Mismatch between created PVC and its reference in the job pod

Author: ElSamhaaCreated Sep 15, 2026Updated Sep 15, 2026

Labelsbug

Component

cli

Describe the bug

Executing a test pipeline through CLI and k8s backend generates mismatched resource names. For instance the below command

KUBECONFIG=~/.kube/myconfig woodpecker-cli exec --log-level debug --pipeline-event push --commit-branch temp --commit-sha "$(git rev-parse HEAD)" --plugins-privileged woodpeckerci/plugin-docker-buildx --backend-engine kubernetes --backend-k8s-storage-class my-rwx-sc --backend-k8s-storage-rwx=true --backend-k8s-volume-size=1Gi ./.woodpecker/build.yaml

gets stuck at

6:24AM INF go.woodpecker-ci.org/woodpecker/v3/cli/exec/exec.go:109 > auto detected workflow in multi workflow setup, parent directory is used as workspace
6:24AM DBG go.woodpecker-ci.org/woodpecker/v3/cli/exec/exec.go:116 > auto selected workspace folder at "/home/myuser/myproject"
6:24AM DBG go.woodpecker-ci.org/woodpecker/v3/cli/common/hooks.go:52 > checking for updates ...
6:24AM DBG go.woodpecker-ci.org/woodpecker/v3/cli/update/updater.go:39 > current version: 3.18.1
# build
6:24AM DBG go.woodpecker-ci.org/woodpecker/v3/pipeline/runtime/workflow.go:113 > executing 1 stages, in order of: CLI=exec
6:24AM DBG go.woodpecker-ci.org/woodpecker/v3/pipeline/runtime/workflow.go:122 > stage CLI=exec StagePos=0 Steps=test
6:24AM DBG go.woodpecker-ci.org/woodpecker/v3/pipeline/runtime/step.go:36 > prepare CLI=exec step=test
6:24AM DBG go.woodpecker-ci.org/woodpecker/v3/pipeline/runtime/step.go:54 > executing CLI=exec step=test
6:24AM DBG go.woodpecker-ci.org/woodpecker/v3/pipeline/backend/kubernetes/secrets.go:59 > Secret names were defined in backend options, but secret access is disallowed by instance configuration.
6:24AM DBG go.woodpecker-ci.org/woodpecker/v3/cli/update/updater.go:78 > no new version available
6:24AM DBG go.woodpecker-ci.org/woodpecker/v3/cli/common/hooks.go:63 > no update required

and is blocked from scheduling due to

  Warning  FailedScheduling   63s   default-scheduler   0/2 nodes are available: persistentvolumeclaim "wp-01m2hhg5xxfm4tm5kykttn7mtm-default" not found. preemption: 0/2 nodes are available: 2 Preemption is not helpful for scheduling., ,
  Normal   NotTriggerScaleUp  40s   cluster-autoscaler  pod didn't trigger scale-up (it wouldn't fit if a new node is added):

but a successful provision of a PVC named wp-01m2hhg5y2gxv7115yd1n6tdhy-0-default happens without issues.

I'm testing a simple workflow exec

when:
- event: push
  branch: temp

steps:
- name: test
  image: debian
  commands:
  - env
  backend_options:
    kubernetes:
      secrets:
      - name: mysecret
        key: mykey
        target:
          env: MYKEY

Steps to reproduce

do a cli exec with k8s backend of the supplied test workflow
mismatched PVC names between actually-provisioned PVC and what is referenced in the pod manifest

Expected behavior

The referenced PVC in the job pod should match the created PVC. This issue is reproducible consistently at my end.

System Info

bash

Helm chart: 3.7.3
WP server: 3.18.1
WP agent: 3.18.1
CLI: 3.18.1
CLI installed through homebrew on Linux arm64

Additional context

No response

Validations

Read the docs.
Check that there isn't already an issue that reports the same bug to avoid creating a duplicate.
Checked that the bug isn't fixed in the next version already [https://woodpecker-ci.org/versions]

Source: woodpecker-ci/woodpecker

View original on GitHub View discussion on GitHub