* Connection pool in connection cache and handle connection errors
1. The connection not has a pool of connections per address, configurable, default 4
2. The connections per address share a lazy initialized endpoint
3. Handle connection issues better, avoid race conditions
4. Various log improvement for help debug connection issues
https://github.com/solana-labs/solana/pull/25744
separated durable nonce and blockhash domains, which will stop double
execution going forward. However it is possible that a durable
transaction has *already* been executed once as a normal transaction and
it is now a valid durable transaction. #25744 cannot stop such
transactions to be re-executed until the nonce accounts are advanced.
This commit adds a new nonce version indicating that the nonce is moved
out of the blockhash domain, and permanently disables durable
transactions for legacy nonces which are in the blockhash domain.
* client: Remove static connection cache, plumb it instead
* Add TpuClient::new_with_connection_cache to not break downstream
* Refactor get_connection and RwLock into ConnectionCache
* Fix merge conflicts from new async TpuClient
* Remove `ConnectionCache::set_use_quic`
* Move DEFAULT_TPU_USE_QUIC to client, use ConnectionCache::default()
AdvanceNonceAccount instruction updates nonce to blockhash. This makes it
possible that a durable transaction is executed twice both as a normal
transaction and a nonce transaction if it uses blockhash (as opposed to nonce)
for its recent_blockhash field.
The commit prevents this double execution by separating nonce and blockhash
domains; when advancing nonce account, blockhash is hashed with a fixed string.
As a result a blockhash cannot be a valid nonce value; and if transaction was
once executed as a normal transaction it cannot be re-executed as a durable
transaction again and vice-versa.
* Spawn QUIC server to receive forwarded txs
* Update validator port range
* forward votes using UDP
* no forwarding from unstaked nodes
* forwarding stats in banking stage
* fix test builds
* fix lifetime of forward sender
1. Move logics related to creating endpoint, creating new connection and retry 0rtt connections into a wrapper construct QuicNewConnection to put the logic together.
2. get_or_add_connection: logic is much simplified to allow manage the connection cache -- the QUIC connection is lazy constructed. get_or_add_connection should no longer be a global hotspot.
3. Per connection stats update is moved out of get_or_add_connection to avoid bad connection impacting good connections.
* Use RwLock instead of Mutex in QUIC connection cache
* replace LruCache with HashMap
* fix tests
* fix tests
* refactor
* add cache eviction for a random connection on reaching upperbound
* cleanup
* Increase connection timeouts
* Bump quic connection cache to 1024
* Use constant for quic connection timeout and add warm cache service
* Fixes to QUIC warmup service
* fix check failure
* fixes after rebase
* fix timeout test
Co-authored-by: Pankaj Garg <pankaj@solana.com>
Add an interface send_wire_transaction_batch_async to TpuConnection to allow for sending batches without waiting for completion
Co-authored-by: Anatoly Yakovenko <anatoly@solana.com>
* Add fallible send methods, and rpc_client helper
* Add helper to return RpcClient url
* Implement BenchTpsClient for TpuClient
* Add cli rpc and identity handling
* Handle different kinds of clients in main, use TpuClient
* Add tpu_client integration test
* Add BenchTpsClient
* Impl BenchTpsClient for used clients
* Use BenchTpsClient in do_bench
* Update integration test to use faucet via rpc
* Support keypairs from file that are not prefunded
* Remove old perf-utils
Refactor the thin_client::create_client to take addresses separately instead of as a tuple
Co-authored-by: Bijie Zhu <bijiezhu@Bijies-MBP.cable.rcn.com>
steveluscher
Michael Vines
Pankaj Garg
Lijun Wang
behzad nouri
Jon Cinque
segfaultdoctor
ryleung-solana
Brooks Prumo
Justin Starry
Jack May
Brennan Watt
Brian Anderson
Tyera Eulberg
sakridge
anatoly yakovenko
BG Zhu
hana