Fix custody context initialization race condition that caused panic (#8391)

Take 2 of #8390.

Fixes the race condition properly instead of propagating the error. I think this is a better alternative, and doesn't seem to look that bad.


  * Lift node id loading or generation from `NetworkService ` startup to the `ClientBuilder`, so that it can be used to compute custody columns for the beacon chain without waiting for Network bootstrap.

I've considered and implemented a few alternatives:
1. passing `node_id` to beacon chain builder and compute columns when creating `CustodyContext`. This approach isn't good for separation of concerns and isn't great for testability
2. passing `ordered_custody_groups` to beacon chain. `CustodyContext` only uses this to compute ordered custody columns, so we might as well lift this logic out, so we don't have to do error handling in `CustodyContext` construction. Less tests to update;.


Co-Authored-By: Jimmy Chen <jchen.tc@gmail.com>
This commit is contained in:
Jimmy Chen
2025-11-17 16:23:12 +11:00
committed by GitHub
parent f2b945a5b5
commit af1d9b9991
15 changed files with 230 additions and 196 deletions

View File

@@ -12,6 +12,7 @@ use futures::future::OptionFuture;
use futures::prelude::*;
use lighthouse_network::Enr;
use lighthouse_network::identity::Keypair;
use lighthouse_network::rpc::InboundRequestId;
use lighthouse_network::rpc::RequestType;
use lighthouse_network::rpc::methods::RpcResponse;
@@ -212,6 +213,7 @@ impl<T: BeaconChainTypes> NetworkService<T> {
executor: task_executor::TaskExecutor,
libp2p_registry: Option<&'_ mut Registry>,
beacon_processor_send: BeaconProcessorSend<T::EthSpec>,
local_keypair: Keypair,
) -> Result<
(
NetworkService<T>,
@@ -284,6 +286,7 @@ impl<T: BeaconChainTypes> NetworkService<T> {
.data_availability_checker
.custody_context()
.custody_group_count_at_head(&beacon_chain.spec),
local_keypair,
)
.await?;
@@ -366,6 +369,7 @@ impl<T: BeaconChainTypes> NetworkService<T> {
executor: task_executor::TaskExecutor,
libp2p_registry: Option<&'_ mut Registry>,
beacon_processor_send: BeaconProcessorSend<T::EthSpec>,
local_keypair: Keypair,
) -> Result<(Arc<NetworkGlobals<T::EthSpec>>, NetworkSenders<T::EthSpec>), String> {
let (network_service, network_globals, network_senders) = Self::build(
beacon_chain,
@@ -373,6 +377,7 @@ impl<T: BeaconChainTypes> NetworkService<T> {
executor.clone(),
libp2p_registry,
beacon_processor_send,
local_keypair,
)
.await?;

View File

@@ -6,6 +6,7 @@ use beacon_chain::BeaconChainTypes;
use beacon_chain::test_utils::BeaconChainHarness;
use beacon_processor::{BeaconProcessorChannels, BeaconProcessorConfig};
use futures::StreamExt;
use lighthouse_network::identity::secp256k1;
use lighthouse_network::types::{GossipEncoding, GossipKind};
use lighthouse_network::{Enr, GossipTopic};
use std::str::FromStr;
@@ -66,6 +67,7 @@ fn test_dht_persistence() {
executor,
None,
beacon_processor_tx,
secp256k1::Keypair::generate().into(),
)
.await
.unwrap();
@@ -134,6 +136,7 @@ fn test_removing_topic_weight_on_old_topics() {
executor.clone(),
None,
beacon_processor_channels.beacon_processor_tx,
secp256k1::Keypair::generate().into(),
)
.await
.unwrap()

View File

@@ -1,4 +1,5 @@
use super::*;
use beacon_chain::test_utils::generate_data_column_indices_rand_order;
use beacon_chain::{
BeaconChain,
builder::{BeaconChainBuilder, Witness},
@@ -73,6 +74,9 @@ impl TestBeaconChain {
Duration::from_secs(recent_genesis_time()),
Duration::from_millis(SLOT_DURATION_MILLIS),
))
.ordered_custody_column_indices(generate_data_column_indices_rand_order::<
MainnetEthSpec,
>())
.shutdown_sender(shutdown_tx)
.rng(Box::new(StdRng::seed_from_u64(42)))
.build()