Skip to content

fix(tests): flaky integration tests test_cp_restart_nodes_reconnect and test_subscription_with_static_link #1692

Description

@Tehsmash

Description

Two control plane integration tests are failing intermittently in the ci-publish-prokect-coverage workflow (coverage job). These appear to be flaky/timing-sensitive tests.

Failing tests:

  • test_cp_restart_nodes_reconnect
  • test_subscription_with_static_link

Observed failure

Seen on PR #1668 in run https://github.com/agntcy/slim/actions/runs/26625998999/job/78462771745

thread 'test_subscription_with_static_link' (23389) panicked at control-plane/tests/integration_test.rs:1506:10:
no link found from node-a to node-b
test result: FAILED. 13 passed; 2 failed; 1 ignored; 0 measured; 0 filtered out; finished in 3.89s

error: test failed, to rerun pass `-p agntcy-slim-control-plane --test integration_test`

Root cause hypothesis

The test test_subscription_with_static_link is asserting that a link from node-a to node-b exists, but it is not present at the time of the assertion. This suggests a timing/race condition where the control plane has not fully propagated the link state before the test checks for it.

Impact

This failure blocks the coverage report from being published. Other CI checks all pass.

Steps to reproduce

Run the full integration test suite with coverage enabled:

task data-plane:test:coverage

The failure is non-deterministic (flaky).

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    Status
    Backlog

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions