bitcoind RPC: Skip validating header chain for newly setup nodes

[tnull]

When we just setup a new node, all chain listeners will still be synced up to the genesis hash. In order to allow for faster startup, we skip validating the entirety of the header chain by continuing with the current best block polled from bitcoind.

Note that this doesn't noticeably change our security model as we're trusting the chain data source, for example to not censor us, anyways. And, we're of course still validating the PoW for the best block we're polling.

[tnull]

Hmm, nevermind. I think this currently doesn't work. At least AFAICT there is no way to skip header validation after the fact, i.e., once the ChannelManager had been initialized with the genesis hash. And of course, we can't poll during/prior to initialization in the builder as we don't have a runtime available.

One way to consider shortening the header validation would be to eventually introducing checkpoints, i.e., initialize the listeners with a more recent best block on mainnet.

[tnull]

When we just setup a new node, all chain listeners will still be synced up to the genesis hash. In order to allow for faster startup, we skip validating the entirety of the header chain by continuing with the current best block polled from bitcoind.

Note that this doesn't noticeably change our security model as we're trusting the chain data source, for example to not censor us, anyways. And, we're of course still validating the PoW for the best block we're polling.

[tnull]

Hmm, nevermind. I think this currently doesn't work. At least AFAICT there is no way to skip header validation after the fact, i.e., once the ChannelManager had been initialized with the genesis hash. And of course, we can't poll during/prior to initialization in the builder as we don't have a runtime available.

One way to consider shortening the header validation would be to eventually introducing checkpoints, i.e., initialize the listeners with a more recent best block on mainnet.

[tnull]

When we just setup a new node, all chain listeners will still be synced up to the genesis hash. In order to allow for faster startup, we skip validating the entirety of the header chain by continuing with the current best block polled from bitcoind.

Note that this doesn't noticeably change our security model as we're trusting the chain data source, for example to not censor us, anyways. And, we're of course still validating the PoW for the best block we're polling.

[tnull]

Hmm, nevermind. I think this currently doesn't work. At least AFAICT there is no way to skip header validation after the fact, i.e., once the ChannelManager had been initialized with the genesis hash. And of course, we can't poll during/prior to initialization in the builder as we don't have a runtime available.

One way to consider shortening the header validation would be to eventually introducing checkpoints, i.e., initialize the listeners with a more recent best block on mainnet.

[tnull]

When we just setup a new node, all chain listeners will still be synced up to the genesis hash. In order to allow for faster startup, we skip validating the entirety of the header chain by continuing with the current best block polled from bitcoind.

Note that this doesn't noticeably change our security model as we're trusting the chain data source, for example to not censor us, anyways. And, we're of course still validating the PoW for the best block we're polling.

[tnull]

Hmm, nevermind. I think this currently doesn't work. At least AFAICT there is no way to skip header validation after the fact, i.e., once the ChannelManager had been initialized with the genesis hash. And of course, we can't poll during/prior to initialization in the builder as we don't have a runtime available.

One way to consider shortening the header validation would be to eventually introducing checkpoints, i.e., initialize the listeners with a more recent best block on mainnet.

[tnull]

When we just setup a new node, all chain listeners will still be synced up to the genesis hash. In order to allow for faster startup, we skip validating the entirety of the header chain by continuing with the current best block polled from bitcoind.

Note that this doesn't noticeably change our security model as we're trusting the chain data source, for example to not censor us, anyways. And, we're of course still validating the PoW for the best block we're polling.

[tnull]

Hmm, nevermind. I think this currently doesn't work. At least AFAICT there is no way to skip header validation after the fact, i.e., once the ChannelManager had been initialized with the genesis hash. And of course, we can't poll during/prior to initialization in the builder as we don't have a runtime available.

One way to consider shortening the header validation would be to eventually introducing checkpoints, i.e., initialize the listeners with a more recent best block on mainnet.

[tnull]

When we just setup a new node, all chain listeners will still be synced up to the genesis hash. In order to allow for faster startup, we skip validating the entirety of the header chain by continuing with the current best block polled from bitcoind.

Note that this doesn't noticeably change our security model as we're trusting the chain data source, for example to not censor us, anyways. And, we're of course still validating the PoW for the best block we're polling.

[tnull]

Hmm, nevermind. I think this currently doesn't work. At least AFAICT there is no way to skip header validation after the fact, i.e., once the ChannelManager had been initialized with the genesis hash. And of course, we can't poll during/prior to initialization in the builder as we don't have a runtime available.

One way to consider shortening the header validation would be to eventually introducing checkpoints, i.e., initialize the listeners with a more recent best block on mainnet.

[tnull]

When we just setup a new node, all chain listeners will still be synced up to the genesis hash. In order to allow for faster startup, we skip validating the entirety of the header chain by continuing with the current best block polled from bitcoind.

Note that this doesn't noticeably change our security model as we're trusting the chain data source, for example to not censor us, anyways. And, we're of course still validating the PoW for the best block we're polling.

[tnull]

Hmm, nevermind. I think this currently doesn't work. At least AFAICT there is no way to skip header validation after the fact, i.e., once the ChannelManager had been initialized with the genesis hash. And of course, we can't poll during/prior to initialization in the builder as we don't have a runtime available.

One way to consider shortening the header validation would be to eventually introducing checkpoints, i.e., initialize the listeners with a more recent best block on mainnet.

[tnull]

When we just setup a new node, all chain listeners will still be synced up to the genesis hash. In order to allow for faster startup, we skip validating the entirety of the header chain by continuing with the current best block polled from bitcoind.

Note that this doesn't noticeably change our security model as we're trusting the chain data source, for example to not censor us, anyways. And, we're of course still validating the PoW for the best block we're polling.

[tnull]

Hmm, nevermind. I think this currently doesn't work. At least AFAICT there is no way to skip header validation after the fact, i.e., once the ChannelManager had been initialized with the genesis hash. And of course, we can't poll during/prior to initialization in the builder as we don't have a runtime available.

One way to consider shortening the header validation would be to eventually introducing checkpoints, i.e., initialize the listeners with a more recent best block on mainnet.