579 lines
20 KiB
Rust
579 lines
20 KiB
Rust
//! Set of functions for emulating windowing functions from a database ledger implementation
|
|
use cluster_info::ClusterInfo;
|
|
use counter::Counter;
|
|
use db_ledger::*;
|
|
use entry::Entry;
|
|
use leader_scheduler::LeaderScheduler;
|
|
use log::Level;
|
|
use packet::{SharedBlob, BLOB_HEADER_SIZE};
|
|
use result::Result;
|
|
use rocksdb::DBRawIterator;
|
|
use solana_metrics::{influxdb, submit};
|
|
use solana_sdk::pubkey::Pubkey;
|
|
use std::cmp;
|
|
use std::net::SocketAddr;
|
|
use std::sync::atomic::AtomicUsize;
|
|
use std::sync::atomic::{AtomicBool, Ordering};
|
|
use std::sync::{Arc, RwLock};
|
|
use streamer::BlobSender;
|
|
|
|
pub fn repair(
|
|
slot: u64,
|
|
db_ledger: &DbLedger,
|
|
cluster_info: &Arc<RwLock<ClusterInfo>>,
|
|
id: &Pubkey,
|
|
times: usize,
|
|
tick_height: u64,
|
|
max_entry_height: u64,
|
|
leader_scheduler_option: &Arc<RwLock<LeaderScheduler>>,
|
|
) -> Result<Vec<(SocketAddr, Vec<u8>)>> {
|
|
let rcluster_info = cluster_info.read().unwrap();
|
|
let mut is_next_leader = false;
|
|
let meta = db_ledger.meta_cf.get(&db_ledger.db, &MetaCf::key(slot))?;
|
|
if meta.is_none() {
|
|
return Ok(vec![]);
|
|
}
|
|
let meta = meta.unwrap();
|
|
|
|
let consumed = meta.consumed;
|
|
let received = meta.received;
|
|
|
|
// Repair should only be called when received > consumed, enforced in window_service
|
|
assert!(received > consumed);
|
|
{
|
|
let ls_lock = leader_scheduler_option.read().unwrap();
|
|
if !ls_lock.use_only_bootstrap_leader {
|
|
// Calculate the next leader rotation height and check if we are the leader
|
|
if let Some(next_leader_rotation_height) = ls_lock.max_height_for_leader(tick_height) {
|
|
match ls_lock.get_scheduled_leader(next_leader_rotation_height) {
|
|
Some((leader_id, _)) if leader_id == *id => is_next_leader = true,
|
|
// In the case that we are not in the current scope of the leader schedule
|
|
// window then either:
|
|
//
|
|
// 1) The replicate stage hasn't caught up to the "consumed" entries we sent,
|
|
// in which case it will eventually catch up
|
|
//
|
|
// 2) We are on the border between seed_rotation_intervals, so the
|
|
// schedule won't be known until the entry on that cusp is received
|
|
// by the replicate stage (which comes after this stage). Hence, the next
|
|
// leader at the beginning of that next epoch will not know they are the
|
|
// leader until they receive that last "cusp" entry. The leader also won't ask for repairs
|
|
// for that entry because "is_next_leader" won't be set here. In this case,
|
|
// everybody will be blocking waiting for that "cusp" entry instead of repairing,
|
|
// until the leader hits "times" >= the max times in calculate_max_repair_entry_height().
