-
Notifications
You must be signed in to change notification settings - Fork 345
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Route blinding MVP #2413
Route blinding MVP #2413
Conversation
f4dc1aa
to
8ed2025
Compare
Codecov ReportPatch coverage:
❗ Your organization needs to install the Codecov GitHub app to enable full functionality. Additional details and impacted files@@ Coverage Diff @@
## main #2413 +/- ##
==========================================
+ Coverage 90.45% 90.61% +0.15%
==========================================
Files 112 113 +1
Lines 58564 59002 +438
Branches 58564 59002 +438
==========================================
+ Hits 52976 53466 +490
+ Misses 5588 5536 -52
☔ View full report in Codecov by Sentry. |
8ed2025
to
223e2de
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Rather than splitting the pr into handle-it-all-then-add-error-handling, can we split it into "handle sending, with all error handling, handle receiving, with all error handling, and then handle forwarding, with all error handling"?
I'm still looking into this but we can't materially test any error handling until we have forwarding, because 1-hop blinded paths are supposed to error the same as unblinded payments, i.e. not wipe the error or return malformed. Another option would be to split forwarding+error handling off into its own PR and land sending/1-hop receiving first. |
223e2de
to
952e537
Compare
Removed forwarding support for now, so all forwarding + error handling will be done in #2540 (should be updated tomorrow). |
c2715bf
to
80405b4
Compare
80405b4
to
43a54c4
Compare
Rebased and removed support for receiving to multi-hop blinded paths. This way we can ship an MVP BOLT 12 in 117 and complete route blinding support in 118, to avoid holding up the current release. |
43a54c4
to
4b39220
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good to squash!
This will be used in the next commit to deserialize encrypted TLVs for receiving to 1-hop blinded paths.
Useful until forwarding and receiving to multi-hop blinded paths is supported.
4b39220
to
ebdc4ae
Compare
let ev = remove_first_msg_event_to_node(&nodes[1].node.get_our_node_id(), &mut events); | ||
pass_along_path(&nodes[0], expected_route[0], amt_msat, payment_hash.clone(), | ||
Some(payment_secret), ev.clone(), false, None); | ||
|
||
let ev = remove_first_msg_event_to_node(&nodes[2].node.get_our_node_id(), &mut events); | ||
pass_along_path(&nodes[0], expected_route[1], amt_msat, payment_hash.clone(), | ||
Some(payment_secret), ev.clone(), true, None); | ||
claim_payment_along_route(&nodes[0], expected_route, false, payment_preimage); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
nit: ev
clones not needed
amt_msat: u64, | ||
total_msat: u64, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Heh, can we get away from the legacy names and use more descriptive names since we're adding new code here? Something about mpp_part_amt_msat and total_payment_msat?
@@ -1234,7 +1234,9 @@ impl OutboundPayments { | |||
if route.paths.len() < 1 { | |||
return Err(PaymentSendFailure::ParameterError(APIError::InvalidRoute{err: "There must be at least one path to send over".to_owned()})); | |||
} | |||
if recipient_onion.payment_secret.is_none() && route.paths.len() > 1 { | |||
if recipient_onion.payment_secret.is_none() && route.paths.len() > 1 | |||
&& !route.paths.iter().any(|p| p.blinded_tail.is_some()) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Shouldnt this be all?
return Err(DecodeError::InvalidValue) | ||
} | ||
let enc_tlvs = encrypted_tlvs_opt.ok_or(DecodeError::InvalidValue)?.0; | ||
let enc_tlvs_ss = node_signer.ecdh(Recipient::Node, &blinding_point, None) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I feel like we should do blinded paths to a random key and not reuse the node id. That wouldn't avoid the new trait pass cause we need it for forwards, but for nodes with the node id on a separate hardware device/VLS we'd avoid the need to call out to the hardware device here for receives, which I think would be very nice.
Support sending to blinded payment paths and receiving to 1-hop paths. Partially addresses #1970. Error handling, forwarding and receiving to multi-hop blinded paths will be completed in follow-up.
Based on #2411, #2412, #2128, #2459, #2514, #2503