Loki
An open-source log aggregation system designed around Prometheus's label model. It indexes only labels rather than log content, uses the LogQL query language, and is characterized by low storage overhead and straightforward operation.
Timeline (last 90 days)
The Loki tool group of Grafana MCP allows executing LogQL log and metric queries, listing labels, and extracting detected log patterns.
PromQL and LogQL are the query languages of Prometheus and Loki respectively; PromQL is oriented toward time series and supports functions such as rate() and histogram_quantile(), used to compute metrics such as error rates and quantiles.
The framework uses characters from Norse mythology as names, including agents such as Thor, Loki, Freya, Heimdal, and Ratatoskr.
Loki queries use LogQL, which requires filtering by labels to narrow down log streams first, then performing regex or keyword matching on the stream content.
Loki indexes labels rather than log bodies, so performance degrades noticeably when searching for content beyond labels.
Grafana's LGTM Stack consists of Loki (logs), Grafana (visualization), Tempo (tracing), and Mimir (metrics).
Loki indexes only the metadata of logs rather than the full text, in order to reduce storage costs.
All Facts (7)
The Loki tool group of Grafana MCP allows executing LogQL log and metric queries, listing labels, and extracting detected log patterns.
50%UnverifiedPromQL and LogQL are the query languages of Prometheus and Loki respectively; PromQL is oriented toward time series and supports functions such as rate() and histogram_quantile(), used to compute metrics such as error rates and quantiles.
50%UnverifiedThe framework uses characters from Norse mythology as names, including agents such as Thor, Loki, Freya, Heimdal, and Ratatoskr.
50%UnverifiedLoki queries use LogQL, which requires filtering by labels to narrow down log streams first, then performing regex or keyword matching on the stream content.
50%UnverifiedLoki indexes labels rather than log bodies, so performance degrades noticeably when searching for content beyond labels.
50%UnverifiedGrafana's LGTM Stack consists of Loki (logs), Grafana (visualization), Tempo (tracing), and Mimir (metrics).
50%UnverifiedLoki indexes only the metadata of logs rather than the full text, in order to reduce storage costs.
50%