Transition block lookup sync to range sync

Author: dapplion

beacon_node/beacon_chain/src/test_utils.rs

@@ -2604,7 +2604,7 @@ pub struct MakeAttestationOptions {
 pub fn build_log(level: slog::Level, enabled: bool) -> Logger {
     let decorator = TermDecorator::new().build();
     let drain = FullFormat::new(decorator).build().fuse();
-    let drain = Async::new(drain).build().fuse();
+    let drain = Async::new(drain).chan_size(10_000).build().fuse();
 
     if enabled {
         Logger::root(drain.filter_level(level).fuse(), o!())

beacon_node/network/src/network_beacon_processor/mod.rs

@@ -23,6 +23,7 @@ use std::sync::Arc;
 use std::time::Duration;
 use store::MemoryStore;
 use task_executor::TaskExecutor;
+use tokio::sync::mpsc::UnboundedSender;
 use tokio::sync::mpsc::{self, error::TrySendError};
 use types::*;
 
@@ -751,7 +752,7 @@ impl<T: BeaconChainTypes> NetworkBeaconProcessor<T> {
     /// Send a message to `sync_tx`.
     ///
     /// Creates a log if there is an internal error.
-    fn send_sync_message(&self, message: SyncMessage<T::EthSpec>) {
+    pub(crate) fn send_sync_message(&self, message: SyncMessage<T::EthSpec>) {
         self.sync_tx.send(message).unwrap_or_else(|e| {
             debug!(self.log, "Could not send message to the sync service";
                    "error" => %e)
@@ -779,6 +780,7 @@ impl<E: EthSpec> NetworkBeaconProcessor<TestBeaconChainType<E>> {
     // processor (but not much else).
     pub fn null_for_testing(
         network_globals: Arc<NetworkGlobals<E>>,
+        sync_tx: UnboundedSender<SyncMessage<E>>,
         chain: Arc<BeaconChain<TestBeaconChainType<E>>>,
         executor: TaskExecutor,
         log: Logger,
@@ -791,7 +793,6 @@ impl<E: EthSpec> NetworkBeaconProcessor<TestBeaconChainType<E>> {
         } = <_>::default();
 
         let (network_tx, _network_rx) = mpsc::unbounded_channel();
-        let (sync_tx, _sync_rx) = mpsc::unbounded_channel();
 
         let network_beacon_processor = Self {
             beacon_processor_send: beacon_processor_tx,

beacon_node/network/src/sync/block_lookups/mod.rs

@@ -28,6 +28,7 @@ use super::network_context::{PeerGroup, RpcResponseError, SyncNetworkContext};
 use crate::metrics;
 use crate::sync::block_lookups::common::ResponseType;
 use crate::sync::block_lookups::parent_chain::find_oldest_fork_ancestor;
+use crate::sync::SyncMessage;
 use beacon_chain::block_verification_types::AsBlock;
 use beacon_chain::data_availability_checker::AvailabilityCheckErrorCategory;
 use beacon_chain::{AvailabilityProcessingStatus, BeaconChainTypes, BlockError};
@@ -53,7 +54,10 @@ mod tests;
 /// The maximum depth we will search for a parent block. In principle we should have sync'd any
 /// canonical chain to its head once the peer connects. A chain should not appear where it's depth
 /// is further back than the most recent head slot.
-pub(crate) const PARENT_DEPTH_TOLERANCE: usize = SLOT_IMPORT_TOLERANCE * 2;
+///
+/// Have the same value as range's sync tolerance to consider a peer synced. Once sync lookup
+/// reaches the maximum depth it will force trigger range sync.
+pub(crate) const PARENT_DEPTH_TOLERANCE: usize = SLOT_IMPORT_TOLERANCE;
 
