Comparison with the original project¶
fritzFluxDB is a fork of bb-Ricardo/fritzinfluxdb by Ricardo Bartels, which laid the entire foundation for collecting FRITZ!Box metrics into InfluxDB. This fork modernises the codebase around operational reliability, a smaller container-first footprint, and QuestDB as a third storage backend.
fritzFluxDB (this fork) | fritzinfluxdb (original) | |
|---|---|---|
| Python | 3.13 | 3.7+ |
| Database backends | InfluxDB v1, InfluxDB v2 and QuestDB | InfluxDB v1 and v2 |
| Database client | httpx — single lightweight HTTP dependency | influxdb + influxdb_client libraries |
| Outage logging | One error on outage, silent retries, one recovery message | Repeated errors per retry |
| Graceful shutdown | Buffered measurements are flushed before exit | Buffer discarded on shutdown |
| HTTP backoff | Exponential backoff on 429/5xx, honours Retry-After, auto-shrinks batch on 413 | Fixed retry interval |
| Parser robustness | Hardened against malformed JSON/XML/CSV with descriptive errors | Basic parsing |
| Measurement identity | Named after the FRITZ!Box serial — swapping hardware keeps history cleanly separated | Single static measurement |
| Secret handling | Credentials masked in logs; refuses to send credentials over plain HTTP to remote hosts | — |
| Docker image | Multi-arch (amd64/arm64), non-root, Tini as PID 1 | Single-arch, runs as root |
| Timezone correctness | Log timestamps are timezone-aware | — |
The original still ships some features this fork intentionally dropped, for example automatic InfluxDB retention-policy creation. If you rely on those, the upstream project may suit you better.