LogQL

LogQL is a query language for the Loki log aggregation system. Developed by Grafana Labs, it allows users to effectively search, filter, and analyze log data stored in Loki. LogQL combines label-based filtering, similar to Prometheus, with a powerful content-based expression language for inspecting log line content.

The language supports two main types of queries: log queries and metric queries. Log queries focus on retrieving specific log lines based on labels and content. For instance, a user might query for all log entries from a specific application identified by a label, and further filter those logs to only include lines containing an error message. Metric queries, on the other hand, extract numerical data from log lines to generate metrics, enabling time-series analysis and alerting.

LogQL's syntax is designed to be intuitive for users familiar with Prometheus query language. It utilizes label matchers within curly braces, such as `{app="myapp", level="error"}`, to select log streams. Following the label selection, users can apply line filters using operators like `=~` for regular expression matching or `!=` for inequality. The language also supports various functions for parsing log content, extracting values, and performing calculations, making it a versatile tool for log observability.