QueryCanary Logo QueryCanary
Login Get Started

Welcome to QueryCanary 🐤

QueryCanary is a lightweight, SQL-powered monitoring tool that helps you catch silent data failures in production databases — before they impact customers or downstream systems.

It was built to proactively alert you to real problems, either from issues you may suspect exist or most importantly things that have bitten you before.

Additionally, it can be used to keep track of values over time and even provide simple dashboards for your data.

Why QueryCanary?

  • Your site has uptime monitoring.
  • Your servers have performance metrics.
  • Your companies software projects have unit tests.
  • But your data is silently breaking — and no one knows until it’s too late.

QueryCanary helps you recognize that by letting you:

  • Define SQL-based data integrity checks
  • Schedule them to run automatically
  • Get alerts when results drift from expectations

What It’s Great For

  • Detecting sudden drops in daily signups
  • Catching null values where they shouldn’t be
  • Monitoring counts, duplicates, or invalid joins
  • Tracking expected combinations (e.g., status + type)
  • Alerting on stale data that hasn’t updated recently

How It Works

1. Connect your database

We currently support Postgres / MySQL + SSH tunneling. Our Quickstart will walk you through setting up a read only user.

More database engines will be introduced as requested, if you need one, please email support@querycanary.com

2. Write a SQL check

We’ll connect to your database and inspect the schema to provide you a editor to build & test a SQL check. The query results can return all sorts of data and we’ll figure out good ways to analyze your results.

Currently, single value metrics will give you solid out of the box features like charting, anomaly detection, thresholds, and more. Multiple column, or row based results will fall back to simpler checks like structure change.

Example:

-- Count all users created yesterday

SELECT COUNT(*) 
FROM users 
WHERE created_at >= CURRENT_DATE - INTERVAL '1 day';

-- Tricks like this ensure you are looking at full data
-- pictures instead of partial days.

3. Set a schedule

QueryCanary supports a full crontab syntax, so you can customize check intervals to pretty much anything you can imagine.

4. Get alerted

When QueryCanary detects a change in your data, it’ll send you or your team an email alert.

Types of Checks

Currently QueryCanary smartly determines the right kind of checks applicable to the type, periodicity, and shape of your data.

Anomaly Detection

Detects anomalies in numerical sequences using statistical methods. Checks Performed:

  • Z-Score Analysis: Calculates the z-score of the latest value compared to historical values. Triggers an alert if the z-score exceeds a specified threshold.
  • Baseline Statistics: Uses the mean and standard deviation of historical values to establish a baseline for comparison.
  • Insufficient Data: Skips anomaly detection if there are not enough historical values to calculate meaningful statistics.

Percent Change

Calculates the percentage change between numeric values. Triggers an alert if the percentage change exceeds a specified threshold.

Value change

Triggered by significant differences in values, types, or structures

  • List Comparison: Compares the length of lists and triggers an alert if the length changes significantly. Detects changes from an empty list to a populated list or vice versa.
  • String Similarity: Uses Levenshtein distance to calculate the similarity between strings. Triggers an alert if the similarity drops below a specified threshold.
  • Structural Change: Compares the number of rows and column names in the result. Detects if the structure of the data (e.g., column names or row count) has changed between runs.

Built for Safety

  • Read-only access recommended
  • SSL by default
  • SSH Tunnel available
  • Credentials are encrypted at rest
  • No raw data is stored — only your SQL and summarized results

Example Use Cases

These SQL checks can be used to monitor common business-critical data integrity conditions. Each includes:

  • Business purpose
  • What it catches
  • SQL example

1. Daily user signups dropped unexpectedly

Purpose: Alert if new users fell below expected volume
Catches: Failed signup flow, low traffic, frontend bugs

SELECT COUNT(*) FROM users
WHERE created_at >= CURRENT_DATE - INTERVAL '1 day';

2. Listings with no price

Purpose: Detect active listings without pricing
Catches: Incomplete inventory, pricing bugs

SELECT COUNT(*) FROM listings
WHERE price IS NULL AND status = 'active';

3. Orders created without associated users

Purpose: Catch data integrity issues or bad joins
Catches: Orphaned records from deleted users or import bugs

SELECT COUNT(*) FROM orders o
LEFT JOIN users u ON u.id = o.user_id
WHERE u.id IS NULL;

4. Too many password reset requests

Purpose: Monitor for spikes in account recovery traffic
Catches: Spam attacks or broken login flows

SELECT COUNT(*) FROM password_resets
WHERE requested_at >= NOW() - INTERVAL '1 hour';

5. Revenue total is lower than expected

Purpose: Alert if revenue is abnormally low
Catches: Payment failures or low sales volume

SELECT SUM(amount) FROM transactions
WHERE created_at >= CURRENT_DATE;

6. Too many failed login attempts

Purpose: Detect potential brute-force attempts
Catches: Security threats or broken auth logic

SELECT COUNT(*) FROM login_attempts
WHERE success = false AND attempted_at >= NOW() - INTERVAL '10 minutes';

7. Missing shipping addresses for shipped orders

Purpose: Catch orders missing fulfillment info
Catches: Bugs in order processing or database sync

SELECT COUNT(*) FROM orders
WHERE status = 'shipped' AND shipping_address_id IS NULL;

8. Duplicate user emails

Purpose: Ensure uniqueness of key identity field
Catches: App logic bugs, import issues

SELECT email, COUNT(*) FROM users
GROUP BY email
HAVING COUNT(*) > 1;

9. Inactive users with active subscriptions

Purpose: Find billing mismatches
Catches: Revenue leakage or status bugs

SELECT COUNT(*) FROM subscriptions s
JOIN users u ON u.id = s.user_id
WHERE u.status = 'inactive' AND s.status = 'active';

10. Stale product catalog

Purpose: Alert when products haven’t been updated
Catches: ETL issues, scraper failure, stale cache

SELECT MAX(updated_at) FROM products;

11. Customer records missing required fields

Purpose: Validate completeness of customer data
Catches: Form bugs, missing data from third-party sources

SELECT COUNT(*) FROM customers
WHERE email IS NULL OR country IS NULL;

12. Unusually low inventory

Purpose: Track low-stock products before they go out
Catches: Stocking issues, demand spikes

SELECT COUNT(*) FROM products
WHERE stock_quantity < 5;

Who It’s For

  • Devs who know SQL
  • Data engineers and analysts
  • SaaS teams tired of finding broken dashboards after it’s too late
  • CTOs with disparate teams constantly launching features

Get Started

Need help? support@querycanary.com or join our Discord