feat(congestion_control) - relax congestion control (#12241)

As per [zulip
discussion](https://near.zulipchat.com/#narrow/channel/295558-core/topic/relax.20congestion.20control.20parameters.3F/near/477318893)
congestion control seems to be too aggressive.

This is relaxing two parameters: 
* The max gas in the delayed receipts queue increased twofold-> this
will allow the queue to grow larger and buffer more receipts to handle
spikes in traffic better. The tradeoff is that the queue will be larger
and the transaction processing time will be slower during those periods.
* The `reject_tx_congestion_threshold` bumped from 50% to 70%. It means
that new transactions will only be rejected once the congestion level
reaches 70%.

Both parameters are picked pretty arbitrarily and there is no science
behind it.

I am not increasing the outgoing or memory limits as they don't seem to
be the bottleneck.

CHANGELOG.md

@@ -2,12 +2,23 @@
 
 ## [unreleased]
 
+### Protocol Changes
+
+* Fixing invalid cost used for wasm_yield_resume_byte. #12192
+* Relaxing Congestion Control to allow accepting and buffering more transactions. #12241
+
+### Non-protocol Changes
+**No Changes**
+
+## [2.3.0]
+
 ### Protocol Changes
 * Sets `chunk_validator_only_kickout_threshold` to 70. Uses this kickout threshold as a cutoff threshold for contribution of endorsement ratio in rewards calculation: if endorsement ratio is above 70%, the contribution of endorsement ratio in average uptime calculation is 100%, otherwise it is 0%. Endorsements received are now included in `BlockHeader` to improve kickout and reward calculation for chunk validators. 
 
 ### Non-protocol Changes
 * Added [documentation](./docs/misc/archival_data_recovery.md) and a [reference](./scripts/recover_missing_archival_data.sh) script to recover the data lost in archival nodes at the beginning of 2024.
 * **Archival nodes only:** Stop saving partial chunks to `PartialChunks` column in the Cold DB. Instead, archival nodes will reconstruct partial chunks from the `Chunks` column.
+* Decentralized state sync: Before, nodes that needed to download state (either because they're several epochs behind the chain or because they're going to start producing chunks for a shard they don't currently track) would download them from a centralized GCS bucket. Now, nodes will attempt to download pieces of the state from peers in the network, and only fallback to downloading from GCS if that fails. Please note that in order to participate in providing state parts to peers, your node may generate snapshots of the state. These snapshots should not take too much space, since they're hard links to database files that get cleaned up on every epoch.
 
 ### 2.2.0
 

chain/chain/src/tests/simple_chain.rs

@@ -34,7 +34,7 @@ fn build_chain() {
     if cfg!(feature = "nightly") {
         insta::assert_snapshot!(hash, @"Hc3bWEd7ikHf9BAe2SknvH2jAAakEtBRU1FBu6Udocm3");
     } else {
-        insta::assert_snapshot!(hash, @"dY6Z6HdATLWK3wwxkNtUs8T1GaEQqpxUCXm7TectWW7");
+        insta::assert_snapshot!(hash, @"GHZFAFiMdGzAfnWTcS9u9wqFvxMrgFpyEr6Use7jk2Lo");
     }
 
     for i in 1..5 {
@@ -52,7 +52,7 @@ fn build_chain() {
     if cfg!(feature = "nightly") {
         insta::assert_snapshot!(hash, @"39R6bDFXkPfwdYs4crV3RyCde85ecycqP5DBwdtwyjcJ");
     } else {
-        insta::assert_snapshot!(hash, @"6RnKeuiGmxkFxNYeEmAbK6NzwvKYuTcKCwqAmqJ6m3DG");
+        insta::assert_snapshot!(hash, @"3Pdm44L71Bk8EokPHF1pxakHojsriNadBdZZSpcoDv9q");
     }
 }
 

chain/jsonrpc/jsonrpc-tests/res/genesis_config.json

@@ -1,5 +1,5 @@
 {
-  "protocol_version": 72,
+  "protocol_version": 73,
   "genesis_time": "1970-01-01T00:00:00.000000000Z",
   "chain_id": "sample",
   "genesis_height": 0,

core/parameters/res/runtime_configs/73.yaml

@@ -1 +1,15 @@
 wasm_yield_resume_byte: { old: 1_195_627_285_210 , new: 47_683_715 }
+
+# Congestion Control
+
+# 40 PGAS
+max_congestion_incoming_gas: {
+  old : 20_000_000_000_000_000,
+  new : 40_000_000_000_000_000,
+}
+
+# 0.7
+reject_tx_congestion_threshold: {
+  old : { numerator: 50, denominator: 100 },
+  new : { numerator: 80, denominator: 100 }
+}