 const FAILED_CHAINS_CACHE_EXPIRY_SECONDS: u64 = 60;
 pub const SINGLE_BLOCK_LOOKUP_MAX_ATTEMPTS: u8 = 4;
@@ -252,22 +256,52 @@ impl<T: BeaconChainTypes> BlockLookups<T> {
                 // blocks on top of A forming A -> C. The malicious peer forces us to fetch C
                 // from it, which will result in parent A hitting the chain_too_long error. Then
                 // the valid chain A -> B is dropped too.
-                if let Ok(block_to_drop) = find_oldest_fork_ancestor(parent_chains, chain_idx) {
-                    // Drop all lookups descending from the child of the too long parent chain
-                    if let Some((lookup_id, lookup)) = self
+                //
+                // `find_oldest_fork_ancestor` should never return Err, unwrapping to tip for
+                // complete-ness
+                let parent_chain_tip = parent_chain.tip;
+                let block_to_drop =
+                    find_oldest_fork_ancestor(parent_chains, chain_idx).unwrap_or(parent_chain_tip);
+                // Drop all lookups descending from the child of the too long parent chain
+                if let Some((lookup_id, lookup)) = self
+                    .single_block_lookups
+                    .iter()
+                    .find(|(_, l)| l.block_root() == block_to_drop)
+                {
+                    // If a lookup chain is too long, we can't distinguish a valid chain from a
+                    // malicious one. We must attempt to sync this chain to not lose liveness. If
+                    // the chain grows too long, we stop lookup sync and transition this head to
+                    // forward range sync. We need to tell range sync which head to sync to, and
+                    // from which peers. The lookup of the very tip of this chain may contain zero
+                    // peers if it's the parent-child lookup. So we do a bit of a trick here:
+                    // - Tell range sync to sync to the tip's root (if available, else its ancestor)
+                    // - But use all peers in the ancestor lookup, which should have at least one
+                    //   peer, and its peer set is a strict superset of the tip's lookup.
+                    let (remote_head_root, remote_head_slot) = match self
                         .single_block_lookups
                         .iter()
-                        .find(|(_, l)| l.block_root() == block_to_drop)
+                        .find(|(_, l)| l.block_root() == parent_chain_tip)
                     {
-                        for &peer_id in lookup.all_peers() {
-                            cx.report_peer(
-                                peer_id,
-                                PeerAction::LowToleranceError,
-                                "chain_too_long",
-                            );
+                        Some((_, tip_lookup)) => {
+                            (parent_chain_tip, tip_lookup.peek_downloaded_block_slot())
                         }
-                        self.drop_lookup_and_children(*lookup_id);
-                    }
+                        None => (lookup.block_root(), lookup.peek_downloaded_block_slot()),
+                    };
+
+                    cx.send_sync_message(SyncMessage::AddPeersForceRangeSync {
+                        peers: lookup.all_peers().copied().collect(),
+                        head_slot: remote_head_slot,
+                        head_root: remote_head_root,
+                    });
+
+                    // Do not downscore peers here. Because we can't distinguish a valid chain from
+                    // a malicious one we may penalize honest peers for attempting to discover us a
+                    // valid chain. Until blocks_by_range allows to specify a tip, for example with
+                    // https://github.com/ethereum/consensus-specs/pull/3845 we will have poor
+                    // attributability. A peer can send us garbage blocks over blocks_by_root, and
+                    // then correct blocks via blocks_by_range.
+
+                    self.drop_lookup_and_children(*lookup_id);
                 }
 
                 return false;

beacon_node/network/src/sync/block_lookups/single_block_lookup.rs

@@ -16,7 +16,7 @@ use std::time::{Duration, Instant};
 use store::Hash256;
 use strum::IntoStaticStr;
 use types::blob_sidecar::FixedBlobSidecarList;
-use types::{DataColumnSidecarList, EthSpec, SignedBeaconBlock};
+use types::{DataColumnSidecarList, EthSpec, SignedBeaconBlock, Slot};
 
 // Dedicated enum for LookupResult to force its usage
 #[must_use = "LookupResult must be handled with on_lookup_result"]
@@ -92,6 +92,14 @@ impl<T: BeaconChainTypes> SingleBlockLookup<T> {
         }
     }
 
+    /// Return the slot of this lookup's block if it's currently cached as `AwaitingProcessing`
+    pub fn peek_downloaded_block_slot(&self) -> Option<Slot> {
+        self.block_request_state
+            .state
+            .peek_downloaded_data()
+            .map(|block| block.slot())
+    }
+
     /// Get the block root that is being requested.
     pub fn block_root(&self) -> Hash256 {
         self.block_root

beacon_node/network/src/sync/block_lookups/tests.rs

@@ -1,5 +1,6 @@
 use crate::network_beacon_processor::NetworkBeaconProcessor;
 use crate::sync::manager::{BlockProcessType, SyncManager};
+use crate::sync::range_sync::RangeSyncType;
 use crate::sync::sampling::SamplingConfig;
 use crate::sync::{SamplingId, SyncMessage};
 use crate::NetworkMessage;
@@ -77,6 +78,8 @@ struct TestRig {
     network_rx: mpsc::UnboundedReceiver<NetworkMessage<E>>,
     /// Stores all `NetworkMessage`s received from `network_recv`. (e.g. outgoing RPC requests)
     network_rx_queue: Vec<NetworkMessage<E>>,
+    /// Receiver for `SyncMessage` from the network
+    sync_rx: mpsc::UnboundedReceiver<SyncMessage<E>>,
     /// To send `SyncMessage`. For sending RPC responses or block processing results to sync.
     sync_manager: SyncManager<T>,
     /// To manipulate sync state and peer connection status
@@ -130,22 +133,20 @@ impl TestRig {
         let chain = harness.chain.clone();
 