|
|
// The impact of this, along with the similar problem from broadcast for the transitioning
|
|
// leader, can be observed in the multinode test, test_full_leader_validator_network(),
|
|
None => (),
|
|
_ => (),
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
let num_peers = rcluster_info.tvu_peers().len() as u64;
|
|
|
|
// Check if there's a max_entry_height limitation
|
|
let max_repair_entry_height = if max_entry_height == 0 {
|
|
calculate_max_repair_entry_height(num_peers, consumed, received, times, is_next_leader)
|
|
} else {
|
|
max_entry_height + 2
|
|
};
|
|
|
|
let idxs = find_missing_data_indexes(slot, db_ledger, consumed, max_repair_entry_height - 1);
|
|
|
|
let reqs: Vec<_> = idxs
|
|
.into_iter()
|
|
.filter_map(|pix| rcluster_info.window_index_request(pix).ok())
|
|
.collect();
|
|
|
|
drop(rcluster_info);
|
|
|
|
inc_new_counter_info!("streamer-repair_window-repair", reqs.len());
|
|
|
|
if log_enabled!(Level::Trace) {
|
|
trace!(
|
|
"{}: repair_window counter times: {} consumed: {} received: {} max_repair_entry_height: {} missing: {}",
|
|
id,
|
|
times,
|
|
consumed,
|
|
received,
|
|
max_repair_entry_height,
|
|
reqs.len()
|
|
);
|
|
for (to, _) in &reqs {
|
|
trace!("{}: repair_window request to {}", id, to);
|
|
}
|
|
}
|
|
|
|
Ok(reqs)
|
|
}
|
|
|
|
// Given a start and end entry index, find all the missing
|
|
// indexes in the ledger in the range [start_index, end_index)
|
|
pub fn find_missing_indexes(
|
|
db_iterator: &mut DBRawIterator,
|
|
slot: u64,
|
|
start_index: u64,
|
|
end_index: u64,
|
|
key: &Fn(u64, u64) -> Vec<u8>,
|
|
index_from_key: &Fn(&[u8]) -> Result<u64>,
|
|
) -> Vec<u64> {
|
|
if start_index >= end_index {
|
|
return vec![];
|
|
}
|
|
|
|
let mut missing_indexes = vec![];
|
|
|
|
// Seek to the first blob with index >= start_index
|
|
db_iterator.seek(&key(slot, start_index));
|
|
|
|
// The index of the first missing blob in the slot
|
|
let mut prev_index = start_index;
|
|
loop {
|
|
if !db_iterator.valid() {
|
|
break;
|
|
}
|
|
let current_key = db_iterator.key().expect("Expect a valid key");
|
|
let current_index =
|
|
index_from_key(¤t_key).expect("Expect to be able to parse index from valid key");
|
|
let upper_index = cmp::min(current_index, end_index);
|
|
for i in prev_index..upper_index {
|
|
missing_indexes.push(i);
|
|
}
|
|
if current_index >= end_index {
|
|
break;
|
|
}
|
|
|
|
prev_index = current_index + 1;
|
|
db_iterator.next();
|
|
}
|
|
|
|
missing_indexes
|
|
}
|
|
|
|
pub fn find_missing_data_indexes(
|
|
slot: u64,
|
|
db_ledger: &DbLedger,
|
|
start_index: u64,
|
|
end_index: u64,
|
|
) -> Vec<u64> {
|
|
let mut db_iterator = db_ledger
|
|
.db
|
|
.raw_iterator_cf(db_ledger.data_cf.handle(&db_ledger.db))
|
|
.expect("Expected to be able to open database iterator");
|
|
|
|
find_missing_indexes(
|
|
&mut db_iterator,
|
|
slot,
|
|
start_index,
|
|
end_index,
|
|
&DataCf::key,
|
|
&DataCf::index_from_key,
|
|
)
|
|
}
|
|
|
|
pub fn find_missing_coding_indexes(
|
|
slot: u64,
|
|
db_ledger: &DbLedger,
|
|
start_index: u64,
|
|
end_index: u64,
|
|
) -> Vec<u64> {
|
|
let mut db_iterator = db_ledger
|
|
.db
|
|
.raw_iterator_cf(db_ledger.erasure_cf.handle(&db_ledger.db))
|
|
