test(node-integration): Harden knex mysql2 healthcheck to fix flaky test by github-actions[bot] · Pull Request #21868 · getsentry/sentry-javascript

github-actions · 2026-07-01T06:45:16Z

The knex + mysql2 node integration test (suites/tracing/knex/mysql2/test.ts) flakes intermittently on CI.

Root cause

The suite's docker-compose.yml gated readiness on:

test: ['CMD-SHELL', 'mysqladmin ping -h 127.0.0.1 -uroot -pdocker']

mysqladmin ping reports the server as alive even when it receives an access-denied error, and MySQL 8 spins up a temporary bootstrap server during first-time initialization (before the root password is applied and the tests database is created). As a result, docker compose up -d --wait — which the runner uses to block until healthy — can return before the real server is ready. The scenario then races the DB init and the query spans never arrive, so the transaction assertion times out.

The pg sibling suite already avoids this class of bug by checking the actual database with pg_isready -U test -d tests.

Fix

Make the mysql2 healthcheck run an authenticated query against the tests database, mirroring the pg pattern:

test: ['CMD-SHELL', 'mysql -h 127.0.0.1 -uroot -pdocker -e "SELECT 1" tests']

This only passes once the real server is up, the root password works, and tests exists — so --wait no longer returns prematurely.

Fixes#21861

The suite gated readiness on `mysqladmin ping`, which reports "alive" even during MySQL 8's init bootstrap (it treats access-denied as alive). So `docker compose up --wait` could return before the real server and root password were ready, and the scenario's first connection was dropped ("server closed the connection") — crashing the run and timing out the test. Gate on an authenticated `SELECT 1` instead, mirroring the knex/mysql2 sibling suite (#21868).

github-actions · 2026-07-01T06:45:16Z

The knex + mysql2 node integration test (suites/tracing/knex/mysql2/test.ts) flakes intermittently on CI.

Root cause

The suite's docker-compose.yml gated readiness on:

test: ['CMD-SHELL', 'mysqladmin ping -h 127.0.0.1 -uroot -pdocker']

mysqladmin ping reports the server as alive even when it receives an access-denied error, and MySQL 8 spins up a temporary bootstrap server during first-time initialization (before the root password is applied and the tests database is created). As a result, docker compose up -d --wait — which the runner uses to block until healthy — can return before the real server is ready. The scenario then races the DB init and the query spans never arrive, so the transaction assertion times out.

The pg sibling suite already avoids this class of bug by checking the actual database with pg_isready -U test -d tests.

Fix

Make the mysql2 healthcheck run an authenticated query against the tests database, mirroring the pg pattern:

test: ['CMD-SHELL', 'mysql -h 127.0.0.1 -uroot -pdocker -e "SELECT 1" tests']

This only passes once the real server is up, the root password works, and tests exists — so --wait no longer returns prematurely.

Fixes#21861

The suite gated readiness on `mysqladmin ping`, which reports "alive" even during MySQL 8's init bootstrap (it treats access-denied as alive). So `docker compose up --wait` could return before the real server and root password were ready, and the scenario's first connection was dropped ("server closed the connection") — crashing the run and timing out the test. Gate on an authenticated `SELECT 1` instead, mirroring the knex/mysql2 sibling suite (#21868).

github-actions · 2026-07-01T06:45:16Z

The knex + mysql2 node integration test (suites/tracing/knex/mysql2/test.ts) flakes intermittently on CI.

Root cause

The suite's docker-compose.yml gated readiness on:

test: ['CMD-SHELL', 'mysqladmin ping -h 127.0.0.1 -uroot -pdocker']

mysqladmin ping reports the server as alive even when it receives an access-denied error, and MySQL 8 spins up a temporary bootstrap server during first-time initialization (before the root password is applied and the tests database is created). As a result, docker compose up -d --wait — which the runner uses to block until healthy — can return before the real server is ready. The scenario then races the DB init and the query spans never arrive, so the transaction assertion times out.

The pg sibling suite already avoids this class of bug by checking the actual database with pg_isready -U test -d tests.

Fix

Make the mysql2 healthcheck run an authenticated query against the tests database, mirroring the pg pattern:

test: ['CMD-SHELL', 'mysql -h 127.0.0.1 -uroot -pdocker -e "SELECT 1" tests']

This only passes once the real server is up, the root password works, and tests exists — so --wait no longer returns prematurely.

