Product Decisions This Supports
- Diagnosing the "slow page" class of bugs with evidence, not guesses. mysqlnd exposes per-query instrumentation that Doctrine's own query log does not: bytes sent/received, rows examined vs. rows returned, connection setup time, and prepared-statement reuse. This separates slow SQL from slow data transfer from slow connection churn — three problems with three completely different fixes (indexing,
SELECT narrowing / pagination, versus persistent connections or a pooler).
- Build-vs-buy on observability. Its existence is a useful data point that low-level DB driver telemetry is worth having in dev, while its 2013 dormancy is a stronger data point that you shouldn't build or maintain this yourself. It argues for buying APM (Blackfire, Tideways, New Relic, Sentry) rather than owning driver-level instrumentation.
- Roadmap prioritization for performance work. If profiling shows connection setup dominating request time, that's an infrastructure roadmap item (persistent connections, ProxySQL/PgBouncer equivalent), not an application-code sprint item. Conversely, high rows-examined-to-rows-returned ratios point at a schema/indexing backlog.
- N+1 and query-count budgets. Feeding query counts and per-query cost into the profiler toolbar makes it cheap to justify enforcing query-count assertions in CI or a per-request budget in code review standards.
- Framework-mismatch as a scoping signal. This is a Symfony2 bundle, not a Laravel package. For a Laravel codebase the equivalent role is filled by Laravel Debugbar and Telescope; the decision it supports there is "Telescope's query panel plus a real APM," not "adopt this."
When to Consider This Package
Adoption criteria — all must be true:
- You are on a legacy Symfony 2.x application still in maintenance, with
mysqlnd as the active MySQL driver.
- You have no APM budget or access and dev-environment query visibility is genuinely absent.
- You accept it as read-only dev tooling, pinned and vendored, with the expectation of zero upstream support.
Look elsewhere — which is nearly always:
- Any Laravel project. Wrong framework entirely; use Debugbar/Telescope plus an APM.
- Any modern Symfony (4/5/6/7). The toolbar extension API it targets no longer exists. Doctrine DBAL's own logging plus Symfony's profiler covers most of it.
- Any PHP 7+/8 runtime. Last release 2013-10-15 means PHP 5.x-era code; expect fatal errors, not just deprecation notices.
- Postgres, SQLite, or non-mysqlnd MySQL drivers. The data source literally doesn't exist.
- Production use of any kind. Debug toolbars leak schema, query text, and credentials; and this one is unmaintained with zero dependents — no community to catch a leak.
mysqli/PDO without mysqlnd. Silent no-op at best.
The opportunity score (~0.76) reads as "a real gap in the ecosystem" rather than "a package worth adopting." Treat it as a spec for what your APM should show you, not as a dependency.
How to Pitch It (Stakeholders)
One line: A small Symfony2 add-on that surfaces MySQL driver–level performance data inside the dev profiler — useful as a checklist of what good DB observability looks like, but 12 years unmaintained and unusable on any current stack.
For executives: The underlying idea — measure the database driver, not just the ORM — is sound and directly shortens performance investigations. The package itself is not a candidate: last release 2013, no dependents, Symfony2-only. Recommendation is to buy the capability (Blackfire/Tideways/New Relic class tooling) rather than adopt this. Cost is a subscription; the alternative is unpaid maintenance of dead code.
For engineering: Read it for the metric list — bytes transferred, rows examined, connection time, prepared-statement reuse — and verify your chosen APM actually reports those. Do not add it as a dependency on any PHP 7+ or Laravel/Symfony-modern codebase; the toolbar-extension API it hooks is gone and the code predates the PHP 7 error model. If you're on a legacy Symfony 2 maintenance branch with no APM, vendor it as dev-only, never enable it in a staging or production environment, and timebox any debugging investment in it.