Skip to content

Bug: Test Harness fails to capture error logs for transient failures (Status 598 result in empty ARS objects) #44

@mohsenht

Description

@mohsenht

During a Pathfinder test run, a test case failed and reported the following result state:

"result": { "shepherd-bte": { "status": "FAILED", "message": "Status code: 500" }, "shepherd-arax": { "status": "FAILED", "message": "Status code: 598" }, "shepherd-aragorn": { "status": "FAILED", "message": "Status code: 598" }, "ars": { "status": "NO_RESULTS", "message": "No results" } }

However, when pulling this PK from the ARS endpoint, it returns an empty object ({}) rather than a 404 or an error log.

Currently, there is a lack of observability in the Test Harness when tests fail due to transient infrastructure errors (e.g., Status code 598). When these failures occur, the Test Harness records a Primary Key (PK) for the failure, but querying that PK in the ARS returns a completely empty object.

Because the query itself is valid and runs successfully when executed manually, this lack of logging creates a blind spot that makes it impossible to debug transient errors during automated test runs. (Confirmed as a Test Harness bug via Slack discussion with @maximusunc).

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions