Repository navigation
Prepare Bunny Tracker for Raspberry Pi deployment - #1
Merged
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Raspberry Pi migration was blocked by unsupported Python setup instructions, incomplete model and annotation transfer instructions, and missing boot services. Automatic motion clips could also keep writing after clipping stopped for low disk space.
Reproduce the storage failure by setting a free-disk floor above available space, triggering a clipping upload, and allowing monitoring to return to normal motion mode. Previously those automatic motion uploads bypassed the disk floor.
Apply the disk floor to tagged motion uploads while preserving manual recordings and alerts. Serialize inference and configuration changes, join cancelled workers, and terminate and reap FFmpeg before replacement. Add shared transient paths, trusted-proxy handling, HTTP-aware CSP, a benchmark driven by promotion metadata, systemd services, model-bundle verification, and updated migration instructions.
Validation on the development laptop: all 21 hermetic check scripts passed across the initial run and targeted reruns after installing the already-declared missing matplotlib dependency. Final capture regressions, Ruff, the production frontend build, an isolated Uvicorn/FFmpeg synthetic-camera smoke test, and service-unit validation with local executable substitutions passed.
Pi hardware, reboot recovery, the production tunnel, and SMTP delivery remain unverified. Classification still requires a validated seven-class model.