lnwire+htlcswitch: report htlc accept height by joostjager · Pull Request #3414 · lightningnetwork/lnd

joostjager · 2019-08-19T10:26:46Z

Implements lightning/bolts#608 by extending the incorrect_or_unknown_payment_details message with the height at which the failed htlc was accepted.

cfromknecht

first pass looks pretty good, only one rpc-related comment 👍

cfromknecht · 2019-09-05T14:15:02Z

lnrpc/routerrpc/router.proto

perhaps something like accept_height, to mimic what the invoice htlc's will show?

I thought about that, but wasn't sure. The idea of the message Failure is that it isn't a oneof to keep it simple. Fields are reused in different messages. This height is meant as a generic height field associated with a failure. But yes, I could also make it more specific.

halseth

Looks more or less good to me :)

invoices/invoiceregistry.go

lnwire/onion_error_test.go

lnwire/onion_error.go

cfromknecht

LGTM! great commit structure 👌

cfromknecht · 2019-09-11T20:52:02Z

invoices/invoiceregistry_test.go:127:29: undefined: testInvoiceCltvDelta

joostjager · 2019-09-11T22:08:59Z

Test failure fixed

halseth

LGTM 👡

halseth · 2019-09-13T06:22:22Z

invoices/invoiceregistry.go

add similar comment as for the settle case

Comment added. While writing the comment, I also realized that the invoice update could be a bit stricter. See commit channeldb+invoices: make htlc cancelation stricter

invoices/invoiceregistry.go

joostjager · 2019-09-13T14:42:08Z

Requires #3503 to be merged so that the linter is tamed first.

halseth · 2019-09-16T07:22:50Z

invoices/invoiceregistry.go

Does this mean that we can cancel a settled one?

Discussed offline. No htlc should be in the settled state here, because then the invoice state would have been settled. Added extra check to return an error if this happens.

Previously it was possible to cancel a canceled htlc. This would subtract the htlc amount from the invoice amount again.

This commit modifies hodl htlc notification from invoice registry from a single notification per hash to distinct notifications per htlc. This prepares for htlc-specific information (accept height) to be added to the notification.

This is a preparation for passing back the accept height in the incorrect payment details failure message to the sender.

This test relied on specific internals of the failure encode function. Changes to failure message pointer receiver everywhere subtly broke this test.

Extends the invalid payment details failure with the new accept height field. This allows sender to distinguish between a genuine invalid details situation and a delay caused by intermediate nodes.

joostjager requested review from Roasbeef and cfromknecht as code owners August 19, 2019 10:26

joostjager changed the title ~~lnwire+htlcswitch: report htlc accept height [wip] [no review]~~ lnwire+htlcswitch: report htlc accept height [wip] [don't review] Aug 23, 2019

joostjager removed request for Roasbeef and cfromknecht August 23, 2019 06:30

joostjager force-pushed the report-accept-height branch from 630fe2a to 7de1e62 Compare September 5, 2019 09:36

joostjager changed the title ~~lnwire+htlcswitch: report htlc accept height [wip] [don't review]~~ lnwire+htlcswitch: report htlc accept height Sep 5, 2019

joostjager force-pushed the report-accept-height branch 2 times, most recently from 2ca1e27 to ae146db Compare September 5, 2019 11:42

joostjager requested a review from cfromknecht September 5, 2019 13:33

cfromknecht reviewed Sep 5, 2019

View reviewed changes

wpaulino added onion routing spec v0.8.0 labels Sep 6, 2019

wpaulino added this to the 0.8.0 milestone Sep 6, 2019

joostjager requested a review from Roasbeef September 9, 2019 09:45

joostjager mentioned this pull request Sep 9, 2019

lnwire+htlcswitch+routing: report and interpret incorrect_or_unknown_payment_details height #3186

Closed

joostjager force-pushed the report-accept-height branch from ae146db to eab6793 Compare September 9, 2019 11:05

halseth reviewed Sep 10, 2019

View reviewed changes

invoices/invoiceregistry.go Outdated Show resolved Hide resolved

lnwire/onion_error_test.go Outdated Show resolved Hide resolved

lnwire/onion_error.go Outdated Show resolved Hide resolved

joostjager force-pushed the report-accept-height branch from eab6793 to ec8602b Compare September 10, 2019 07:58

joostjager requested a review from halseth September 10, 2019 07:58

joostjager force-pushed the report-accept-height branch from ec8602b to 3894c26 Compare September 11, 2019 11:44

cfromknecht approved these changes Sep 11, 2019

View reviewed changes

joostjager force-pushed the report-accept-height branch from 3894c26 to 91766ed Compare September 11, 2019 22:08

halseth approved these changes Sep 13, 2019

View reviewed changes

joostjager force-pushed the report-accept-height branch 2 times, most recently from 343c376 to 81d6637 Compare September 13, 2019 07:40

joostjager requested a review from halseth September 13, 2019 07:42

joostjager force-pushed the report-accept-height branch from 81d6637 to 783d9db Compare September 13, 2019 18:28

halseth reviewed Sep 16, 2019

View reviewed changes

joostjager added 5 commits September 16, 2019 10:10

channeldb+invoices: make htlc cancelation stricter

49a20a8

Previously it was possible to cancel a canceled htlc. This would subtract the htlc amount from the invoice amount again.

invoices: return accept height in hodl event

d3e206e

This is a preparation for passing back the accept height in the incorrect payment details failure message to the sender.

lnwire/test: fix TestFailIncorrectDetailsOptionalAmount

1de08c4

This test relied on specific internals of the failure encode function. Changes to failure message pointer receiver everywhere subtly broke this test.

lnwire+htlcswitch: report height for invalid payment details failure

f60e4b1

Extends the invalid payment details failure with the new accept height field. This allows sender to distinguish between a genuine invalid details situation and a delay caused by intermediate nodes.

joostjager force-pushed the report-accept-height branch from 783d9db to f60e4b1 Compare September 16, 2019 08:10

joostjager merged commit 95502da into lightningnetwork:master Sep 16, 2019

Conversation

joostjager commented Aug 19, 2019 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

cfromknecht left a comment

Choose a reason for hiding this comment

Uh oh!

cfromknecht Sep 5, 2019

Choose a reason for hiding this comment

Uh oh!

joostjager Sep 5, 2019

Choose a reason for hiding this comment

Uh oh!

halseth left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

cfromknecht left a comment

Choose a reason for hiding this comment

Uh oh!

cfromknecht commented Sep 11, 2019

Uh oh!

joostjager commented Sep 11, 2019

Uh oh!

halseth left a comment

Choose a reason for hiding this comment

Uh oh!

halseth Sep 13, 2019

Choose a reason for hiding this comment

Uh oh!

joostjager Sep 13, 2019

Choose a reason for hiding this comment

Uh oh!

Uh oh!

joostjager commented Sep 13, 2019

Uh oh!

halseth Sep 16, 2019

Choose a reason for hiding this comment

Uh oh!

joostjager Sep 16, 2019

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

joostjager commented Aug 19, 2019 •

edited

Loading