Duckdive

Search and visualize log files from S3, HTTPS URLs or local disk, in the browser. Nothing leaves your machine.

Duckdive searches and charts log files in the browser. It reads from Amazon S3 and S3-compatible storage, from HTTPS URLs, or from files on your computer. Discover shows matching records with a time histogram, a field sidebar and filters. Visualize builds area, line, bar, table and metric charts from the same query. The Data source page takes an S3 path pattern with date tokens, an HTTPS URL, or local files. Common AWS log formats are recognized automatically; other files are read as Parquet, CSV or JSON. Files are read straight from where they are stored and queried in the browser. Nothing is sent anywhere else, and data that has been read once is kept on your machine for later queries. Queries run on DuckDB, compiled to WebAssembly and bundled with the extension. The SQL generated for each query can be displayed. Credentials never leave your browser. Sign in at your own identity provider to get temporary AWS credentials, or enter access keys yourself.