 in `/test/custom-task-ctrls/wait-task-beta`
- Updated `k8s.io/apimachinery` dependency in wait-task-beta
- Simplified e2e test health status result
- Images used in `tekton/publish.yaml` updated
- `chainguard-dev/actions` updated from 1.5.13 to 1.5.14
- `github/codeql-action` updated to 4.32.1
- `actions/cache` updated to 5.0.3
- `github.com/tektoncd/pipeline` dependency updated from 1.7.0 to 1.9.0
- wait-task-beta updated to `github.com/tektoncd/pipeline` v1.7.0
- Controller flag and environment-variable documentation clarified and corrected
- `releases.md` release lifecycle listings updated
- `roadmap.md` updated with GitHub Projects roadmap guidance
- PipelineRun and workspace documentation examples updated

---

Summaries are produced by AI, so they can be wrong, and they are provided without warranty.
