And make most places warn or error. Places where we're operating
on a channel tend to be "warn and close connection" since we want to
forget the mistake they just sent, and closing the connection does that.
We now use the same words everywhere:
1. "fail channel" means to go onchain (if necessary).
2. "send `error`" means to send an error message.
3. "send `warning`" means to send a warning message.
4. "close connection" means close the connection.
These are all spelled out explicitly, rather than having "fail channel"
imply sending an error packet, for example.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
In this commit, we add a new feature bit to gate the new explicit
channel type funding via the new `channel_type` TLV. The addition of
this new bit allows peers to seek out other peers that understand the
new explicit channel negotiation. This is useful in practice, as it
allows peers to avoid needing to "downgrade" the feature bits advertised
at the connection level due to one peer not understanding a new
required feature bit while it has a channel with a connecting peer.
Such a workaround is already deployed on the network between lnd peers
and certain eclair peers, as the `lnd` peers require static key, but the
feature bit is unknown to eclair peers. This situation (forced
downgrade) is undesirable, as until the connected peer updates (or the
channel is closed) and "worst" feature bit set must always be advertised
in order to maintain connectivity.
The other benefit of adding this feature bit is that it allows
implementations to simplify their code by ensuring that the new feature
will be used before sending any messages that include or reference that
feature. Without a feature bit, peers are instead forced to essentially
guess if a peer understands that feature, with logic to be able to "bail
out" of an invalid state.
The addition of this feature bit matches the prior precedent of adding
feature bits when new fields in the channel negotiation message (last
one was upfront shutdown) are added.
Since HTLCs below this amount will not appear in the commitment tx, they
are effectively converted to miner fees. The peer could use this to grief
you by broadcasting its commitment once it contains a lot of dust HTLCs.
Add network dust thresholds computation details, as implemented in Bitcoin
Core's default relay policy.
Drop non-segwit support in shutdown: this allows dust limit to go as low
as 354 sats without creating relay issues with default node policies.
We add a requirement that dust limit cannot be lower than 354 sats.
This ensures implementers don't have to figure this subtlety on their own.
Fixes#696 and #905
With anchor outputs, we can keep the commit tx feerate lower than the real
on-chain feerate. That means that when closing the channel, the resulting
fee will not necessarily be lower than the current commit tx fee, this
requirement doesn't make sense to be strict.
Both sides can optionally include a preferred fee range in their
`closing_signed`. This lets their peer know what fees they find acceptable
and simplifies the closing negotiation.
This is extracted from channel_upgrade (#868), but used for opening
negotiation as suggested by @roasbeef on the last spec meeting.
It's a trivial change, fully backwards compatible, but now each channel
has a channel_type, which defines its behavior, rather than an ad-hoc
set of "sticky" feature bits. It also means both peers can *support* a
feature without endorsing it.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We previously had a restriction on HTLC amounts to avoid big losses during
the early phases of the network, but it shouldn't be necessary anymore.
As long as we honor `max_htlc_value_in_flight_msat` and implementations
provide safe defaults for that parameter, we don't need that additional
restriction.
In bitcoin 0.19.0, standardness rules are going to be relaxed to allow
future witness versions. Once this is widely deployed, it will be safe
to accept them, smoothing use of future segwit versions.
See: https://github.com/bitcoin/bitcoin/pull/15846
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Routing nodes have an incentive to use low fees when opening channels to
ensure their activity is economically viable.
However, when a funding transaction takes too long to confirm, the fundee
may have forgotten the channel. In that case the funder is forced to
broadcast the first commit tx to get his funds back and then open a new
channel, which is costly.
We can avoid this issue by simply knowing how long the fundee will wait,
and ensuring the funding tx confirms before that period ends. We set this
timeout to 2016 blocks (2 weeks).
The existing requirements were not specifying the case where both a
`commitment_signed` and `revoke_and_ack` need to be retransmitted.
This is an important case to specify because if the relative order is not
preserved, the channel will close.
Fixes#794
Many channels use a value below 6, which is really insecure (there are
more than 2k such channels on mainnet).
While less risky, there are more than 7k channels with a value below 12.
This indicates that the spec should probably make the risks a bit more
clear to help guide node operators.
This commit extends the specification with a new commitment format that
adds two anchor outputs to the commitment transaction. Anchor outputs
are a safety feature that allows a channel party to unilaterally increase
the fee of the commitment transaction using CPFP and ensure timely
confirmation on the chain. There is no cooperation required from the
remote party.
Add an additional "reserve" for funders on top of the real reserve to
avoid getting in a state where the channel is unusable because
of the increased commit tx cost of a new HTLC.
Requirements are only added for the funder sending an HTLC.
Fundee receiving HTLCs may choose to verify that funders apply
this, but it may lead to an unusable UX.
Fixes#728.
The spec already prepared a hook to add additional information to existing
messages (additional bytes at the end of a message must be ignored).
Since we're using TLV in many places, it makes sense to use that optional
additional space at the end of each message to allow an optional tlv stream.
