# Tekton v1.9.0

2026-01-30 · [Original release notes](https://github.com/tektoncd/pipeline/releases/tag/v1.9.0) · [Full summary](https://ratatosk.io/en/releases/tekton/v1.9.0)

Security 1 · Breaking changes 5 · Important fixes 7 · Enhancements 2 · Dependency updates 30 · Other changes 39

## Check before upgrading

### Defects found after release

- Sidecar-logs result extraction no longer drops all TaskRun results · Fixed in v1.15.0

## Security (1)

### Pipeline parameter defaults resolve references with a 10-level limit

- Affects: PipelineRuns with parameter references in defaults

Previously, defaults containing `$(params.X)` references were copied verbatim, leading to late task pod-creation failures with misleading messages. These references are now resolved, while dependency chains over 10 levels fail with a maximum-recursion-depth error.

> * :bug: Fix parameter resolution for defaults with  references (#9271)

## Breaking changes (5)

### Duplicate protobuf tag for `PriorityClassName` fixed

- Affects: `pod.Template` protobuf serialization of `PriorityClassName` (Inferred)

Previously, `PriorityClassName` shared protobuf tag 7 with `Env`; it now uses tag 8, leaving `Env` on tag 7.

Evidence: “- Changed `PriorityClassName` protobuf tag from 7 to 8” https://github.com/tektoncd/pipeline/pull/9229

> * :bug: Fix duplicated protobuf tag in pod.Template struct (#9229)

### Shortened `v1beta1` CRD descriptions fix installation errors

- Affects: installations applying `v1beta1` CRDs

Applying CRDs with verbose embedded descriptions could fail with a `metadata.annotations: Too long` error. `kubectl explain` now shows less detail; detailed API field descriptions remain available in the Tekton documentation or in `v1` fields.

Evidence: “To resolve "metadata.annotations: Too long" errors during installation, the field descriptions embedded in `v1beta1` CRDs (visible via `kubectl explain`) have been significantly shortened. Users should refer to the [Tekton documentation](https://tekton.dev/docs/) for detailed API field descriptions or use descriptions in v1 fields.” https://github.com/tektoncd/pipeline/pull/9252

> * :hammer: fix: reduce CRD size by shortening verbose descriptions (#9252)

### Corrected streaming-list handling for `PipelineRunList`

- Affects: the v1 or v1beta1 `PipelineRunList` API types

Previously, `items` could be omitted from the JSON and was not marked required in the generated schema, so Kubernetes API streaming-list handling did not correctly interpret these types. Now `items` is always serialized and marked required.

Evidence: “- Removes ',omitempty' from the 'Items' field in the v1 and v1beta1 PipelineRunList structs.” https://github.com/tektoncd/pipeline/pull/9240

> * :hammer: fix(codegen): Correct JSON tags for streaming lists (#9240)

### `ListMeta` JSON tag gains `omitempty` in v1beta1 `ResolutionRequestList`

- Affects: setups using the v1beta1 `ResolutionRequestList` type and its JSON representation (Inferred)

The `ListMeta` field's JSON tag changes from `json:"metadata"` to `json:"metadata,omitempty"`.

Evidence: “- Adds ',omitempty' to the 'ListMeta' field in the v1beta1 ResolutionRequestList struct.” https://github.com/tektoncd/pipeline/pull/9240

> * :hammer: fix(codegen): Correct JSON tags for streaming lists (#9240)

### Custom task API versions now require `group/version` format

- Affects: Pipelines using custom tasks with `taskRef.apiVersion` or `taskSpec.apiVersion`

Previously, invalid values such as `invalid-api-version` could leave PipelineRuns creating unhandled CustomRuns that timed out. Now, Pipeline creation fails immediately with a clear error when either value is invalid.

Evidence: “add validation to ensure taskRef.apiVersion follows the correct format (group/version) when specified. Previously, invalid apiVersion values like 'invalid-api-version' were accepted, causing PipelineRuns to create unhandled CustomRuns that would timeout. Now Pipeline creation fails immediately with a clear error message when an invalid apiVersion is provided.” https://github.com/tektoncd/pipeline/pull/9045

> * :bug: fix: validate taskRef.apiVersion format for custom tasks (#9045)

## Important fixes (7)

### Affinity Assistant ServiceAccount inheritance fixed

- Affects: PipelineRuns with Affinity Assistant pods and a `taskRunTemplate.serviceAccountName` set

Previously, these pods always used the namespace's `default` ServiceAccount, which could lack required SCC permissions and cause workspace-coscheduled PipelineRuns to fail in OpenShift. Now they use the PipelineRun's `taskRunTemplate.serviceAccountName` unless `AffinityAssistantTemplate` specifies an override; if neither provides a value, they use the namespace's `default` ServiceAccount.

