feat(consensus): Send duplicate slots from shred network to consensus (revival of #1027) by InKryption · Pull Request #1172 · Syndica/sig

InKryption · 2026-01-09T16:05:22Z

No description provided.

codecov · 2026-01-09T16:36:11Z

Codecov Report

❌ Patch coverage is 89.81571% with 105 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/shred_network/duplicate_shred_listener.zig	89.48%	57 Missing ⚠️
...hred_network/collector/duplicate_shred_handler.zig	90.65%	34 Missing ⚠️
src/shred_network/service.zig	67.85%	9 Missing ⚠️
src/gossip/service.zig	44.44%	5 Missing ⚠️

Files with missing lines	Coverage Δ
src/core/epoch_tracker.zig	`94.92% <100.00%> (ø)`
src/core/leader_schedule.zig	`96.62% <100.00%> (ø)`
src/gossip/data.zig	`76.41% <100.00%> (-0.11%)`	⬇️
src/ledger/Reader.zig	`89.68% <100.00%> (+0.12%)`	⬆️
src/ledger/ResultWriter.zig	`93.71% <100.00%> (+0.82%)`	⬆️
src/ledger/shred.zig	`90.28% <100.00%> (-0.42%)`	⬇️
src/shred_network/collector/shred_receiver.zig	`83.60% <100.00%> (+0.18%)`	⬆️
src/shred_network/collector/shred_tracker.zig	`87.25% <100.00%> (+0.56%)`	⬆️
src/gossip/service.zig	`86.61% <44.44%> (-0.27%)`	⬇️
src/shred_network/service.zig	`84.55% <67.85%> (-5.84%)`	⬇️
... and 2 more

... and 1 file with indirect coverage changes

🚀 New features to boost your workflow:

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

src/shred_network/collector/duplicate_shred_handler.zig

src/shred_network/duplicate_shred_listener.zig

src/shred_network/collector/duplicate_shred_handler.zig

yewman · 2026-01-30T15:44:29Z

src/shred_network/collector/duplicate_shred_handler.zig

+                    // TODO: Check feature flag for chained_merkle_conflict_duplicate_proofs
+                    // For now, we'll store it unconditionally. When feature checking is implemented,
+                    // this should check: if (!chained_merkle_conflict_duplicate_proofs) continue;
+
+                    // Although this proof can be immediately stored on detection, we wait until
+                    // here in order to check the feature flag, as storage in ledger can
+                    // preclude the detection of other duplicate proofs in this slot


Can we not pass the feature set and check it now?

I've looked through it, and it's actually not entirely clear to me what the best way would be to pass the feature set through.

What issue do you see with passing the feature set in? We could potentially add it to EpochInfo in the EpochTracker unless we move the slot tracker outside replay. While the feature set is constant for most of the duration of an epoch, it can be different across forks in an epoch until the first slot in said epoch is rooted. This could be addressed on top of epoch-state-tracking if required. Another thing to check is whether this feature is activated on all clusters. If it is, just add a note here and we can remove the check.

src/shred_network/collector/duplicate_shred_handler.zig

Drop dep on gossip table

Co-authored-by: Drew Nutter <dnut@users.noreply.github.com>

InKryption requested review from dnut, ultd and yewman as code owners January 9, 2026 16:05

github-project-automation bot added this to Sig Jan 9, 2026

github-project-automation bot moved this to 🏗 In progress in Sig Jan 9, 2026

InKryption force-pushed the ink/send-duplicate-slot-to-consensus branch 3 times, most recently from 920dcef to 5a16138 Compare January 13, 2026 14:18

dnut mentioned this pull request Jan 14, 2026

feat(consensus): Send duplicated slot from shred network to consensus #1027

Closed

InKryption force-pushed the ink/send-duplicate-slot-to-consensus branch 2 times, most recently from a974315 to 994f7c0 Compare January 16, 2026 20:25

dnut requested changes Jan 30, 2026

View reviewed changes

github-project-automation bot moved this from 🏗 In progress to 👀 In review in Sig Jan 30, 2026

yewman reviewed Jan 30, 2026

View reviewed changes

dnut linked an issue Feb 2, 2026 that may be closed by this pull request

DuplicateShredListener #1025

Open

InKryption force-pushed the ink/send-duplicate-slot-to-consensus branch 6 times, most recently from e1ffafe to 6ee04e6 Compare February 4, 2026 07:03

dnut previously approved these changes Feb 6, 2026

View reviewed changes

InKryption dismissed dnut’s stale review via 1e2810a February 13, 2026 22:01

InKryption force-pushed the ink/send-duplicate-slot-to-consensus branch from e2a2b4d to 1e2810a Compare February 13, 2026 22:01

InKryption mentioned this pull request Feb 13, 2026

Sanitize duplicate proofs #1231

Open

InKryption force-pushed the ink/send-duplicate-slot-to-consensus branch 2 times, most recently from eccc9fb to b6ca584 Compare February 13, 2026 22:30

dadepo added 2 commits February 14, 2026 19:40

Send duplicates to consensus

2805e8c

Add storeDuplicateSlot & isDuplicateSlot + tests

81dd4be

dadepo and others added 21 commits February 14, 2026 19:40

Store/check duplicate in blockstore

0413e1c

Send duplicate proof to gossip

4137283

Add duplicate shred listener

939ca27

Some code hygiene refactors & renames

f5eb6e5

use more cache

1e54411

Pass the gossip table

56b2580

Cleaner deps passing to DuplicateShredHandler

f29fddb

Move duplicate handling to its own file

a40e10a

Renamed DuplicateShredHandler in gossip to GossipDuplicateShredHandler

d04ead2

test dupe shred listener,receiver,handler,tracker

bac58d8

Do not depend on the whole gossip service

49d9300

Keep track of slots processed and dont consult gossip table

74bdff8

Drop dep on gossip table

Make gossip send duplicate shreds directly

c4a09ad

Use usize instead of u64

8ce5b6a

Co-authored-by: Drew Nutter <dnut@users.noreply.github.com>

use actual shred type

ec36094

Co-authored-by: Drew Nutter <dnut@users.noreply.github.com>

Stop ignoring OOM

f330dfe

Remove unused constant

408df1e

correct errdefer

bd3fedd

Add feature_set to EpochInfo

8860131

Handle chained_merkle_conflict_duplicate_proofs

31c0bd8

Link to issue for troubling TODO

500e989

InKryption force-pushed the ink/send-duplicate-slot-to-consensus branch from b6ca584 to 500e989 Compare February 14, 2026 18:41

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat(consensus): Send duplicate slots from shred network to consensus (revival of #1027)#1172

feat(consensus): Send duplicate slots from shred network to consensus (revival of #1027)#1172
InKryption wants to merge 23 commits intomainfrom
ink/send-duplicate-slot-to-consensus

InKryption commented Jan 9, 2026

Uh oh!

codecov bot commented Jan 9, 2026 •

edited

Loading

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

yewman Jan 30, 2026

Uh oh!

InKryption Feb 5, 2026

Uh oh!

yewman Feb 5, 2026 •

edited

Loading

Uh oh!

InKryption Feb 13, 2026

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Conversation

InKryption commented Jan 9, 2026

Uh oh!

codecov bot commented Jan 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

yewman Jan 30, 2026

Choose a reason for hiding this comment

Uh oh!

InKryption Feb 5, 2026

Choose a reason for hiding this comment

Uh oh!

yewman Feb 5, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

InKryption Feb 13, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

codecov bot commented Jan 9, 2026 •

edited

Loading

yewman Feb 5, 2026 •

edited

Loading