Jagalite/superseedr v1.0.7

on GitHub

Standard Builds (Recommended)

• macOS Universal: superseedr-v1.0.7-universal-macos.pkg
• Linux (Debian): superseedr_v1.0.7_amd64.deb
• Windows (MSI): superseedr_v1.0.7_x64_en-US.msi

Private Builds (Advanced)

These builds do not contain PEX or DHT in the final binary. Not recommended for normal users unless you have privacy requirements.

• macOS Universal: superseedr-v1.0.7-private-universal-macos.pkg
• Linux (Debian): superseedr-private_v1.0.7_amd64.deb
• Windows (MSI): superseedr-private_v1.0.7_x64_en-US.msi

Release v1.0.7

New Features

• Synthetic Benchmark Harness: Added feature-gated benchmark tooling for local download, upload, and swarm capacity testing, with adaptive step sizing, JSON summaries, ETA reporting, and generated run artifacts; see docs/synthetic-benchmark.md.
• Adaptive Disk Backpressure: Added live download throttling that reacts to disk write pressure so high-throughput sessions can back off when storage latency becomes the bottleneck.
• Additional Watch Paths: Restored support for additional watch folders so operators can ingest torrents from multiple configured locations.

Improvements

• TUI Swarm Visibility: Refined the peer table, files panel, and swarm heatmap with inactive-peer summaries, saturated-peer file strips, more compact peer-action labels, and better behavior in short terminal layouts.
• Performance Diagnostics: Added measured footer FPS, disk-queue write latency timing, clearer benchmark capacity reporting, and better synthetic peer ramping for load investigations.
• Docker and Client Interop: Made the Docker Compose setup wait for Gluetun health, grouped interop lanes by client, and improved qBittorrent login and add-response compatibility.
• Dependency and Runtime Maintenance: Removed config/logging helper dependencies, hardened the replacement config, logging, and atomic-write paths, refreshed dependency pins, and kept CI on the released Rust 1.95.0 toolchain.

Bug Fixes

• Piece Request Correctness: Fixed duplicate requests during piece writes, adjusted endgame request scheduling, and removed defensive pending-block scheduling that could interfere with active work.
• Telemetry and Logging Accuracy: Fixed logging flush behavior and telemetry percentile regressions so runtime diagnostics stay reliable under load.
• Layout Edge Cases: Fixed files-panel and heatmap sizing edge cases in constrained terminal space while preserving useful peer and availability context.
• Synthetic Load Stability: Fixed synthetic-load permit accounting and Windows clippy issues in the benchmark-only build path.