feat: track in-flight amount for pending payments #4366

shruti2522 · 2026-01-31T03:50:20Z

adds inflight_msat to RecentPaymentDetails::Pending to expose the amount currently locked in HTLCs. this allows wallets to avoid balance flicker during payment retries by distinguishing between inflight funds and funds waiting to retry

ldk-reviews-bot · 2026-01-31T03:50:23Z

👋 I see @joostjager was un-assigned.
If you'd like another reviewer assignment, please click here.

codecov · 2026-01-31T05:36:36Z

Codecov Report

❌ Patch coverage is 94.03974% with 9 lines in your changes missing coverage. Please review.
✅ Project coverage is 86.08%. Comparing base (e9ce486) to head (0934df6).
⚠️ Report is 163 commits behind head on main.

Files with missing lines	Patch %	Lines
lightning/src/ln/channelmanager.rs	93.28%	8 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #4366      +/-   ##
==========================================
- Coverage   86.54%   86.08%   -0.47%     
==========================================
  Files         158      156       -2     
  Lines      103166   103536     +370     
  Branches   103166   103536     +370     
==========================================
- Hits        89287    89126     -161     
- Misses      11456    11898     +442     
- Partials     2423     2512      +89

Flag	Coverage Δ
fuzzing	`?`
tests	`86.08% <94.03%> (+0.24%)`	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

ldk-reviews-bot · 2026-02-02T04:01:06Z

🔔 1st Reminder

Hey @wpaulino! This PR has been waiting for your review.
Please take a look when you have a chance. If you're unable to review, please let us know so we can find another reviewer.

ldk-reviews-bot · 2026-02-04T04:01:38Z

🔔 2nd Reminder

Hey @wpaulino! This PR has been waiting for your review.
Please take a look when you have a chance. If you're unable to review, please let us know so we can find another reviewer.

wpaulino

Please include the description of the changes in the commit message as well

wpaulino · 2026-02-05T19:08:12Z

lightning/src/ln/channelmanager.rs

+		/// Amount (in msat) currently locked in HTLCs.
+		///
+		/// `total_msat - inflight_msat` gives the amount waiting to be retried
+		/// Reserve both from spendable balance.


This line is not very clear to me, can you rephrase it?

wpaulino · 2026-02-05T19:08:56Z

lightning/src/ln/payment_tests.rs

+
+	let payment_amt = 100_000;
+	let (payment_preimage, _payment_hash, _, payment_id) =
+		route_payment(&nodes[0], &[&nodes[1], &nodes[2]], payment_amt);


Let's route another payment but still only claim one so we can assert the in-flight amount of the remaining payment

ldk-reviews-bot · 2026-02-05T19:09:36Z

👋 The first review has been submitted!

Do you think this PR is ready for a second reviewer? If so, click here to assign a second reviewer.

ldk-reviews-bot · 2026-02-05T19:10:02Z

✅ Added second reviewer: @joostjager

TheBlueMatt

I'm not convinced this API is quite sufficient for what we need. We really want a single, consistent view of users' current balance. The problem with fetching information from two separate sources (here ChannelMonitor(s) + ChannelManager) is that we end up racing retries across them - calling them both in succession may double- or under-count pending payments. I'm not quite sure with the right answer is, maybe we need to list HTLCs in pending payments and have some utility that removes duplicated entries, or maybe we need to include total payment amounts in the ChannelMonitor balance list and have some logic to compare the pending payments list to remove payments that have partially MPP-timed-out.

TheBlueMatt · 2026-02-07T16:26:24Z

We'll have to explore it a bit, but I think the second option is compelling - we already have payment info in the monitors (I hope enough) so communicating that back may suffice. The only case we'd maybe want to worry about is whether we over-state the in-flight amount for payments where we've given up retrying but there are stuck MPP parts. In that case we'll need the pending payment info like you exposed here, but may need more information about the retry state.

shruti2522 · 2026-02-09T13:08:52Z

We'll have to explore it a bit, but I think the second option is compelling - we already have payment info in the monitors (I hope enough) so communicating that back may suffice. The only case we'd maybe want to worry about is whether we over-state the in-flight amount for payments where we've given up retrying but there are stuck MPP parts. In that case we'll need the pending payment info like you exposed here, but may need more information about the retry state.

Got it, will look into this and follow up

Tracks actual in-flight amounts for pending payments to enable accurate balance calculations in wallets - Added payment_id to Balance::MaybeTimeoutClaimableHTLC so monitors can report which payment each HTLC belongs to - Added is_retryable to RecentPaymentDetails::Pending to distinguish active retries from stuck HTLCs - Added reconcile_inflight_payments() using ChannelMonitor as authoritative source for amounts to prevent double-counting - Added aggregate_outbound_htlcs_by_payment() to ChannelMonitor for aggregating HTLCs by payment This solves the race condition when fetching from ChannelMonitor and ChannelManager separately during retries

shruti2522 · 2026-02-11T14:46:21Z

I went with the second option, added payment info to the monitors, using them as the source of truth for in-flight amounts. Added payment_id to Balance::MaybeTimeoutClaimableHTLC so the monitors can report which payment each HTLC belongs to.

For the stuck MPP parts concern, I added is_retryable to the pending payment info to track retry state. That way we can tell the difference between payment is actively retrying vs we've given up but HTLCs are still stuck

Added reconcile_inflight_payments() utility that takes both sources and gives back a consistent view, it uses the monitor amounts (what's actually in channels) but pulls the retry status from ChannelManager. Returns a simple map of PaymentId -> (amount, is_retryable). This should eliminate the double-counting race since we're not trying to add up amounts from two separate calls.

TheBlueMatt · 2026-02-11T18:38:23Z

lightning/src/chain/channelmonitor.rs

+	/// returns msats.
+	///
+	/// [`reconcile_inflight_payments`]: crate::ln::channelmanager::reconcile_inflight_payments
+	pub fn aggregate_outbound_htlcs_by_payment(balances: &[Balance]) -> HashMap<PaymentId, u64> {


This is a weird public API?

TheBlueMatt · 2026-02-11T18:39:31Z

pending_changelog/3374-pending-retry-amount.txt

@@ -0,0 +1,27 @@
+## API Updates


No, please keep pending changelog entries condensed. Max a line or two wrapped at 80 chars. There should almost certainly only be one or two entries, each one or two sentences for a PR like this.

TheBlueMatt · 2026-02-11T18:40:17Z

lightning/src/ln/channelmanager.rs

+/// ```
+pub fn reconcile_inflight_payments(
+	balances: &[Balance], recent_payments: &[RecentPaymentDetails],
+) -> HashMap<PaymentId, (u64, bool)> {


Do you feel like this is the right API? I really want to answer the question "what is my balance" not "what is the amount being sent for pending payments".

shruti2522 force-pushed the feat/pending-balance branch from 4c7a57d to 0c440b6 Compare January 31, 2026 03:57

ldk-reviews-bot requested a review from wpaulino January 31, 2026 04:00

wpaulino reviewed Feb 5, 2026

View reviewed changes

ldk-reviews-bot requested a review from joostjager February 5, 2026 19:10

TheBlueMatt reviewed Feb 7, 2026

View reviewed changes

TheBlueMatt removed the request for review from joostjager February 7, 2026 16:25

shruti2522 force-pushed the feat/pending-balance branch 2 times, most recently from 048827d to ca6d216 Compare February 11, 2026 14:13

shruti2522 force-pushed the feat/pending-balance branch from ca6d216 to 0934df6 Compare February 11, 2026 14:27

TheBlueMatt reviewed Feb 11, 2026

View reviewed changes

feat: track in-flight amount for pending payments #4366

Are you sure you want to change the base?

feat: track in-flight amount for pending payments #4366

Uh oh!

Conversation

shruti2522 commented Jan 31, 2026

Uh oh!

ldk-reviews-bot commented Jan 31, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

codecov bot commented Jan 31, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

ldk-reviews-bot commented Feb 2, 2026

Uh oh!

ldk-reviews-bot commented Feb 4, 2026

Uh oh!

wpaulino left a comment

Choose a reason for hiding this comment

Uh oh!

wpaulino Feb 5, 2026

Choose a reason for hiding this comment

Uh oh!

wpaulino Feb 5, 2026

Choose a reason for hiding this comment

Uh oh!

ldk-reviews-bot commented Feb 5, 2026

Uh oh!

ldk-reviews-bot commented Feb 5, 2026

Uh oh!

TheBlueMatt left a comment

Choose a reason for hiding this comment

Uh oh!

TheBlueMatt commented Feb 7, 2026

Uh oh!

shruti2522 commented Feb 9, 2026

Uh oh!

shruti2522 commented Feb 11, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

TheBlueMatt Feb 11, 2026

Choose a reason for hiding this comment

Uh oh!

TheBlueMatt Feb 11, 2026

Choose a reason for hiding this comment

Uh oh!

TheBlueMatt Feb 11, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

ldk-reviews-bot commented Jan 31, 2026 •

edited

Loading

codecov bot commented Jan 31, 2026 •

edited

Loading

shruti2522 commented Feb 11, 2026 •

edited

Loading