Monero develops pruning and pruned-sync fixes. P2Pool documents an Arti proxy option, while a restricted-RPC block check remains proposed.
1. Monero pruning and pruned-sync fixes enter development
September 5–9, 2026 · Watch release and backport status
Affects Pruned Monero databases and synchronization.
A development commit saves pruning progress so interrupted work can resume. PR #11223 addresses a separate problem during pruned sync. A pruned node does not keep all transaction data, but its database still needs the correct hash for the omitted portion. The patch stores the supplied hash instead of trying to reconstruct it from incomplete data. Both changes still needed to reach a stable package.
- Keep a recoverable database copy before testing pruning changes. An interrupted older pruning run may require explicit follow-up rather than automatically benefiting from new checkpoints.
Sources: Pruning progress commit · Prunable hash fix
2. Restricted RPC block submission check proposed
September 4, 2026 · Watch only
Affects Restricted submit_block requests.
PR #11239 proposes rejecting blocks submitted through restricted RPC when they use an old RandomX seed epoch. The change would let the node reject those blocks earlier. It was still open on September 11.
- Wait for upstream review and release inclusion. Keep unrestricted RPC access private.
Sources: Proposed seed-epoch check
3. P2Pool documents Arti proxy setup
September 8, 2026 · Optional configuration
Affects P2Pool connections through Tor.
P2Pool now documents how to send outbound connections through an Arti SOCKS proxy. The instructions also explain the separate setup needed to accept incoming onion connections. Operators whose existing Tor setup works do not need to migrate.
- Try an alternative proxy separately and confirm peer connectivity before changing an established setup.
Sources: P2Pool Arti documentation change
Monero Research Lab — September 9, 2026
Rucknium presented tests of Dandelion++, the transaction-relay method that helps hide a sender’s network location. In the simulated network, the chance of identifying the true origin tracked the share of spy nodes, matching the paper’s prediction. Participants noted limits: the test omitted unreachable nodes and did not model every hostile behavior. It supported the implementation under those conditions, not a universal privacy guarantee.
Phase 2 FCMP++ integration proposals were ready for review. Audit quotes were being evaluated, and hot/cold wallet support was close to completion. P2Pool v5 testing was planned for the stressnet.
MAGIC Grants also reported a review of concerns about the Bulletproofs+ security proofs. Further analysis had not established an inflation flaw in Monero’s use. A full public report was still forthcoming.