test(node-integration): Skip flaky tedious tests#15798
Merged
Merged
Conversation
lforst
approved these changes
Mar 24, 2025
Contributor
|
you could say the tests were ... tedious 👉😎👉 |
mydea
added a commit
that referenced
this pull request
May 18, 2026
…s` (#20961) ## Summary Migrates the remaining integration-specific tracing tests under `dev-packages/node-integration-tests/suites/tracing/` to use the `createEsmAndCjsTests` helper, which auto-runs each scenario in both ESM and CJS mode from a single `.mjs` source. For each folder: - `scenario*.js` → `scenario*.mjs` (ESM imports) - Sentry.init extracted into a separate `instrument.mjs` - `test.ts` switched to `createEsmAndCjsTests(__dirname, scenarioPath, instrumentPath, callback, options?)` - `failsOnEsm: true` added only where CJS passed and ESM legitimately fails (i.e. instrumentation only patches the CJS module shape) ## Migrated folders | Folder | Notes | |---|---| | `lru-memoizer` | `failsOnEsm: true` — OTel `instrumentation-lru-memoizer` only patches the CJS function-shaped export. | | `mongodb` | mongodb v3 is CJS-only and has no named ESM exports — uses `import mongodb from 'mongodb'` + destructure. | | `mongoose` | Straight migration. | | `mysql2` | Straight migration (docker). | | `mysql` | `failsOnEsm: true` on all 3 scenarios — mysql v2 is CJS-only. Three scenarios kept (`withConnect`, `withoutCallback`, `withoutConnect`). | | `postgres` | 3 scenarios. `ignoreConnectSpans` reuses the default scenario but with a separate `instrument-ignoreConnect.mjs`. `pg-native` scenario migrated as-is (still relies on local `setupCommand: 'yarn'` to build the native bindings). | | `postgresjs` | Consolidated previously-parallel `.cjs`/`.mjs` scenarios into a single `.mjs` each. The `wait-for-postgres.js` helper is still required from the scenarios via `createRequire(import.meta.url)` (revert of the .mjs-helper conversion approach — see notes below). The error stacktrace assertion uses `expect.stringMatching(/postgres(\.cjs)?\.src:connection/)` so the same expectation matches both modes. | | `redis` | Straight migration (docker). | | `redis-cache` | 2 scenarios (`ioredis`, `redis-4`). Each scenario has its own `instrument-*.mjs` because they pass different `cachePrefixes` to `redisIntegration`. | | `redis-dc` | Uses `await import('redis-5')` inside `run()` to preserve the comment-documented ordering where the DC subscriber must be registered (via `Promise.resolve().then`) before node-redis eagerly creates its native TracingChannels on require/import. | | `tedious` | Migrated but kept `describe.skip` (the test was previously skipped as flaky in #15798). Manually verified both modes pass when un-skipped. | | `apollo-graphql` | 3 scenarios (query, mutation, error). Shared `apollo-server.mjs` helper copied into the tmp dir via `copyPaths: ['apollo-server.mjs']` and loaded with `await import('./apollo-server.mjs')` from each scenario. | | `apollo-graphql/useOperationNameForRootSpan` | 6 scenarios. Reuses the parent `apollo-server.mjs` via `../../apollo-server.mjs` (the scenarios run from `useOperationNameForRootSpan/tmp_xxx/`, so this resolves back to the apollo-graphql folder). Scenarios use `Sentry.getClient().tracer` instead of capturing the client returned by `Sentry.init` (since init now lives in `instrument.mjs`). | ## Out of scope The following folders under `tracing/` are still using `createRunner` directly — they are not integrations for a specific package but general tracing-behavior tests, so they're intentionally left alone: - `meta-tags*`, `meta-tags-twp*` - `sample-rate-propagation/*`, `sample-rand-propagation` - `tracePropagationTargets/*` - `traceid-recycling*` - `dsc-txn-name-update` - `http-client-spans/*` - `maxSpans`, `envelope-header/*`, `linking` 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This test now failed twice in a row for me and I've seen it fail in countless PRs. Skipping it for now but we should prioritize #15579 to eventually unflake and unskip them again.