Skip to content

fix: stabilize Firo node config to prevent RPC deadlocks#3282

Merged
bernd2022 merged 1 commit intodevelopfrom
fix/firo-node-stability
Feb 27, 2026
Merged

fix: stabilize Firo node config to prevent RPC deadlocks#3282
bernd2022 merged 1 commit intodevelopfrom
fix/firo-node-stability

Conversation

@bernd2022
Copy link
Collaborator

Summary

  • Restrict Firo node to IPv4-only (onlynet=ipv4) to eliminate hanging connections to unreachable IPv6 peers
  • Reduce max peer connections from 125 to 40 to lower P2P load and lock contention
  • Increase RPC work queue from 16 to 32 to prevent "work queue depth exceeded" rejections
  • Lower dbcache from 1024 to 512 for more frequent chainstate flushes (less block rewind on restart)

Context

The Firo node was repeatedly freezing due to a cascade: IPv6 connection attempts → socket timeouts → mutex contention → RPC deadlock. See FIRO-NODE-ANALYSIS.md for full root cause analysis.

Test plan

  • Deploy to dev and verify node syncs without socket sending timeout or Network is unreachable errors
  • Confirm RPC responds reliably under normal API polling load
  • Monitor for 24-48h to verify no freezes occur

Restrict to IPv4-only to eliminate hanging IPv6 connections,
reduce max peer connections (125→40), increase RPC work queue
(16→32), and lower dbcache (1024→512) for more frequent flushes.
@bernd2022 bernd2022 merged commit 9a43aea into develop Feb 27, 2026
8 checks passed
@bernd2022 bernd2022 deleted the fix/firo-node-stability branch February 27, 2026 08:23
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants