The original title is about a Modbus proxy with specific features. Let me rewrite it for a mobile feed.

Short summary

A detailed engineering post on building a robust Modbus cache proxy that handles nighttime inverter shutdowns, firmware reboots, and stale data gracefully. The author implements hard timeouts per register batch, asymmetric retry logic with backoff, and explicit stale-cache logging so Home Assistant can alert on outdated values. The key insight is distinguishing survivable TimeoutErrors from dead-connection exceptions to avoid serving silently stale data.

• •Every Modbus read gets a hard timeout with 50ms spacing between batch reads
• •Asymmetric retry: fast retry on failure, normal interval on success, stale warning after 120s
• •Distinguishing TimeoutError (survivable) from other exceptions (connection dead) is critical for reliability

Generated with AI, which can make mistakes.