delvingbitcoin

Casual research on running `mempoolfullrbf`

Casual research on running `mempoolfullrbf`

Original Postby valuedmammal

Posted on: August 27, 2024 01:31 UTC

The email discusses a test that showcased the full-rbf node (B) experiencing significantly less variance in score compared to other nodes, pointing towards an interesting observation regarding node behavior and network health.

The sender mentioned that node uptime could be a confounding variable affecting network visibility, with the fullrbf node being an always-on server in contrast to the bip125 node, which is a pruned node operating on an old laptop and frequently turned off. Efforts were made to minimize this variation by ensuring these nodes were peered, allowing for the sharing of mempool contents. This approach underlines the significance of node operators monitoring statistics to observe changes in network usage, aiming to ensure the robustness of the p2p network.

The purpose of this exploration was to initiate a broader discussion about the p2p network's health, emphasizing the importance of tracking node statistics to understand usage trends. The score metric, despite being simple, is not expected to always be perfect due to the distributed nature of the network where different nodes see different transactions. A deviation from the ideal score is not necessarily alarming; however, large or prolonged deviations might indicate a discrepancy between local policy and the wider network or that transactions are confirming without ever entering the mempool. This observation could signal issues that need addressing to maintain network health.

The sender abstains from taking a specific stance on whether nodes should align with miner practices or the reverse. However, there is an advocacy for finding a balance between maintaining reliable defaults and adapting policies to evolve with network demands. The ultimate goal highlighted is to optimize the mempool as an efficient platform conducive for user transactions, reflecting a need for continuous adaptation and policy evolution to support the network's functionality and reliability.