core/parameters/res/runtime_configs/parameters.snap

@@ -147,7 +147,7 @@ wasm_alt_bn128_g1_sum_element                  5_000_000_000
 wasm_yield_create_base                       153_411_779_276
 wasm_yield_create_byte                            15_643_988
 wasm_yield_resume_base                     1_195_627_285_210
-wasm_yield_resume_byte                     1_195_627_285_210
+wasm_yield_resume_byte                            47_683_715
 wasm_bls12381_p1_sum_base                     16_500_000_000
 wasm_bls12381_p1_sum_element                   6_000_000_000
 wasm_bls12381_p2_sum_base                     18_600_000_000
@@ -208,7 +208,7 @@ vm_kind                                 NearVm
 eth_implicit_accounts                   true
 yield_resume                            true
 discard_custom_sections                 true
-max_congestion_incoming_gas             20_000_000_000_000_000
+max_congestion_incoming_gas             40_000_000_000_000_000
 max_congestion_outgoing_gas             10_000_000_000_000_000
 max_congestion_memory_consumption              1_000_000_000
 max_congestion_missed_chunks                               5
@@ -217,5 +217,5 @@ min_outgoing_gas                        1_000_000_000_000_000
 allowed_shard_outgoing_gas              1_000_000_000_000_000
 max_tx_gas                               500_000_000_000_000
 min_tx_gas                                20_000_000_000_000
-reject_tx_congestion_threshold          50 / 100
+reject_tx_congestion_threshold          80 / 100
 use_state_stored_receipt                true

core/parameters/src/config.rs

@@ -4,7 +4,7 @@ use crate::config_store::INITIAL_TESTNET_CONFIG;
 use crate::cost::RuntimeFeesConfig;
 use crate::parameter_table::ParameterTable;
 use near_account_id::AccountId;
-use near_primitives_core::types::{Balance, Gas};
+use near_primitives_core::types::{Balance, Gas, ProtocolVersion};
 use near_primitives_core::version::PROTOCOL_VERSION;
 use std::sync::Arc;
 
@@ -49,8 +49,12 @@ impl RuntimeConfig {
     }
 
     pub fn test() -> Self {
+        Self::test_protocol_version(PROTOCOL_VERSION)
+    }
+
+    pub fn test_protocol_version(protocol_version: ProtocolVersion) -> Self {
         let config_store = super::config_store::RuntimeConfigStore::new(None);
-        let runtime_config = config_store.get_config(PROTOCOL_VERSION);
+        let runtime_config = config_store.get_config(protocol_version);
 
         let mut wasm_config = crate::vm::Config::clone(&runtime_config.wasm_config);
         // Lower the yield timeout length so that we can observe timeouts in integration tests.

core/parameters/src/config_store.rs

@@ -49,6 +49,7 @@ static CONFIG_DIFFS: &[(ProtocolVersion, &str)] = &[
     (70, include_config!("70.yaml")),
     // Increase main_storage_proof_size_soft_limit and introduces StateStoredReceipt
     (72, include_config!("72.yaml")),
+    // Fix wasm_yield_resume_byte and relax congestion control.
     (73, include_config!("73.yaml")),
     (129, include_config!("129.yaml")),
 ];

core/parameters/src/snapshots/near_parameters__config_store__tests__129.json.snap

@@ -246,7 +246,7 @@ expression: config_view
     "registrar_account_id": "registrar"
   },
   "congestion_control_config": {
-    "max_congestion_incoming_gas": 20000000000000000,
+    "max_congestion_incoming_gas": 40000000000000000,
     "max_congestion_outgoing_gas": 10000000000000000,
     "max_congestion_memory_consumption": 1000000000,
     "max_congestion_missed_chunks": 5,
@@ -255,7 +255,7 @@ expression: config_view
     "allowed_shard_outgoing_gas": 1000000000000000,
     "max_tx_gas": 500000000000000,
     "min_tx_gas": 20000000000000,
-    "reject_tx_congestion_threshold": 0.5,
+    "reject_tx_congestion_threshold": 0.8,
     "outgoing_receipts_usual_size_limit": 102400,
     "outgoing_receipts_big_size_limit": 4718592
   },

core/parameters/src/snapshots/near_parameters__config_store__tests__73.json.snap

@@ -246,7 +246,7 @@ expression: config_view
     "registrar_account_id": "registrar"
   },
   "congestion_control_config": {
-    "max_congestion_incoming_gas": 20000000000000000,
+    "max_congestion_incoming_gas": 40000000000000000,
     "max_congestion_outgoing_gas": 10000000000000000,
     "max_congestion_memory_consumption": 1000000000,
     "max_congestion_missed_chunks": 5,
@@ -255,7 +255,7 @@ expression: config_view
     "allowed_shard_outgoing_gas": 1000000000000000,
     "max_tx_gas": 500000000000000,
     "min_tx_gas": 20000000000000,
-    "reject_tx_congestion_threshold": 0.5,
+    "reject_tx_congestion_threshold": 0.8,
     "outgoing_receipts_usual_size_limit": 102400,
     "outgoing_receipts_big_size_limit": 4718592
   },

core/parameters/src/snapshots/near_parameters__config_store__tests__testnet_129.json.snap

