InfluxDB v3.10.0 released on 17-06-2026
InfluxDB v3.10.0 is out now. Operators and developers get a compact on-disk catalog v3 migration and an offline inspection tool, plus processing-engine controls, readiness checks and additional Enterprise preview features.
See the full release notes on InfluxData’s documentation and the changelog on GitHub for complete details and upgrade steps.
What’s in this release
- Catalog v3: the on-disk catalog migrates automatically to a compact binary v3 format (~5–6× smaller). Migration is automatic, idempotent and crash-safe; use the new influxdb3 debug catalog tool to inspect catalog state from object storage without a running server.
- Processing engine: plugins can read from any database using database= on influxdb3_local.query(); new serve flags –restrict-plugin-triggers-to and –plugin-dir-only restrict plugin-trigger behaviour for finer security control.
- Readiness and observability: new GET /ready returns 200 when object storage is reachable and 503 when not; always-on heap profiling is enabled at startup and exposed via pprof (disable with MALLOC_CONF=prof:false); Enterprise adds 36 influxdb3_compactor_* Prometheus metrics including influxdb3_compactor_snapshot_lag_seconds.
Upgrade notes
- Catalog v3 migration is one-way — back up {prefix}/catalogs/ and {prefix}/_catalog_checkpoint before upgrading because v3.9.x binaries cannot read a v3 catalog.
- CLI flag rename: –pt-partition-count was renamed to –pt-shard-count with no alias; update startup scripts. Also note API/permission changes: /api/v2/write now returns 403 for valid tokens without write permission and line-protocol parse errors return 400 (was 500).
Share any issues or observations from your upgrade and how the new catalog format behaves in your environment — the community will appreciate real-world feedback.
