I believe this is our fifth proposal for Hive-Engine development funding, and it carries a straightforward message: we have delivered, and we are willing to keep going.
Our previous proposals all closed "within budget", with transparent reporting and more work completed than promised. This cycle follows the same open-ended mandate: fix what needs fixing as it comes up, keep building what the ecosystem requires, and document everything publicly along the way.
Achievements This Cycle
Hive-Engine Core: The Failover Fix
We diagnosed and resolved a long-standing reliability issue in hivesmartcontracts where nodes failed to fail over cleanly during primary Hive RPC degradation. PR #134 is now merged into main, delivering:
Request-level failover across the complete streamNodes list.
Scheduler-level node demotion to temporarily pause traffic to failing endpoints.
We conducted an audit on the legacy ssc_tokens_history repository, resolving 11 vulnerabilities (including a high-severity CSRF issue in the axios chain). PR #33 updated sscjs, pg, axios, and eslint.
We also published a scoped modernization roadmap for the next cycle:
Flexible single-sided timestamp filtering.
Cursor-based pagination to replace deep skip/limit queries.
Index strategy revisions based on actual account-history query patterns.
NFT history lookup path improvements.
Pre-intervention query telemetry.
hive-nectar Library Upgrades
v0.2.14 and v1.0.0 shipped with the following improvements:
Dual beacon support (Peakd + v4v), with thanks to Hive account@brianoflondon for the secondary beacon infrastructure.
5-minute on-disk node cache to reduce beacon load and accelerate application startup.
Docker/SQLite path fixes for containerized deployments.
Documentation cleanup for both hive-nectar and nectarengine, ensuring clean API references on ReadTheDocs.
Infrastructure & Operator Documentation
We published two critical diagnostic guides to support the Hive-Engine node community:
MongoDB WiredTiger Throttle Fix: Documented the "Triple Pincer" method for parallel archive restoration on high-core-count hardware, including the track_restore.sh progress script.
MongoDB Segfault Isolation: Traced a post-power-outage crash to a systemd environment variable (GLIBC_TUNABLES), documenting the exact diagnostic path and resolution.
HiveTools and Market-Viewr
Rebranded neo-beeswap to Mithril Diesel Pool with a responsive, theme-aware UI.
Implemented dynamic OpenGraph preview images for pool links using a Pillow backend.
Added searchable, scrollable token dropdowns to handle long lists efficiently.
Resolved dark mode rendering bugs across all pages.
Wired a dynamic sitemap to the Flask url_map.
Fixed the timeframe filtering bug on the Market-Viewr history chart backend.
Legendary 1 line PR
We can't forget about our favorite submission, a one-line quality of life fix for node operators:
Upcoming Objectives
Execute ssc_tokens_history Modernization: Implement the published roadmap, prioritizing cursor pagination and index enhancements.
hivesmartcontracts Maintenance: Continue to evaluate a dedicated RPC manager abstraction (so-called Option B) with the community.
Emergency Response Infrastructure: Maintain ongoing, un-scoped technical support for critical failures and debugging.
Tooling Updates: Provide continuous maintenance for hive-nectar, nectarengine, HiveTools, and Market-Viewr.
Funding Request
We expect to receive approximately 317 BEE per day in practice at present. We are requesting the maximum cap of 547 BEE per day, functioning as a "sweeper" proposal to utilize remaining daily budget funds after higher-priority proposals are serviced.
for the secondary beacon infrastructure.
