Skip to content

Flaky test: //score/mw/com/test/bigdata/integration_test:test_bigdata_exchange #840

Closed

Description

@github-actions

Flaky test: //score/mw/com/test/bigdata/integration_test:test_bigdata_exchange

Cumulative failed runs observed: 3 (over 300 total runs, 3 nightlies)
First seen: 2026-08-05 · Last seen: 2026-10-07

Per-config cumulative

Config Failed Total
gcc15-integration 2 200
qnx-integration 1 100
{
  "configs": {
    "gcc15-integration": {
      "failed_runs": 2,
      "total_runs": 200
    },
    "qnx-integration": {
      "failed_runs": 1,
      "total_runs": 100
    }
  },
  "cumulative_failed_runs": 3,
  "cumulative_total_runs": 300,
  "first_seen": "2026-08-05",
  "last_seen": "2026-10-07",
  "nightly_count": 3,
  "target": "//score/mw/com/test/bigdata/integration_test:test_bigdata_exchange"
}

Activity

  1. github-actions commented on Aug 5, 2026

    @github-actions
    Author

    Nightly run 30964957957 on 2026-08-05: 1/100 runs failed.

    Per config: qnx-integration 1/100.

  2. github-actions commented on Aug 25, 2026

    @github-actions
    Author

    Nightly run 32795042173 on 2026-08-25: 1/100 runs failed.

    Per config: gcc15-integration 1/100.

  3. github-actions commented on Oct 7, 2026

    @github-actions
    Author

    Nightly run 37556206847 on 2026-10-07: 1/100 runs failed.

    Per config: gcc15-integration 1/100.

  4. castler commented on Oct 30, 2026

    @castler
    Contributor

    A failed execution of this flaky test was verified from the CI log to be a QNX guest kernel panic during boot (muon_sched), before the test body ran. SSH failures are a consequence. Tracked in #1513. Only the sampled executions were verified.

  5. castler commented on Nov 3, 2026

    @castler
    Contributor

    Closing as mitigated by itf 0.7.0 (itf#150), now used in communication via #1545 (merged 2026-11-03).

    Root cause: the QNX SDP 8.0.4 guest sporadically kernel-panics during boot (muon_sched@458, about 1 in 2,200 boots; tracked in #1513, vendor report eclipse-score/qnx_sdp#9), which appeared here as SSH/test failures before any test ran. The ITF now detects the dead guest and restarts it (up to 3 boot attempts, --qemu-boot-attempts), so the panic no longer fails the test. Please reopen if a boot failure still fails a test.

    This issue also saw the docker exec output loss, fixed by itf#151 in the same release.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    flaky-testDetected flaky test target

    Type

    No type

    Projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions