Flamingo Raises $4.5M to Get MSPs Out of the Ticket Queue

Skip to content
Grafana Loki logo

Grafana Loki

Security Information and Event Mgmt. (SIEM)

Open Source
Selected
Gen1
E
Enterprise
Free Tier
Paid Plans
OpenMSP Score
52
40
Reddit Impact Score
Github Score
833M
29KStars
4KForks
14KCommits
GNU Affero General Public License v3.0License
Nov 20, 2026Last commit
Grafana Loki is an open-source log aggregation system from Grafana Labs, licensed under AGPLv3. Instead of indexing the full text of every log line, it indexes only a small set of labels (as Prometheus does for metrics) and stores compressed log chunks in object storage such as S3, GCS or Azure Blob. That keeps storage and operating costs low, at the price of slower searches across very large unfiltered datasets. Logs are shipped with Grafana Alloy (Promtail has been deprecated in favour of Alloy), the OpenTelemetry Collector, Fluent Bit, Fluentd or Vector, and queried with LogQL from Grafana, where they sit next to metrics and traces. Loki can run as a single binary or as scalable microservices on Kubernetes, and supports alerting rules, retention policies and native multi-tenancy keyed by a tenant ID, which suits MSPs keeping each client's logs apart. Grafana Labs also offers Loki as a managed service in Grafana Cloud (free for 50 GB of logs a month with 14-day retention; Pro from a $19 monthly platform fee plus usage) and a self-managed commercial edition, Grafana Enterprise Logs.
image media

Key Features

Label-Based Indexing

Indexes only labels, not log content, which keeps the index small and storage cheap

Object Storage Backend

Stores compressed chunks in S3, GCS, Azure Blob or compatible storage for long retention

LogQL

Query language for filtering logs and turning them into metrics, inspired by PromQL

Native Multi-Tenancy

Separates data per tenant ID, with per-tenant limits and retention

Grafana Integration

Explore logs next to metrics and traces in Grafana dashboards

Flexible Collection

Ingest with Grafana Alloy, OpenTelemetry, Fluent Bit, Fluentd, Vector and more

Pros and Cons

Pros

Low storage cost

Label-only indexing and object storage keep costs well below full-text log platforms

Built-in multi-tenancy

Tenant IDs separate client data in one cluster, useful for MSPs

Open source (AGPLv3)

Free to self-host with all core features

Fits the Grafana stack

Works naturally with Prometheus, Mimir, Tempo and Grafana dashboards

Managed option

Grafana Cloud offers hosted Loki with a free tier

Cons

Label design matters

Too many or high-cardinality labels hurt performance and cost; good labels need planning

Weak full-text search

Searching content across large time ranges without label filters is slow

Not a SIEM

No built-in security detections, correlation or case management

Operational complexity at scale

Distributed mode has many components to tune and run

Feature Comparison

Features
Cost and Licensing
Loading the comparison

Comments

No reviews yet. Be the first to share your experience with Grafana Loki.

Frequently Asked Questions

/vendor/grafana-loki