Fixes#21861

The suite gated readiness on `mysqladmin ping`, which reports "alive" even during MySQL 8's init bootstrap (it treats access-denied as alive). So `docker compose up --wait` could return before the real server and root password were ready, and the scenario's first connection was dropped ("server closed the connection") — crashing the run and timing out the test. Gate on an authenticated `SELECT 1` instead, mirroring the knex/mysql2 sibling suite (#21868).

github-actions · 2026-07-01T06:45:16Z

The knex + mysql2 node integration test (suites/tracing/knex/mysql2/test.ts) flakes intermittently on CI.

Root cause

The suite's docker-compose.yml gated readiness on:

test: ['CMD-SHELL', 'mysqladmin ping -h 127.0.0.1 -uroot -pdocker']

mysqladmin ping reports the server as alive even when it receives an access-denied error, and MySQL 8 spins up a temporary bootstrap server during first-time initialization (before the root password is applied and the tests database is created). As a result, docker compose up -d --wait — which the runner uses to block until healthy — can return before the real server is ready. The scenario then races the DB init and the query spans never arrive, so the transaction assertion times out.

The pg sibling suite already avoids this class of bug by checking the actual database with pg_isready -U test -d tests.

Fix

Make the mysql2 healthcheck run an authenticated query against the tests database, mirroring the pg pattern:

test: ['CMD-SHELL', 'mysql -h 127.0.0.1 -uroot -pdocker -e "SELECT 1" tests']

This only passes once the real server is up, the root password works, and tests exists — so --wait no longer returns prematurely.

Fixes#21861

The suite gated readiness on `mysqladmin ping`, which reports "alive" even during MySQL 8's init bootstrap (it treats access-denied as alive). So `docker compose up --wait` could return before the real server and root password were ready, and the scenario's first connection was dropped ("server closed the connection") — crashing the run and timing out the test. Gate on an authenticated `SELECT 1` instead, mirroring the knex/mysql2 sibling suite (#21868).

github-actions · 2026-07-01T06:45:16Z

The knex + mysql2 node integration test (suites/tracing/knex/mysql2/test.ts) flakes intermittently on CI.

Root cause

The suite's docker-compose.yml gated readiness on:

test: ['CMD-SHELL', 'mysqladmin ping -h 127.0.0.1 -uroot -pdocker']

mysqladmin ping reports the server as alive even when it receives an access-denied error, and MySQL 8 spins up a temporary bootstrap server during first-time initialization (before the root password is applied and the tests database is created). As a result, docker compose up -d --wait — which the runner uses to block until healthy — can return before the real server is ready. The scenario then races the DB init and the query spans never arrive, so the transaction assertion times out.

The pg sibling suite already avoids this class of bug by checking the actual database with pg_isready -U test -d tests.

Fix

Make the mysql2 healthcheck run an authenticated query against the tests database, mirroring the pg pattern:

test: ['CMD-SHELL', 'mysql -h 127.0.0.1 -uroot -pdocker -e "SELECT 1" tests']

This only passes once the real server is up, the root password works, and tests exists — so --wait no longer returns prematurely.

Fixes#21861

The suite gated readiness on `mysqladmin ping`, which reports "alive" even during MySQL 8's init bootstrap (it treats access-denied as alive). So `docker compose up --wait` could return before the real server and root password were ready, and the scenario's first connection was dropped ("server closed the connection") — crashing the run and timing out the test. Gate on an authenticated `SELECT 1` instead, mirroring the knex/mysql2 sibling suite (#21868).

github-actions · 2026-07-01T06:45:16Z

The knex + mysql2 node integration test (suites/tracing/knex/mysql2/test.ts) flakes intermittently on CI.

Root cause

The suite's docker-compose.yml gated readiness on:

test: ['CMD-SHELL', 'mysqladmin ping -h 127.0.0.1 -uroot -pdocker']

mysqladmin ping reports the server as alive even when it receives an access-denied error, and MySQL 8 spins up a temporary bootstrap server during first-time initialization (before the root password is applied and the tests database is created). As a result, docker compose up -d --wait — which the runner uses to block until healthy — can return before the real server is ready. The scenario then races the DB init and the query spans never arrive, so the transaction assertion times out.

The pg sibling suite already avoids this class of bug by checking the actual database with pg_isready -U test -d tests.

Fix

Make the mysql2 healthcheck run an authenticated query against the tests database, mirroring the pg pattern:

test: ['CMD-SHELL', 'mysql -h 127.0.0.1 -uroot -pdocker -e "SELECT 1" tests']

This only passes once the real server is up, the root password works, and tests exists — so --wait no longer returns prematurely.