         let (network_tx, network_rx) = mpsc::unbounded_channel();
-        // TODO(das): make the generation of the ENR use the deterministic rng to have consistent
-        // column assignments
+        let (sync_tx, sync_rx) = mpsc::unbounded_channel::<SyncMessage<E>>();
         let globals = Arc::new(NetworkGlobals::new_test_globals(
             Vec::new(),
             &log,
             chain.spec.clone(),
         ));
         let (beacon_processor, beacon_processor_rx) = NetworkBeaconProcessor::null_for_testing(
             globals,
+            sync_tx,
             chain.clone(),
             harness.runtime.task_executor.clone(),
             log.clone(),
         );
 
-        let (_sync_send, sync_recv) = mpsc::unbounded_channel::<SyncMessage<E>>();
-
         let fork_name = chain.spec.fork_name_at_slot::<E>(chain.slot().unwrap());
 
         // All current tests expect synced and EL online state
@@ -159,13 +160,15 @@ impl TestRig {
             beacon_processor_rx_queue: vec![],
             network_rx,
             network_rx_queue: vec![],
+            sync_rx,
             rng,
             network_globals: beacon_processor.network_globals.clone(),
             sync_manager: SyncManager::new(
                 chain,
                 network_tx,
                 beacon_processor.into(),
-                sync_recv,
+                // Pass empty recv not tied to any tx
+                mpsc::unbounded_channel().1,
                 SamplingConfig::Custom {
                     required_successes: vec![SAMPLING_REQUIRED_SUCCESSES],
                 },
@@ -232,6 +235,13 @@ impl TestRig {
         self.send_sync_message(SyncMessage::SampleBlock(block_root, block_slot))
     }
 
+    /// Drain all sync messages in the sync_rx attached to the beacon processor
+    fn drain_sync_rx(&mut self) {
+        while let Ok(sync_message) = self.sync_rx.try_recv() {
+            self.send_sync_message(sync_message);
+        }
+    }
+
     fn rand_block(&mut self) -> SignedBeaconBlock<E> {
         self.rand_block_and_blobs(NumBlobs::None).0
     }
@@ -288,6 +298,10 @@ impl TestRig {
         self.sync_manager.active_parent_lookups().len()
     }
 
+    fn active_range_sync_chain(&self) -> (RangeSyncType, Slot, Slot) {
+        self.sync_manager.get_range_sync_chains().unwrap().unwrap()
+    }
+
     fn assert_single_lookups_count(&self, count: usize) {
         assert_eq!(
             self.active_single_lookups_count(),
@@ -1665,7 +1679,18 @@ fn test_parent_lookup_too_deep_grow_ancestor() {
         )
     }
 
-    rig.expect_penalty(peer_id, "chain_too_long");
+    // Should create a new syncing chain
+    rig.drain_sync_rx();
+    assert_eq!(
+        rig.active_range_sync_chain(),
+        (
+            RangeSyncType::Head,
+            Slot::new(0),
+            Slot::new(PARENT_DEPTH_TOLERANCE as u64 - 1)
+        )
+    );
+    // Should not penalize peer, but network is not clear because of the blocks_by_range requests
+    rig.expect_no_penalty_for(peer_id);
     rig.assert_failed_chain(chain_hash);
 }
 
@@ -1689,7 +1714,18 @@ fn test_parent_lookup_too_deep_grow_tip() {
         );
     }
 
-    rig.expect_penalty(peer_id, "chain_too_long");
+    // Should create a new syncing chain
+    rig.drain_sync_rx();
+    assert_eq!(
+        rig.active_range_sync_chain(),
+        (
+            RangeSyncType::Head,
+            Slot::new(0),
+            Slot::new(PARENT_DEPTH_TOLERANCE as u64 - 2)
+        )
+    );
+    // Should not penalize peer, but network is not clear because of the blocks_by_range requests
+    rig.expect_no_penalty_for(peer_id);
     rig.assert_failed_chain(tip.canonical_root());
 }