Bug Hunt
How a Timezone Bug Silently Corrupted Three Months of Analytics
The dashboard looked right. The numbers were wrong by a few percent every day, and the reason was one implicit conversion in a reporting query.
Query performance, deadlocks, N+1 patterns and data integrity problems across relational and non relational stores.
Bug Hunt
The dashboard looked right. The numbers were wrong by a few percent every day, and the reason was one implicit conversion in a reporting query.
Networking
At-least-once delivery means duplicates are normal operation, not an edge case. Here is how to build a handler that survives them.
Databases
Reading from cache is easy. Knowing when to remove something is where the bugs live, and most of them are races.
PostgreSQL
Adding a column should be instant. Sometimes it is not, and the difference is worth knowing before you run it in production.
Databases
The code reads beautifully and issues four thousand queries. Here is how to detect them automatically instead of discovering them in production.
PostgreSQL
Connection pooling explained through the failure it prevents, and why raising max_connections is usually the wrong fix.