Description of problem: 

When running a Tekton PipelineRun on OCP 4.15, several TaskRuns fail due to repeated

Operation cannot be fulfilled... the object has been modified

errors when updating finalizers, along with image pull failures for

openshift/python:3.12-ubi9

This causes the PipelineRun to fail intermittently, even though the referenced image should be available in the OpenShift internal registry.

Workaround:

Workaround is not working as it works for other OCP versions with same image. It is only failing with 4.15 OCP version

Prerequisites (if any, like setup, operators/versions):

• OpenShift Container Platform 4.15

• Tekton Pipelines Operator installed (version matching OCP 4.15)

• Internal image registry enabled and accessible

• Pipeline referencing image-registry.openshift-image-registry.svc:5000/openshift/python:3.12-ubi9{}

Steps to Reproduce

 # <steps>

• Deploy a Tekton PipelineRun that references the image image-registry.openshift-image-registry.svc:5000/openshift/python:3.12-ubi9 in one of its steps.
• Ensure the image has not been pre-pulled on the node.
• Observe the PipelineRun execution.

Actual results:

• The PipelineRun fails with logs showing:

  • Failed to update finalizers... the object has been modified; please apply your changes to the latest version and try again

• • ErrImagePull and ImagePullBackOff for openshift/python:3.12-ubi9
• Errors occur in multiple TaskRuns within the PipelineRun.

Expected results:

• PipelineRun should successfully pull the specified image from the OpenShift internal registry and execute without finalizer update conflicts.