.expect("Expected to be able to open database iterator");
|
|
|
|
find_missing_indexes(
|
|
&mut db_iterator,
|
|
slot,
|
|
start_index,
|
|
end_index,
|
|
&ErasureCf::key,
|
|
&ErasureCf::index_from_key,
|
|
)
|
|
}
|
|
|
|
pub fn retransmit_all_leader_blocks(
|
|
dq: &[SharedBlob],
|
|
leader_scheduler: &LeaderScheduler,
|
|
retransmit: &BlobSender,
|
|
) -> Result<()> {
|
|
let mut retransmit_queue: Vec<SharedBlob> = Vec::new();
|
|
for b in dq {
|
|
// Check if the blob is from the scheduled leader for its slot. If so,
|
|
// add to the retransmit_queue
|
|
let slot = b.read().unwrap().slot()?;
|
|
if let Some(leader_id) = leader_scheduler.get_leader_for_slot(slot) {
|
|
add_blob_to_retransmit_queue(b, leader_id, &mut retransmit_queue);
|
|
}
|
|
}
|
|
|
|
submit(
|
|
influxdb::Point::new("retransmit-queue")
|
|
.add_field(
|
|
"count",
|
|
influxdb::Value::Integer(retransmit_queue.len() as i64),
|
|
).to_owned(),
|
|
);
|
|
|
|
if !retransmit_queue.is_empty() {
|
|
inc_new_counter_info!("streamer-recv_window-retransmit", retransmit_queue.len());
|
|
retransmit.send(retransmit_queue)?;
|
|
}
|
|
Ok(())
|
|
}
|
|
|
|
pub fn add_blob_to_retransmit_queue(
|
|
b: &SharedBlob,
|
|
leader_id: Pubkey,
|
|
retransmit_queue: &mut Vec<SharedBlob>,
|
|
) {
|
|
let p = b.read().unwrap();
|
|
if p.id().expect("get_id in fn add_block_to_retransmit_queue") == leader_id {
|
|
let nv = SharedBlob::default();
|
|
{
|
|
let mut mnv = nv.write().unwrap();
|
|
let sz = p.meta.size;
|
|
mnv.meta.size = sz;
|
|
mnv.data[..sz].copy_from_slice(&p.data[..sz]);
|
|
}
|
|
retransmit_queue.push(nv);
|
|
}
|
|
}
|
|
|
|
/// Process a blob: Add blob to the ledger window. If a continuous set of blobs
|
|
/// starting from consumed is thereby formed, add that continuous
|
|
/// range of blobs to a queue to be sent on to the next stage.
|
|
pub fn process_blob(
|
|
leader_scheduler: &LeaderScheduler,
|
|
db_ledger: &mut DbLedger,
|
|
blob: &SharedBlob,
|
|
max_ix: u64,
|
|
pix: u64,
|
|
consume_queue: &mut Vec<Entry>,
|
|
tick_height: &mut u64,
|
|
done: &Arc<AtomicBool>,
|
|
) -> Result<()> {
|
|
let is_coding = blob.read().unwrap().is_coding();
|
|
|
|
// Check if the blob is in the range of our known leaders. If not, we return.
|
|
let slot = blob.read().unwrap().slot()?;
|
|
let leader = leader_scheduler.get_leader_for_slot(slot);
|
|
|
|
if leader.is_none() {
|
|
return Ok(());
|
|
}
|
|
|
|
// Insert the new blob into the window
|
|
let mut consumed_entries = if is_coding {
|
|
let erasure_key = ErasureCf::key(slot, pix);
|
|
let rblob = &blob.read().unwrap();
|
|
let size = rblob.size()?;
|
|
db_ledger.erasure_cf.put(
|
|
&db_ledger.db,
|
|
&erasure_key,
|
|
&rblob.data[..BLOB_HEADER_SIZE + size],
|
|
)?;
|
|
vec![]
|
|
} else {
|
|
let data_key = ErasureCf::key(slot, pix);
|
|
db_ledger.insert_data_blob(&data_key, &blob.read().unwrap())?
|
|
};
|
|
|
|
// TODO: Once erasure is fixed, readd that logic here
|
|
|
|
for entry in &consumed_entries {
|
|
*tick_height += entry.is_tick() as u64;
|
|
}
|
|
|
|
// For downloading storage blobs,
|
|
// we only want up to a certain index
|
|
// then stop
|
|
if max_ix != 0 && !consumed_entries.is_empty() {
|
|
let meta = db_ledger
|
|
.meta_cf
|
|
.get(&db_ledger.db, &MetaCf::key(slot))?
|
|
.expect("Expect metadata to exist if consumed entries is nonzero");
|
|
|
|
let consumed = meta.consumed;
|
|
|
|
// Check if we ran over the last wanted entry
|
|
if consumed > max_ix {
|
|
let extra_unwanted_entries_len = consumed - (max_ix + 1);
|
|
let consumed_entries_len = consumed_entries.len();
|
|
consumed_entries.truncate(consumed_entries_len - extra_unwanted_entries_len as usize);
|
|
done.store(true, Ordering::Relaxed);
|
|
}
|
|
}
|
|
|
|
consume_queue.extend(consumed_entries);
|
|
Ok(())
|
|
}
|
|
|
|
pub fn calculate_max_repair_entry_height(
|
|
num_peers: u64,
|
|
consumed: u64,
|
|
received: u64,
|
|
times: usize,
|
|
is_next_leader: bool,
|
|
) -> u64 {
|
|
// Calculate the highest blob index that this node should have already received
|
|
// via avalanche. The avalanche splits data stream into nodes and each node retransmits
|
|
// the data to their peer nodes. So there's a possibility that a blob (with index lower
|
|
// than current received index) is being retransmitted by a peer node.
|
|
if times >= 8 || is_next_leader {
|
|
// if repair backoff is getting high, or if we are the next leader,
|
|
// don't wait for avalanche. received - 1 is the index of the highest blob.
|
|
received
|
|
} else {
|
|
cmp::max(consumed, received.saturating_sub(num_peers))
|
|
}
|
|
}
|
|
|
|
#[cfg(test)]
|
|
mod test {
|
|
use super::*;
|
|
use ledger::{get_tmp_ledger_path, make_tiny_test_entries, Block};
|
|
use packet::{Blob, Packet, Packets, SharedBlob, PACKET_DATA_SIZE};
|
|
use rocksdb::{Options, DB};
|
|
use signature::{Keypair, KeypairUtil};
|
|
use std::io;
|
|
use std::io::Write;
|
|
use std::net::UdpSocket;
|
|
use std::sync::atomic::{AtomicBool, Ordering};
|
|
use std::sync::mpsc::channel;
|
|
use std::sync::Arc;
|
|
use std::time::Duration;
|
|
use streamer::{receiver, responder, PacketReceiver};
|
|
|
|
fn get_msgs(r: PacketReceiver, num: &mut usize) {
|
|
for _t in 0..5 {
|
|
let timer = Duration::new(1, 0);
|
|
match r.recv_timeout(timer) {
|
|
Ok(m) => *num += m.read().unwrap().packets.len(),
|
|
e => info!("error {:?}", e),
|
|
}
|
|
if *num == 10 {
|
|
break;
|
|
}
|
|
}
|
|
}
|
|
#[test]
|
|
pub fn streamer_debug() {
|
|
write!(io::sink(), "{:?}", Packet::default()).unwrap();
|
|
write!(io::sink(), "{:?}", Packets::default()).unwrap();
|
|
write!(io::sink(), "{:?}", Blob::default()).unwrap();
|
|
}
|
|
|
|
#[test]
|
|
pub fn streamer_send_test() {
|
|
let read = UdpSocket::bind("127.0.0.1:0").expect("bind");
|
|
read.set_read_timeout(Some(Duration::new(1, 0))).unwrap();
|
|
|
|
let addr = read.local_addr().unwrap();
|
|
let send = UdpSocket::bind("127.0.0.1:0").expect("bind");
|
|
let exit = Arc::new(AtomicBool::new(false));
|
|
let (s_reader, r_reader) = channel();
|
|
let t_receiver = receiver(
|
|
Arc::new(read),
|
|
exit.clone(),
|
|
s_reader,
|
|
"window-streamer-test",
|
|
);
|
|
let t_responder = {
|
|
let (s_responder, r_responder) = channel();
|
|
let t_responder = responder("streamer_send_test", Arc::new(send), r_responder);
|
|
let mut msgs = Vec::new();
|
|
for i in 0..10 {
|
|
let mut b = SharedBlob::default();
|
|
{
|
|
let mut w = b.write().unwrap();
|
|
w.data[0] = i as u8;
|
|
w.meta.size = PACKET_DATA_SIZE;
|
|
w.meta.set_addr(&addr);
|
|
}
|
|
msgs.push(b);
|
|
}
|
|
s_responder.send(msgs).expect("send");
|
|
t_responder
|
|
};
|
|
|
|
let mut num = 0;
|
|
get_msgs(r_reader, &mut num);
|
|
assert_eq!(num, 10);
|
|
exit.store(true, Ordering::Relaxed);
|
|
t_receiver.join().expect("join");
|
|
t_responder.join().expect("join");
|
|
}
|
|
|
|
#[test]
|
|
pub fn test_calculate_max_repair_entry_height() {
|
|
assert_eq!(calculate_max_repair_entry_height(20, 4, 11, 0, false), 4);
|
|
assert_eq!(calculate_max_repair_entry_height(0, 10, 90, 0, false), 90);
|
|
assert_eq!(calculate_max_repair_entry_height(15, 10, 90, 32, false), 90);
|
|
assert_eq!(calculate_max_repair_entry_height(15, 10, 90, 0, false), 75);
|
|
assert_eq!(calculate_max_repair_entry_height(90, 10, 90, 0, false), 10);
|
|
assert_eq!(calculate_max_repair_entry_height(90, 10, 50, 0, false), 10);
|
|
assert_eq!(calculate_max_repair_entry_height(90, 10, 99, 0, false), 10);
|
|
assert_eq!(calculate_max_repair_entry_height(90, 10, 101, 0, false), 11);
|
|
assert_eq!(calculate_max_repair_entry_height(90, 10, 101, 0, true), 101);
|
|
assert_eq!(
|
|
calculate_max_repair_entry_height(90, 10, 101, 30, true),
|
|
101
|
|
);
|
|
}
|
|
|
|
#[test]
|
|
pub fn test_retransmit() {
|
|
let leader = Keypair::new().pubkey();
|
|
let nonleader = Keypair::new().pubkey();
|
|
let leader_scheduler = LeaderScheduler::from_bootstrap_leader(leader);
|
|
let blob = SharedBlob::default();
|
|
|
|
let (blob_sender, blob_receiver) = channel();
|
|
|
|
// Expect blob from leader to be retransmitted
|
|
blob.write().unwrap().set_id(&leader).unwrap();
|
|
retransmit_all_leader_blocks(&vec![blob.clone()], &leader_scheduler, &blob_sender)
|
|
.expect("Expect successful retransmit");
|
|
let output_blob = blob_receiver
|
|
.try_recv()
|
|
.expect("Expect input blob to be retransmitted");
|
|
|
|
// Retransmitted blob should be missing the leader id
|
|
assert_ne!(*output_blob[0].read().unwrap(), *blob.read().unwrap());
|
|
// Set the leader in the retransmitted blob, should now match the original
|
|
output_blob[0].write().unwrap().set_id(&leader).unwrap();
|
|
assert_eq!(*output_blob[0].read().unwrap(), *blob.read().unwrap());
|
|
|
|
// Expect blob from nonleader to not be retransmitted
|
|
blob.write().unwrap().set_id(&nonleader).unwrap();
|
|
retransmit_all_leader_blocks(&vec![blob], &leader_scheduler, &blob_sender)
|
|
.expect("Expect successful retransmit");
|
|
assert!(blob_receiver.try_recv().is_err());
|
|
}
|
|
|
|
#[test]
|
|
pub fn test_find_missing_data_indexes_sanity() {
|
|
let slot = 0;
|
|
|
|
// Create RocksDb ledger
|
|
let db_ledger_path = get_tmp_ledger_path("test_find_missing_data_indexes_sanity");
|
|
let mut db_ledger = DbLedger::open(&db_ledger_path).unwrap();
|
|
|
|
// Early exit conditions
|
|
let empty: Vec<u64> = vec![];
|
|
assert_eq!(find_missing_data_indexes(slot, &db_ledger, 0, 0), empty);
|
|
assert_eq!(find_missing_data_indexes(slot, &db_ledger, 5, 5), empty);
|
|
assert_eq!(find_missing_data_indexes(slot, &db_ledger, 4, 3), empty);
|
|
|
|
let shared_blob = &make_tiny_test_entries(1).to_blobs()[0];
|
|
let first_index = 10;
|
|
{
|
|
let mut bl = shared_blob.write().unwrap();
|
|
bl.set_index(10).unwrap();
|
|
}
|
|
|
|
// Insert one blob at index = first_index
|
|
db_ledger
|
|
.write_blobs(slot, &vec![&*shared_blob.read().unwrap()])
|
|
.unwrap();
|
|
|
|
// The first blob has index = first_index. Thus, for i < first_index,
|
|
// given the input range of [i, first_index], the missing indexes should be
|
|
// [i, first_index - 1]
|
|
for i in 0..first_index {
|
|
let result = find_missing_data_indexes(slot, &db_ledger, i, first_index);
|
|
let expected: Vec<u64> = (i..first_index).collect();
|
|
|
|
assert_eq!(result, expected);
|
|
}
|
|
|
|
drop(db_ledger);
|
|
DB::destroy(&Options::default(), &db_ledger_path)
|
|
.expect("Expected successful database destruction");
|
|
}
|
|
|
|
#[test]
|
|
pub fn test_find_missing_data_indexes() {
|
|
let slot = 0;
|
|
// Create RocksDb ledger
|
|
let db_ledger_path = get_tmp_ledger_path("test_find_missing_data_indexes");
|
|
let mut db_ledger = DbLedger::open(&db_ledger_path).unwrap();
|
|
|
|
// Write entries
|
|
let gap = 10;
|
|
let num_entries = 10;
|
|
let shared_blobs = make_tiny_test_entries(num_entries).to_blobs();
|
|
for (b, i) in shared_blobs.iter().zip(0..shared_blobs.len() as u64) {
|
|
b.write().unwrap().set_index(i * gap).unwrap();
|
|
}
|
|
let blob_locks: Vec<_> = shared_blobs.iter().map(|b| b.read().unwrap()).collect();
|
|
let blobs: Vec<&Blob> = blob_locks.iter().map(|b| &**b).collect();
|
|
db_ledger.write_blobs(slot, &blobs).unwrap();
|
|
|
|
// Index of the first blob is 0
|
|
// Index of the second blob is "gap"
|
|
// Thus, the missing indexes should then be [1, gap - 1] for the input index
|
|
// range of [0, gap)
|
|
let expected: Vec<u64> = (1..gap).collect();
|
|
assert_eq!(
|
|
find_missing_data_indexes(slot, &db_ledger, 0, gap),
|
|
expected
|
|
);
|
|
assert_eq!(
|
|
find_missing_data_indexes(slot, &db_ledger, 1, gap),
|
|
expected,
|
|
);
|
|
assert_eq!(
|
|
find_missing_data_indexes(slot, &db_ledger, 0, gap - 1),
|
|
&expected[..expected.len() - 1],
|
|
);
|
|
|
|
for i in 0..num_entries as u64 {
|
|
for j in 0..i {
|
|
let expected: Vec<u64> = (j..i)
|
|
.flat_map(|k| {
|
|
let begin = k * gap + 1;
|
|
let end = (k + 1) * gap;
|
|
(begin..end)
|
|
}).collect();
|
|
assert_eq!(
|
|
find_missing_data_indexes(slot, &db_ledger, j * gap, i * gap),
|
|
expected,
|
|
);
|
|
}
|
|
}
|
|
|
|
drop(db_ledger);
|
|
DB::destroy(&Options::default(), &db_ledger_path)
|
|
.expect("Expected successful database destruction");
|
|
}
|
|
|
|
#[test]
|
|
pub fn test_no_missing_blob_indexes() {
|
|
let slot = 0;
|
|
// Create RocksDb ledger
|
|
let db_ledger_path = get_tmp_ledger_path("test_find_missing_data_indexes");
|
|
let mut db_ledger = DbLedger::open(&db_ledger_path).unwrap();
|
|
|
|
// Write entries
|
|
let num_entries = 10;
|
|
let shared_blobs = make_tiny_test_entries(num_entries).to_blobs();
|
|
let blob_locks: Vec<_> = shared_blobs.iter().map(|b| b.read().unwrap()).collect();
|
|
let blobs: Vec<&Blob> = blob_locks.iter().map(|b| &**b).collect();
|
|
db_ledger.write_blobs(slot, &blobs).unwrap();
|
|
|
|
let empty: Vec<u64> = vec![];
|
|
for i in 0..num_entries as u64 {
|
|
for j in 0..i {
|
|
assert_eq!(find_missing_data_indexes(slot, &db_ledger, j, i), empty);
|
|
}
|
|
}
|
|
|
|
drop(db_ledger);
|
|
DB::destroy(&Options::default(), &db_ledger_path)
|
|
.expect("Expected successful database destruction");
|
|
}
|
|
}
|