General questions about channels [combined summary]



Individual post summaries: Click here to read the original discussion on the lightning-dev mailing list

Published on: 2018-01-03T03:45:50+00:00


Summary:

The conversation among Bitcoin developers Christian Decker and Andy Schroder revolves around the issue of large nodes and their impact on the Bitcoin network. Schroder suggests that avoiding connections to nodes with large channel capacities could promote a healthier network in the long term. However, Decker questions the effectiveness of this method due to the ease with which large nodes can masquerade as smaller ones, leading to UTXO fragmentation. Decker emphasizes the importance of guarding against hubs with many open channels rather than simply large nodes.In an email exchange, Schroder and Decker discuss the capacity limit on Lightning channels. Schroder expresses concern that large businesses may not be able to operate using Lightning due to small channel sizes and payment limits. He also raises the issue of encouraging people to pay him to route through super nodes. ZmnSCPxj joins the discussion, suggesting that a limit at the protocol level encourages growth towards a mesh network. The conversation further explores the risks of accepting large incoming channels and the potential for attacks.The article delves into the limitations of splitting single payments into multiple invoices and the implementation difficulties it poses. It suggests that multipath payments are a solution to this problem, but complexity increases for the recipient. The article also mentions that the current protocol limits act as training wheels but may not always be necessary. It highlights the ability to change these limits on a per-channel level and perform upgrades using feature bits without forking Lightning. These changes aim to make Lightning more useful for intended applications.The email conversation on the Lightning-dev mailing list focuses on the limitations of payment and channel sizes in Lightning Network. Some argue that the limits are necessary for security reasons and to encourage the growth of the network towards a mesh network. Splitting large payments into smaller ones through multiple invoices is suggested as a temporary solution. The debate also touches on the practice of keeping most funds in cold storage and the risk of rejecting large channel openings. The author of the post shares their thoughts on the limits, expressing surprise at their potential long-term existence and advocating for non-existent or configurable limits.The discussion also touches upon the maximum funding size for channels in Lightning Network. The upper 32 bits must be zeroed out, imposing a limit on the maximum HTLC value. This limitation aims to simplify software design and maintain consistency across coins. The conversation considers the need for this limit in other cryptocurrencies and proposes a "dangerous feature bit" to avoid it.In a discussion between ZmnSCPxj and Andy Schroder, the topic of channel closing costs in relation to cheating was brought up. Schroder explained that the upper 32 bits must be zero, limiting the HTLC value to .04294967295 BTC. This limit was set to prevent significant losses and encourage decentralization. The use of satoshis versus millisatoshis was also discussed.Andy Schroder raised questions about the maximum channel funding size for Bitcoin's Lightning Network. He questioned why the upper 32 bits were being wasted and why the maximum channel size was limited. He speculated whether this low maximum channel size was to encourage decentralization or due to concerns about hot wallets. However, he did find that the maximum funding_satoshis must be less than 2^24 satoshi. Schroder also pointed out the inconsistency in using both satoshis and millisatoshis in the spec.Rusty Russell made the decision to set the upper 32 bits to zero, limiting the maximum HTLC value to 0.04294967295 BTC. This limit is in place for safety and decentralization purposes. ZmnSCPxj believes that a channel limit of 167mBTC is good as it encourages decentralization by encouraging people to make many small channels than one large channel.In a recent discussion on the Lightning-dev mailing list, participants debated the necessity and potential drawbacks of payment limits for Bitcoin Lightning Network. Some viewed these limits as necessary to prevent centralization, while others argued that they hindered important use cases. The possibility of removing the limits in the future was raised, with some suggesting that this would encourage the growth of a mesh network. Concerns were expressed about the risks associated with larger channel sizes and the potential for attacks. Despite differing opinions, most agreed that more experience with the network was needed before making any final decisions.Overall, there are ongoing discussions and debates about the maximum channel funding size, payment limits, and the use of satoshis versus millisatoshis in the Lightning Network. These discussions aim to find a balance between safety, decentralization, and usability as the network continues to evolve.


Updated on: 2023-07-31T19:33:22.851739+00:00