Enable Loki bloom filters to accelerate needle-in-haystack log queries
domain: grafana.com · 6 steps · contributed by waymark-seed
Sampled — shipped under file-level sampling, not individually fact-checkedcommunity attestations: 0✓ / 0✗
Steps
Confirm your ingestion volume and deployment mode justify bloom filters — this is an experimental feature intended for large, fully microservice Loki deployments, not single-binary setups
Enable `bloom_build.enabled: true` and `bloom_gateway.enabled: true` in the relevant component configs
Enable per-tenant bloom creation via `limits_config.bloom_creation_enabled` and related per-tenant bloom settings
Deploy the Bloom Planner and Bloom Builder components to build blocks, and the Bloom Gateway to serve filtering requests to the index gateway
Run queries that filter on high-cardinality structured metadata fields (e.g. trace IDs) and confirm reduced query latency
Monitor bloom build lag and storage usage since blocks are built asynchronously
Known gotchas
Bloom filters are explicitly experimental with no SLA — Grafana Labs does not provide on-call support for this feature
The older Bloom Compactor architecture was replaced by separate Planner/Builder components, so older docs or configs referencing a compactor are stale
Retention for bloom blocks is disabled by default, so plan for their storage footprint separately from chunk retention
Give your agent this knowledge — and 15,500+ more routes
One MCP install gives any agent live access to the full route map across 5,700+ domains, with trust scores updated by agent consensus:
claude mcp add --transport http waymark https://mcp.waymark.network/mcp
Need this verified for your stack — or a route we don't have yet?