replay: do not set block_id for leader_blocks

[AshwinSekar]

Problem

For leader banks, broadcast is responsible for setting the block_id once the block has finished shredding:

agave/turbine/src/broadcast_stage/broadcast_utils.rs

Lines 200 to 209 in a7c8d68

	/// Set the block id on the bank and send it for consideration in voting
	pub(super) fn set_block_id_and_send(
	migration_status: &MigrationStatus,
	votor_event_sender: &VotorEventSender,
	bank: Arc<Bank>,
	block_id: Hash,
	) -> Result<()> {
	bank.set_block_id(Some(block_id));
	if bank.is_frozen() && migration_status.should_send_votor_event(bank.slot()) {
	votor_event_sender.send(VotorEvent::Block(CompletedBlock {

In replay we accidentally set the None block id for leader blocks

agave/core/src/replay_stage.rs

Lines 3525 to 3558 in fb1208f

	let block_id = if !is_leader_block {
	// If the block does not have at least DATA_SHREDS_PER_FEC_BLOCK correctly retransmitted
	// shreds in the last FEC set, mark it dead. No reason to perform this check on our leader block.
	match process_active_banks_context
	.blockstore
	.check_last_fec_set_and_get_block_id(
	bank.slot(),
	bank.hash(),
	&bank.feature_set,
	) {
	Ok(block_id) => block_id,
	Err(result_err) => {
	let root = bank_forks.read().unwrap().root();
	Self::mark_dead_slot(
	&process_active_banks_context.blockstore,
	bank,
	root,
	&result_err,
	process_active_banks_context.rpc_subscriptions.as_deref(),
	process_active_banks_context.slot_status_notifier.as_ref(),
	progress,
	duplicate_slots_to_repair,
	&process_active_banks_context.ancestor_hashes_replay_update_sender,
	purge_repair_slot_counter,
	&mut tbft_structs,
	&process_active_banks_context.replay_vote_sender,
	);
	continue;
	}
	}
	} else {
	None
	};
	bank.set_block_id(block_id);

In rare cases, shredding can complete before bank freeze on a leader block. This will lead to:

• broadcast setting Some(block_id)
• replay setting block_id back to None

Summary of Changes

Do not set block_id in replay for leader blocks.

This was found and correctly addressed in the alpenglow repo - just didn't fix it here

https://github.com/anza-xyz/alpenglow/blob/8a139cd8c68b7ee9582b55f6f9c472cb9444917b/core/src/replay_stage.rs#L3689-L3690

[codecov-commenter]

Codecov Report

❌ Patch coverage is 66.66667% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 83.1%. Comparing base (981d3d3) to head (162aaad).
⚠️ Report is 2 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master   #11613   +/-   ##
=======================================
  Coverage    83.1%    83.1%           
=======================================
  Files         850      850           
  Lines      320849   320900   +51     
=======================================
+ Hits       266759   266836   +77     
+ Misses      54090    54064   -26     

🚀 New features to boost your workflow:

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

[bw-solana]

would probably be nice to have a comment explaining the race between broadcast/replay in setting block ID

[AshwinSekar]

i'm planning to upstream the alpenglow code which splits this out in a separate fn soon, will add the comment then