> * :sparkles: feat: add ServiceAccount inheritance to Affinity Assistants (#9253)

### Redundant PipelineRun status writes fixed

- Affects: PipelineRuns reconciled with child references written to status

Unordered child-reference arrays caused massive repeated status writes, particularly for PipelineRuns with more than 64 child TaskRuns, and were associated with apiserver memory spikes and high write QPS. Child references are now consistently ordered, mitigating the redundant writes and their associated apiserver load and stability impact.

> * :bug: fix(pipelinerun): fix the issue of massive invalid status updates caused by unordered arrays, which will greatly impact the resource load and stability of the apiserver. (#9295)

### Pipeline parameter defaults now resolve references

- Affects: PipelineSpecs with parameter defaults that reference other parameters

Previously, the literal reference could reach a task and cause pod creation to fail. Now the task receives the resolved parameter value, allowing the pod to be created successfully.

> * :bug: Fix parameter resolution for defaults with  references (#9271)

### Circular parameter dependencies detected at PipelineRun creation

- Affects: PipelineRuns with circular references between parameter defaults

These cycles previously surfaced during task pod creation as misleading errors. They now fail validation when the PipelineRun is created, with an error identifying the circular dependency.

> * :bug: Fix parameter resolution for defaults with  references (#9271)

### Undefined PipelineRun parameter references caught during validation

- Affects: PipelineRun parameter defaults

When a default referenced an undefined parameter, resolution did not catch it, and a misleading error could appear later during task pod creation. Now validation fails with an error identifying the undefined reference.

> * :bug: Fix parameter resolution for defaults with  references (#9271)

### Excessive reconciliation for `TaskRuns` and `PipelineRuns` without timeouts fixed

- Affects: `TaskRuns` and `PipelineRuns` with no configured timeout: timeout set to `0`, or unset while `default-timeout-minutes` is `"0"`

Previously, these runs were requeued immediately, causing excessive reconciliation. With no timeout, reconciliation now relies on watch events; configured task- or finally-specific timeouts still trigger a requeue for their timeout.

> * :bug: fix: Prevent excessive reconciliation when timeout disabled (#9202)

### TaskRun pod container errors now fail fast

- Affects: TaskRuns with pod containers configured to use ConfigMaps or Secrets through `valueFrom`

When pod containers had configuration or container-creation errors, including missing ConfigMaps or Secrets referenced through `valueFrom`, TaskRuns could wait and report only a generic timeout. They now fail as soon as the error is detected, with actionable errors such as `CreateContainerConfigError` or `PodCreationFailed`.

> * :bug: fix: Detect pod configuration errors early instead of timeout (#9197)

## Enhancements (2)

- User namespace selection in TaskRun and PipelineRun PodTemplates (To use: The Kubernetes feature gate `UserNamespacesSupport` must be enabled to use this field.)
- HTTP resolver digest verification added (To use: `enable-http-resolver` feature flag, enabled by default)

## Dependency updates (30)

30 more, none mentioning security

- `* :hammer: build(deps): bump go.uber.org/zap from 1.27.0 to 1.27.1 (#9333)`
- `* :hammer: build(deps): bump github.com/google/cel-go from 0.26.0 to 0.27.0 (#9330)`
- `* :hammer: build(deps): bump golang.org/x/sync from 0.18.0 to 0.19.0 (#9293)`
- `* :hammer: build(deps): bump github.com/hashicorp/go-version from 1.7.0 to 1.8.0 (#9288)`
- `* :hammer: build(deps): bump k8s.io/apiextensions-apiserver from 0.32.8 to 0.32.11 (#9286)`
- `* :hammer: build(deps): bump github.com/spiffe/spire-api-sdk from 1.12.4 to 1.14.0 (#9261)`
- `* :hammer: build(deps): bump go.opentelemetry.io/otel/sdk from 1.38.0 to 1.39.0 (#9259)`
- `* :hammer: build(deps): bump github.com/cloudevents/sdk-go/v2 from 2.16.1 to 2.16.2 (#9258)`
- `* :hammer: build(deps): bump k8s.io/client-go from 0.32.8 to 0.32.11 (#9256)`
- `* :hammer: build(deps): bump google.golang.org/protobuf from 1.36.10 to 1.36.11 (#9254)`
- `* :hammer: build(deps): bump github/codeql-action from 4.31.8 to 4.31.9 (#9247)`
- `* :hammer: build(deps): bump the all group in /tekton with 2 updates (#9246)`
- `* :hammer: build(deps): bump go.opentelemetry.io/otel/trace from 1.37.0 to 1.39.0 (#9245)`
- `* :hammer: build(deps): bump k8s.io/apimachinery from 0.33.3 to 0.33.7 (#9244)`
- `* :hammer: build(deps): bump the all group in /tekton with 4 updates (#9236)`
- `* :hammer: build(deps): bump step-security/harden-runner from 2.13.3 to 2.14.0 (#9235)`
- `* :hammer: build(deps): bump github/codeql-action from 4.31.7 to 4.31.8 (#9234)`
- `* :hammer: build(deps): bump actions/cache from 4.3.0 to 5.0.1 (#9233)`
- `* :hammer: build(deps): bump tj-actions/changed-files from abdd2f68ea150cee8f236d4a9fb4e0f2491abf1b to e0021407031f5be11a464abee9a0776171c79891 (#9232)`
- `* :hammer: build(deps): bump actions/upload-artifact from 5.0.0 to 6.0.0 (#9231)`
- `* :hammer: build(deps): bump the all group across 1 directory with 4 updates (#9220)`
- `* :hammer: build(deps): bump actions/checkout from 6.0.0 to 6.0.1 (#9219)`
- `* :hammer: build(deps): bump golangci/golangci-lint-action from 9.1.0 to 9.2.0 (#9218)`
- `* :hammer: build(deps): bump peter-evans/slash-command-dispatch from 4.0.0 to 5.0.1 (#9217)`
- `* :hammer: build(deps): bump step-security/harden-runner from 2.13.2 to 2.13.3 (#9216)`
- `* :hammer: build(deps): bump github/codeql-action from 4.31.6 to 4.31.7 (#9215)`
- `* :hammer: build(deps): bump actions/github-script from 7.0.1 to 8.0.0 (#9195)`
- `* :hammer: build(deps): bump github/codeql-action from 4.31.5 to 4.31.6 (#9194)`
- `* :hammer: build(deps): bump github.com/jenkins-x/go-scm from 1.15.4 to 1.15.16 (#9183)`
- `* :hammer: build(deps): bump google.golang.org/grpc from 1.75.0 to 1.77.0 (#9177)`

## Other changes (39)

- TaskRun result-size errors report the actual size
- Cherry-pick workflow handles multi-commit rebase merges
- Two e2e tests moved out of CI temporarily
- Docker-in-Docker sidecar example test probes updated
- GitHub Actions workflows cache ko layer metadata between runs
- Reconciliation and excessive-reconciliation test updates
- GHCR migration notice removed from the README
- Release documentation and cheat sheet revised
- `github/codeql-action` updated to `4.32.0`
- Tekton release-pipeline images updated
- `chainguard-dev/actions` updated to 1.5.13
- Repository automation updates `actions/checkout` to 6.0.2
- Repository automation updates `step-security/harden-runner` to 2.14.1
- Release-pipeline image versions updated
- `chainguard-dev/actions` updated to 1.5.12
- Repository automation update to `actions/setup-go` `6.2.0`
- Repository automation `actions/cache` updated from 5.0.1 to 5.0.2
- Updated `/tekton` release-pipeline images
- Repository automation updates `chainguard-dev/actions` to 1.5.11
- Tekton release pipeline images updated
- Updated `/tekton` release-pipeline images and tools
- `chainguard/go` updated in `/tekton`
- Repository automation updates `peter-evans/slash-command-dispatch` to 5.0.2
- GitHub workflow references now include `# main` comments
- Large-diff handling fixed in the `golangci-lint` GitHub Actions step
- CI workflow file-change detection fixed
- `twoGiants` added to pipeline maintainer aliases
- Cherry-pick-command workflow startup failure fixed
- End-to-end test categorization
- Cherry-pick workflow migrated to the shared plumbing workflow
- go-coverage comment posting fixed
- Automated cherry-pick PRs get descriptive titles and original context
- Invalid `github-script` action SHAs corrected in the `e2e-matrix-extras` workflow
- Examples test parallelism reduced from 6 to 2
- Coverage workflow uses `CHATOPS_TOKEN` for pull request comments
- Copied `chatops_retest` replaced by shared workflow
- YAML indentation fixed in release cheat sheet templates
- Beta-feature notes removed for `projected` and `csi` workspaces
- `releases.md` updated with v1.7 release information

---

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