[updatecli] Update latest snapshot to 9.6.0-e52b5e0b

[github-actions]

Bump latest 9.x-SNAPSHOT elastic-stack test version

[updatecli] Update latest snapshot to 9.6.0-e52b5e0b

1 file(s) updated with "$1 9.6.0-e52b5e0b-SNAPSHOT": * Makefile

GitHub Action workflow link

---

Created automatically by Updatecli Options: Most of Updatecli configuration is done via its manifest(s). If you close this pull request, Updatecli will automatically reopen it, the next time it runs. If you close this pull request and delete the base branch, Updatecli will automatically recreate it, erasing all previous commits made. Feel free to report any issues at github.com/updatecli/updatecli. If you find this tool useful, do not hesitate to star our GitHub repository as a sign of appreciation, and/or to tell us directly on our chat!

[infra-vault-gh-plugin-prod]

💛 Build succeeded, but was flaky

• Buildkite Build
• Commit: 7c8038f

Failed CI Steps

• :go: Integration test: sql_input

History

[github-actions]

TL;DR

The :go: Integration test: sql_input Buildkite step failed, but the prefetched log starts after the actual failure and only shows cleanup/teardown, so the specific failing assertion is not present yet. Immediate action: inspect the uploaded sql_input-system-*.xml test artifact (or rerun with full logs) to extract the first real error line.

Remediation

• Download and inspect build/test-results/sql_input-system-*.xml from build artifacts to capture the exact failing system test and assertion message.
• If the XML confirms an intermittent external/service startup error, retry the Buildkite step; if it shows a deterministic assertion, patch the referenced test/package path and rerun test-check-packages-parallel -p sql_input.
• Keep the detective signal anchored on the first failure line (before cleanup), not the wrapper exit code.

Investigation details

Root Cause

Current evidence is inconclusive for a code-level root cause because the available log excerpt contains only post-failure cleanup output and wrapper exits. This is most consistent with an infrastructure/observability gap in the provided artifact, not a verified regression in this PR.

Relevant code path for this job:

• .buildkite/scripts/integration_tests.sh:175 runs make SERVERLESS=... PACKAGE_UNDER_TEST=... <target>
• .buildkite/scripts/integration_tests.sh:184 emits the generic wrapper failure line
• Makefile:128 maps test-check-packages-parallel to ./scripts/test-check-packages.sh

PR scope check:

• The PR changes only Makefile (test-stack-command-9x version bump), while the failing target is test-check-packages-parallel; no direct sql_input package/test files changed in this PR.

Evidence

• Build: https://buildkite.com/elastic/elastic-package/builds/8791
• Job/step: :go: Integration test: sql_input
• Key log excerpt:
  • /tmp/gh-aw/buildkite-logs/elastic-package-go-integration-test-sql_input.txt:30 → make: *** [Makefile:128: test-check-packages-parallel] Error 1
  • /tmp/gh-aw/buildkite-logs/elastic-package-go-integration-test-sql_input.txt:101 → make SERVERLESS=false PACKAGE_UNDER_TEST=sql_input test-check-packages-parallel failed with 2
  • /tmp/gh-aw/buildkite-logs/elastic-package-go-integration-test-sql_input.txt:115-137 shows uploaded xUnit artifacts including sql_input-system-...xml (the likely source of the concrete assertion detail).

Flaky issue cross-check:

• Searched repo:elastic/elastic-package label:flaky-test sql_input → no matching issue found.

Verification

• Local reproduction not run in this workflow because the provided CI artifact does not include the failing assertion output and this environment does not provide Buildkite artifact download context for the hidden logs.

Follow-up

• Once sql_input-system-*.xml is available in-thread, I can map the exact failing test case back to source (test/packages/parallel/sql_input/_dev/test/system/*.yml) and provide a precise patch recommendation.

---

What is this? | From workflow: Observability Agentic Workflow — Status

Give us feedback! React with 🚀 if perfect, 👍 if helpful, 👎 if not.