Fixes#21861

The suite gated readiness on `mysqladmin ping`, which reports "alive" even during MySQL 8's init bootstrap (it treats access-denied as alive). So `docker compose up --wait` could return before the real server and root password were ready, and the scenario's first connection was dropped ("server closed the connection") — crashing the run and timing out the test. Gate on an authenticated `SELECT 1` instead, mirroring the knex/mysql2 sibling suite (#21868).

github-actions · 2026-07-01T06:45:16Z

The knex + mysql2 node integration test (suites/tracing/knex/mysql2/test.ts) flakes intermittently on CI.

Root cause

The suite's docker-compose.yml gated readiness on:

test: ['CMD-SHELL', 'mysqladmin ping -h 127.0.0.1 -uroot -pdocker']

mysqladmin ping reports the server as alive even when it receives an access-denied error, and MySQL 8 spins up a temporary bootstrap server during first-time initialization (before the root password is applied and the tests database is created). As a result, docker compose up -d --wait — which the runner uses to block until healthy — can return before the real server is ready. The scenario then races the DB init and the query spans never arrive, so the transaction assertion times out.

The pg sibling suite already avoids this class of bug by checking the actual database with pg_isready -U test -d tests.

Fix

Make the mysql2 healthcheck run an authenticated query against the tests database, mirroring the pg pattern:

test: ['CMD-SHELL', 'mysql -h 127.0.0.1 -uroot -pdocker -e "SELECT 1" tests']

This only passes once the real server is up, the root password works, and tests exists — so --wait no longer returns prematurely.

Fixes#21861

The suite gated readiness on `mysqladmin ping`, which reports "alive" even during MySQL 8's init bootstrap (it treats access-denied as alive). So `docker compose up --wait` could return before the real server and root password were ready, and the scenario's first connection was dropped ("server closed the connection") — crashing the run and timing out the test. Gate on an authenticated `SELECT 1` instead, mirroring the knex/mysql2 sibling suite (#21868).

github-actions · 2026-07-01T06:45:16Z

The knex + mysql2 node integration test (suites/tracing/knex/mysql2/test.ts) flakes intermittently on CI.

Root cause

The suite's docker-compose.yml gated readiness on:

test: ['CMD-SHELL', 'mysqladmin ping -h 127.0.0.1 -uroot -pdocker']

mysqladmin ping reports the server as alive even when it receives an access-denied error, and MySQL 8 spins up a temporary bootstrap server during first-time initialization (before the root password is applied and the tests database is created). As a result, docker compose up -d --wait — which the runner uses to block until healthy — can return before the real server is ready. The scenario then races the DB init and the query spans never arrive, so the transaction assertion times out.

The pg sibling suite already avoids this class of bug by checking the actual database with pg_isready -U test -d tests.

Fix

Make the mysql2 healthcheck run an authenticated query against the tests database, mirroring the pg pattern:

test: ['CMD-SHELL', 'mysql -h 127.0.0.1 -uroot -pdocker -e "SELECT 1" tests']

This only passes once the real server is up, the root password works, and tests exists — so --wait no longer returns prematurely.

Fixes#21861

The suite gated readiness on `mysqladmin ping`, which reports "alive" even during MySQL 8's init bootstrap (it treats access-denied as alive). So `docker compose up --wait` could return before the real server and root password were ready, and the scenario's first connection was dropped ("server closed the connection") — crashing the run and timing out the test. Gate on an authenticated `SELECT 1` instead, mirroring the knex/mysql2 sibling suite (#21868).

test(node-integration): Harden knex mysql2 healthcheck to fix flaky test
6961788
Fixes#21861

github-actionsBot requested a review from a team as a code owner July 1, 2026 06:45

github-actionsBot requested review from JPeer264, andreiborza and mydea and removed request for a teamJuly 1, 2026 06:45

mydea approved these changes Jul 1, 2026
View reviewed changes

mydea enabled auto-merge (squash) July 1, 2026 06:51

mydea merged commit 690f778 into developJul 1, 2026
87 of 89 checks passed

mydea deleted the fix/knex-mysql2-healthcheck-flaky branch July 1, 2026 07:30

sentry-release-botBot mentioned this pull request Jul 1, 2026
publish: getsentry/sentry-javascript@10.63.0 getsentry/publish#8758
Closed
49 tasks

github-actionsBot mentioned this pull request Jul 1, 2026
[Flaky CI]: Node Integration Tests - suites/tracing/knex/mysql2/test.ts > knex auto instrumentation > with mysql2 client > should auto-instrument knex package [cjs] #21861
Closed