This requires making a few previously optional fields mandatory:
- channel_reestablish commitment points: it makes sense to always include those
regardless of whether `option_dataloss_protect` or `option_static_remotekey` are set.
- option_upfront_shutdown_script: if you're not using one, just set the length to 0.
That field is moved to a TLV record because luckily, the resulting bytes are the same.
This provides more flexibility to later remove the requirement of making this field mandatory.
No need to change the `channel_update`'s `htlc_maximum_msat` because
the `message_flags` encode its presence/absence.
It can still be either included or omitted without causing issues to the extension stream.
The spec already prepared a hook to add additional information to existing
messages (additional bytes at the end of a message must be ignored).
Since we're using TLV in many places, it makes sense to use that optional
additional space at the end of each message to allow an optional tlv stream.
This requires making a few previously optional fields mandatory:
- channel_reestablish commitment points: it makes sense to always include those
regardless of whether `option_dataloss_protect` or `option_static_remotekey` are set.
- option_upfront_shutdown_script: if you're not using one, just set the length to 0.
That field is moved to a TLV record because luckily, the resulting bytes are the same.
This provides more flexibility to later remove the requirement of making this field mandatory.
No need to change the `channel_update`'s `htlc_maximum_msat` because
the `message_flags` encode its presence/absence.
It can still be either included or omitted without causing issues to the extension stream.
ECDSA signatures in Bitcoin are DER-encoded but public keys are not.
The compressed format for public keys is for example standardized in
Sections 2.3.3 and 2.3.4 of
Standards for Efficient Cryptography, SEC 1: Elliptic Curve
Cryptography, Certicom Research, Version 2, 2009,
https://www.secg.org/sec1-v2.pdf
We simply specify, in each case, where they will appear ("Context").
Because `globalfeatures` is already in use, we fold that into the
renamed `localfeatures` field to unify them (now called `features`),
but dissuade further use.
Note also: we REQUIRE minimal `features` field in
channel_announcement, since otherwise both sides of channel will not
agree and not be able to create their signatures!
Consider these theoretical future features:
`opt_dlog_chan`: a new channel type which uses a new discrete log HTLC
type, but can't support traditional HTLC:
* `init`: presents as odd (optional) or even (if traditional channels
not supported)
* `node_announcement`: the same as above, so you can seek suitable peers.
* `channel_announcement`: presents as even (compulsory), since users need
to use the new HTLCs.
`opt_wumbochan`: a node which allows channels > 2^24 satoshis:
* `init`: presents as odd (optional), or maybe even (if you only want
giant channels)
* `node_announcement`: the same as above, so you can seek suitable peers.
* `channel_announcement`: not present, since size of channel indicates
capacity.
`opt_wumbohtlc`: a channel which allows HTLCs > 2^32 millisatoshis:
* `init`: presents as odd (optional), or even (compulsory)
* `node_announcement`: the same as above, so you can seek suitable peers.
* `channel_announcement`: odd (optional) since you can use the channel
without understanding what this option means.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Co-Authored-By: Bastien Teinturier <31281497+t-bast@users.noreply.github.com>
We might argue this does not apply if you set `minimum_depth` to 0, since
you're assuming trust (TurboChannels-style), but it needs to be specified.
See: CVE-2019-12998 / CVE-2019-12999 / CVE-2019-13000
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This separates out the static remotekey changes from the more ambitious
option_simplified_commitment (which also included pushme outputs and
bring-your-own-fee for HTLC outputs).
As per http://www.erisian.com.au/meetbot/lightning-dev/2019/lightning-dev.2019-09-02-20.06.html
Thanks to everyone for feedback: @araspitzu @roasbeef @bitconner
Suggested-by: @roasbeef
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
* Added descriptions of how a 2-of-2 multisignature verification is used for enforcing timelocks when timing out on-chain offered HTLCs as well as spending on-chain received HTLCs in the success case.
(No spec change, just wording)
The "local" and "remote" here are just *confusing*. Each side says
where it's at, and the other side retransmits based on that.
We could call it 'number_of_next_commitment_i_expect_to_receive' and
'number_of_next_revocation_i_expect_to_receive' but that's getting
silly.
These names were a major source of confusion while writing tests!
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
It's trivial to make types->lengths, but not so much the other way.
The types I used here are the ones I found useful in implementation, and
I think add some clarity, though we can certainly argue about them.
There's no normative changes to the spec in here.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Saying "the risk is only to the node *accepting* the HTLC" is confusing
because merely accepting an HTLC is risk-free. The risk comes from
accepting *responsibility to route the payment*, i.e. offering an HTLC
of your own in the next channel on the path, where a too-small
difference in the HTLC values could end up with you cheated out of a
payment.
This revised paragraph hopefully makes that clearer.
Uses more specific and consistent language ("the"/"that" instead of "an"
where possible). Also helps avoid confusion with unrelated HTLCs that
serve other payments but are shared by the same two nodes.
We express it has how the outputs are ordered, but the only way you can
detect that is by the htlc_signatures order, which is the part which really
matters.
I finally reproduced this, BTW, which is why I'm digging it up!
Closes: #448
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