@@ -246,7 +246,7 @@ expression: config_view
     "registrar_account_id": "registrar"
   },
   "congestion_control_config": {
-    "max_congestion_incoming_gas": 20000000000000000,
+    "max_congestion_incoming_gas": 40000000000000000,
     "max_congestion_outgoing_gas": 10000000000000000,
     "max_congestion_memory_consumption": 1000000000,
     "max_congestion_missed_chunks": 5,
@@ -255,7 +255,7 @@ expression: config_view
     "allowed_shard_outgoing_gas": 1000000000000000,
     "max_tx_gas": 500000000000000,
     "min_tx_gas": 20000000000000,
-    "reject_tx_congestion_threshold": 0.5,
+    "reject_tx_congestion_threshold": 0.8,
     "outgoing_receipts_usual_size_limit": 102400,
     "outgoing_receipts_big_size_limit": 4718592
   },

core/parameters/src/snapshots/near_parameters__config_store__tests__testnet_73.json.snap

@@ -246,7 +246,7 @@ expression: config_view
     "registrar_account_id": "registrar"
   },
   "congestion_control_config": {
-    "max_congestion_incoming_gas": 20000000000000000,
+    "max_congestion_incoming_gas": 40000000000000000,
     "max_congestion_outgoing_gas": 10000000000000000,
     "max_congestion_memory_consumption": 1000000000,
     "max_congestion_missed_chunks": 5,
@@ -255,7 +255,7 @@ expression: config_view
     "allowed_shard_outgoing_gas": 1000000000000000,
     "max_tx_gas": 500000000000000,
     "min_tx_gas": 20000000000000,
-    "reject_tx_congestion_threshold": 0.5,
+    "reject_tx_congestion_threshold": 0.8,
     "outgoing_receipts_usual_size_limit": 102400,
     "outgoing_receipts_big_size_limit": 4718592
   },

core/parameters/src/snapshots/near_parameters__view__tests__runtime_config_view.snap

@@ -174,7 +174,7 @@ expression: "&view"
       "yield_create_base": 153411779276,
       "yield_create_byte": 15643988,
       "yield_resume_base": 1195627285210,
-      "yield_resume_byte": 1195627285210,
+      "yield_resume_byte": 47683715,
       "bls12381_p1_sum_base": 16500000000,
       "bls12381_p1_sum_element": 6000000000,
       "bls12381_p2_sum_base": 18600000000,
@@ -246,7 +246,7 @@ expression: "&view"
     "registrar_account_id": "registrar"
   },
   "congestion_control_config": {
-    "max_congestion_incoming_gas": 20000000000000000,
+    "max_congestion_incoming_gas": 40000000000000000,
     "max_congestion_outgoing_gas": 10000000000000000,
     "max_congestion_memory_consumption": 1000000000,
     "max_congestion_missed_chunks": 5,
@@ -255,7 +255,7 @@ expression: "&view"
     "allowed_shard_outgoing_gas": 1000000000000000,
     "max_tx_gas": 500000000000000,
     "min_tx_gas": 20000000000000,
-    "reject_tx_congestion_threshold": 0.5,
+    "reject_tx_congestion_threshold": 0.8,
     "outgoing_receipts_usual_size_limit": 102400,
     "outgoing_receipts_big_size_limit": 4718592
   },

core/primitives-core/src/version.rs

@@ -267,7 +267,7 @@ impl ProtocolFeature {
 }
 
 /// Current protocol version used on the mainnet with all stable features.
-const STABLE_PROTOCOL_VERSION: ProtocolVersion = 72;
+const STABLE_PROTOCOL_VERSION: ProtocolVersion = 73;
 
 // On nightly, pick big enough version to support all features.
 const NIGHTLY_PROTOCOL_VERSION: ProtocolVersion = 145;

core/primitives/src/congestion_info.rs

@@ -507,8 +507,10 @@ mod tests {
     use super::*;
 
     fn get_config() -> CongestionControlConfig {
+        // Fix the initial configuration of congestion control for the tests.
+        let protocol_version = ProtocolFeature::CongestionControl.protocol_version();
         let runtime_config_store = RuntimeConfigStore::new(None);
-        let runtime_config = runtime_config_store.get_config(PROTOCOL_VERSION);
+        let runtime_config = runtime_config_store.get_config(protocol_version);
         runtime_config.congestion_control_config
     }
 

core/primitives/src/snapshots/near_primitives__views__tests__runtime_config_view.snap

@@ -174,7 +174,7 @@ expression: "&view"
       "yield_create_base": 153411779276,
       "yield_create_byte": 15643988,
       "yield_resume_base": 1195627285210,
-      "yield_resume_byte": 1195627285210,
+      "yield_resume_byte": 47683715,
       "bls12381_p1_sum_base": 16500000000,
       "bls12381_p1_sum_element": 6000000000,
       "bls12381_p2_sum_base": 18600000000,
@@ -246,7 +246,7 @@ expression: "&view"
     "registrar_account_id": "registrar"
   },
   "congestion_control_config": {
-    "max_congestion_incoming_gas": 20000000000000000,
+    "max_congestion_incoming_gas": 40000000000000000,
     "max_congestion_outgoing_gas": 10000000000000000,
     "max_congestion_memory_consumption": 1000000000,
     "max_congestion_missed_chunks": 5,
@@ -255,7 +255,7 @@ expression: "&view"
     "allowed_shard_outgoing_gas": 1000000000000000,
     "max_tx_gas": 500000000000000,
     "min_tx_gas": 20000000000000,
-    "reject_tx_congestion_threshold": 0.5,
+    "reject_tx_congestion_threshold": 0.8,
     "outgoing_receipts_usual_size_limit": 102400,
     "outgoing_receipts_big_size_limit": 4718592
   },

integration-tests/src/tests/client/features/congestion_control.rs

@@ -53,7 +53,7 @@ fn setup_test_runtime(sender_id: AccountId, protocol_version: ProtocolVersion) -
     // Chain must be sharded to test cross-shard congestion control.
     genesis.config.shard_layout = ShardLayout::v1_test();
 
-    let mut config = RuntimeConfig::test();
+    let mut config = RuntimeConfig::test_protocol_version(protocol_version);
     adjust_runtime_config(&mut config);
     let runtime_configs = vec![RuntimeConfigStore::with_one_config(config)];
 
@@ -515,11 +515,16 @@ fn submit_n_cheap_fns(
 /// with remote traffic.
 #[test]
 fn test_transaction_limit_for_local_congestion() {
+    init_test_logger();
+
     if !ProtocolFeature::CongestionControl.enabled(PROTOCOL_VERSION) {
         return;
     }
     let runtime_config_store = RuntimeConfigStore::new(None);
-    let config = runtime_config_store.get_config(PROTOCOL_VERSION);
+
+    // Fix the initial configuration of congestion control for the tests.
+    let protocol_version = ProtocolFeature::CongestionControl.protocol_version();
+    let config = runtime_config_store.get_config(protocol_version);
     // We don't want to go into the TX rejection limit in this test.
     let upper_limit_congestion = config.congestion_control_config.reject_tx_congestion_threshold;
 
@@ -528,7 +533,7 @@ fn test_transaction_limit_for_local_congestion() {
     let contract_id: AccountId = CONTRACT_ID.parse().unwrap();
     let sender_id = contract_id.clone();
     let dummy_receiver: AccountId = "a_dummy_receiver".parse().unwrap();
-    let env = setup_test_runtime("test0".parse().unwrap(), PROTOCOL_VERSION);
+    let env = setup_test_runtime("test0".parse().unwrap(), protocol_version);
 
     let (
         remote_tx_included_without_congestion,

pytest/tests/sanity/congestion_control.py

@@ -8,7 +8,6 @@
 import unittest
 import pathlib
 import sys
-import json
 import time
 import threading
 
@@ -141,7 +140,7 @@ def __run_under_congestion(self, node):
 
     def __run_after_congestion(self, node):
         logger.info("Checking the chain after congestion")
-        for height, hash in poll_blocks(node, __target=50):
+        for height, hash in poll_blocks(node, __target=100):
             chunk = self.__get_chunk(node, hash, 0)
 
             gas_used = chunk['header']['gas_used']
@@ -200,8 +199,10 @@ def __start_load(self, node: BaseNode, accounts):
         self.finished = False
         self.lock = threading.Lock()
 
+        self.txs = []
         target_account = accounts[0]
-        for account in accounts:
+        # Spawn two thread per each account to get more transactions in.
+        for account in accounts + accounts:
             thread = threading.Thread(
                 target=self.__load,
                 args=[node, account, target_account],
@@ -221,7 +222,6 @@ def __load(self, node: BaseNode, sender_account, target_account):
         logger.debug(
             f"Starting load thread {sender_account.account_id} -> {target_account.account_id}"
         )
-        self.txs = []
         while not self.finished:
             tx_hash = self.__call_contract(node, sender_account, target_account)
             with self.lock: