Begin work on sync...

This commit is contained in:
Alex Auvolat 2020-04-16 14:50:49 +02:00
parent 43ce5e4ab4
commit f01c1e71b5
9 changed files with 289 additions and 86 deletions

View File

@ -68,7 +68,7 @@ impl BackgroundRunner {
let _: Result<_, _> = self.queue_in.clone().send((boxed, true));
}
pub async fn spawn_worker<F, T>(self: Arc<Self>, worker: F)
pub async fn spawn_worker<F, T>(&self, worker: F)
where
F: FnOnce(watch::Receiver<bool>) -> T + Send + 'static,
T: Future<Output = JobOutput> + Send + 'static,

View File

@ -5,6 +5,7 @@ mod proto;
mod background;
mod membership;
mod table;
mod table_sync;
mod block;
mod block_ref_table;
@ -36,11 +37,11 @@ pub struct Opt {
#[structopt(short = "h", long = "rpc-host", default_value = "127.0.0.1:3901")]
rpc_host: SocketAddr,
#[structopt(long="ca-cert")]
#[structopt(long = "ca-cert")]
ca_cert: Option<String>,
#[structopt(long="client-cert")]
#[structopt(long = "client-cert")]
client_cert: Option<String>,
#[structopt(long="client-key")]
#[structopt(long = "client-key")]
client_key: Option<String>,
#[structopt(subcommand)]
@ -86,13 +87,11 @@ async fn main() {
let opt = Opt::from_args();
let tls_config = match (opt.ca_cert, opt.client_cert, opt.client_key) {
(Some(ca_cert), Some(client_cert), Some(client_key)) => {
Some(TlsConfig{
(Some(ca_cert), Some(client_cert), Some(client_key)) => Some(TlsConfig {
ca_cert,
node_cert: client_cert,
node_key: client_key,
})
}
}),
(None, None, None) => None,
_ => {
eprintln!("Missing one of: --ca-cert, --node-cert, --node-key. Not using TLS.");

View File

@ -157,7 +157,7 @@ impl Ring {
self.walk_ring_from_pos(start, n)
}
fn walk_ring_from_pos(&self, start: usize, n: usize) -> Vec<UUID> {
pub fn walk_ring_from_pos(&self, start: usize, n: usize) -> Vec<UUID> {
let mut ret = vec![];
let mut datacenters = vec![];
@ -282,13 +282,13 @@ impl System {
.collect::<Vec<_>>();
self.clone().ping_nodes(bootstrap_peers).await;
self.background
.clone()
self.clone()
.background
.spawn_worker(|stop_signal| self.ping_loop(stop_signal).map(Ok))
.await;
}
pub async fn ping_nodes(self: Arc<Self>, peers: Vec<(SocketAddr, Option<UUID>)>) {
async fn ping_nodes(self: Arc<Self>, peers: Vec<(SocketAddr, Option<UUID>)>) {
let ping_msg = self.make_ping();
let ping_resps = join_all(peers.iter().map(|(addr, id_option)| {
let sys = self.clone();

View File

@ -142,7 +142,10 @@ impl RpcClient {
let resp = tokio::time::timeout(timeout, resp_fut)
.await?
.map_err(|e| {
eprintln!("RPC HTTP client error when connecting to {}: {}", to_addr, e);
eprintln!(
"RPC HTTP client error when connecting to {}: {}",
to_addr, e
);
e
})?;
@ -158,4 +161,3 @@ impl RpcClient {
}
}
}

View File

@ -76,9 +76,7 @@ async fn handler(
// and the request handler simply sits there waiting for the task to finish.
// (if it's cancelled, that's not an issue)
// (TODO FIXME except if garage happens to shut down at that point)
let write_fut = async move {
garage.block_manager.write_block(&m.hash, &m.data).await
};
let write_fut = async move { garage.block_manager.write_block(&m.hash, &m.data).await };
tokio::spawn(write_fut).await?
}
Message::GetBlock(h) => garage.block_manager.read_block(&h).await,

View File

@ -93,7 +93,7 @@ impl Garage {
&db,
"block_ref".to_string(),
data_rep_param.clone(),
);
).await;
let version_table = Table::new(
VersionTable {
background: background.clone(),
@ -103,7 +103,7 @@ impl Garage {
&db,
"version".to_string(),
meta_rep_param.clone(),
);
).await;
let object_table = Table::new(
ObjectTable {
background: background.clone(),
@ -113,7 +113,7 @@ impl Garage {
&db,
"object".to_string(),
meta_rep_param.clone(),
);
).await;
let mut garage = Self {
db,

View File

@ -12,6 +12,7 @@ use crate::error::Error;
use crate::membership::System;
use crate::proto::*;
use crate::rpc_client::*;
use crate::table_sync::TableSyncer;
pub struct Table<F: TableSchema> {
pub instance: F,
@ -20,7 +21,6 @@ pub struct Table<F: TableSchema> {
pub system: Arc<System>,
pub store: sled::Tree,
pub partitions: Vec<Partition>,
pub param: TableReplicationParams,
}
@ -61,12 +61,6 @@ pub enum TableRPC<F: TableSchema> {
Update(Vec<Arc<ByteBuf>>),
}
pub struct Partition {
pub begin: Hash,
pub end: Hash,
pub other_nodes: Vec<UUID>,
}
pub trait PartitionKey {
fn hash(&self) -> Hash;
}
@ -124,7 +118,7 @@ pub trait TableSchema: Send + Sync {
}
impl<F: TableSchema + 'static> Table<F> {
pub fn new(
pub async fn new(
instance: F,
system: Arc<System>,
db: &sled::Db,
@ -132,14 +126,15 @@ impl<F: TableSchema + 'static> Table<F> {
param: TableReplicationParams,
) -> Arc<Self> {
let store = db.open_tree(&name).expect("Unable to open DB tree");
Arc::new(Self {
let table = Arc::new(Self {
instance,
name,
system,
store,
partitions: Vec::new(),
param,
})
});
TableSyncer::launch(table.clone()).await;
table
}
pub fn rpc_handler(self: Arc<Self>) -> Box<dyn TableRpcHandler + Send + Sync> {
@ -207,7 +202,11 @@ impl<F: TableSchema + 'static> Table<F> {
}
}
pub async fn get(&self, partition_key: &F::P, sort_key: &F::S) -> Result<Option<F::E>, Error> {
pub async fn get(
self: &Arc<Self>,
partition_key: &F::P,
sort_key: &F::S,
) -> Result<Option<F::E>, Error> {
let hash = partition_key.hash();
let who = self
.system
@ -245,17 +244,19 @@ impl<F: TableSchema + 'static> Table<F> {
}
if let Some(ret_entry) = &ret {
if not_all_same {
let _: Result<_, _> = self.repair_on_read(&who[..], &ret_entry).await;
self.system
.background
.spawn(self.clone().repair_on_read(who, ret_entry.clone()));
}
}
Ok(ret)
}
async fn repair_on_read(&self, who: &[UUID], what: &F::E) -> Result<(), Error> {
let what_enc = Arc::new(ByteBuf::from(rmp_to_vec_all_named(what)?));
async fn repair_on_read(self: Arc<Self>, who: Vec<UUID>, what: F::E) -> Result<(), Error> {
let what_enc = Arc::new(ByteBuf::from(rmp_to_vec_all_named(&what)?));
self.rpc_try_call_many(&who[..], &TableRPC::<F>::Update(vec![what_enc]), who.len())
.await
.map(|_| ())
.await?;
Ok(())
}
async fn rpc_try_call_many(

204
src/table_sync.rs Normal file
View File

@ -0,0 +1,204 @@
use rand::Rng;
use std::sync::Arc;
use std::time::Duration;
use std::collections::BTreeSet;
use futures::{pin_mut, select};
use futures_util::future::*;
use tokio::sync::watch;
use tokio::sync::Mutex;
use crate::data::*;
use crate::error::Error;
use crate::membership::{Ring, System};
use crate::table::*;
const SCAN_INTERVAL: Duration = Duration::from_secs(3600);
pub struct TableSyncer<F: TableSchema> {
pub table: Arc<Table<F>>,
pub todo: Mutex<SyncTodo>,
}
pub struct SyncTodo {
pub todo: Vec<Partition>,
}
#[derive(Debug, Clone)]
pub struct Partition {
pub begin: Hash,
pub end: Hash,
}
impl<F: TableSchema + 'static> TableSyncer<F> {
pub async fn launch(table: Arc<Table<F>>) -> Arc<Self> {
let todo = SyncTodo { todo: Vec::new() };
let syncer = Arc::new(TableSyncer {
table: table.clone(),
todo: Mutex::new(todo),
});
let s1 = syncer.clone();
table
.system
.background
.spawn_worker(move |must_exit: watch::Receiver<bool>| s1.watcher_task(must_exit))
.await;
let s2 = syncer.clone();
table
.system
.background
.spawn_worker(move |must_exit: watch::Receiver<bool>| s2.syncer_task(must_exit))
.await;
syncer
}
async fn watcher_task(
self: Arc<Self>,
mut must_exit: watch::Receiver<bool>,
) -> Result<(), Error> {
self.todo.lock().await.add_full_scan(&self.table);
let mut next_full_scan = tokio::time::delay_for(SCAN_INTERVAL).fuse();
let mut prev_ring: Arc<Ring> = self.table.system.ring.borrow().clone();
let mut ring_recv: watch::Receiver<Arc<Ring>> = self.table.system.ring.clone();
loop {
let s_ring_recv = ring_recv.recv().fuse();
let s_must_exit = must_exit.recv().fuse();
pin_mut!(s_ring_recv, s_must_exit);
select! {
_ = next_full_scan => {
next_full_scan = tokio::time::delay_for(SCAN_INTERVAL).fuse();
self.todo.lock().await.add_full_scan(&self.table);
}
new_ring_r = s_ring_recv => {
if let Some(new_ring) = new_ring_r {
self.todo.lock().await.add_ring_difference(&self.table, &prev_ring, &new_ring);
prev_ring = new_ring;
}
}
must_exit_v = s_must_exit => {
if must_exit_v.unwrap_or(false) {
return Ok(())
}
}
}
}
}
async fn syncer_task(
self: Arc<Self>,
mut must_exit: watch::Receiver<bool>,
) -> Result<(), Error> {
loop {
let s_pop_task = self.pop_task().fuse();
let s_must_exit = must_exit.recv().fuse();
pin_mut!(s_must_exit, s_pop_task);
select! {
task = s_pop_task => {
if let Some(partition) = task {
let res = self.sync_partition(&partition).await;
if let Err(e) = res {
eprintln!("Error while syncing {:?}: {}", partition, e);
}
} else {
tokio::time::delay_for(Duration::from_secs(1)).await;
}
}
must_exit_v = s_must_exit => {
if must_exit_v.unwrap_or(false) {
return Ok(())
}
}
}
}
}
async fn pop_task(&self) -> Option<Partition> {
self.todo.lock().await.pop_task()
}
async fn sync_partition(self: &Arc<Self>, partition: &Partition) -> Result<(), Error> {
unimplemented!()
}
}
impl SyncTodo {
fn add_full_scan<F: TableSchema>(&mut self, table: &Table<F>) {
let my_id = table.system.id.clone();
self.todo.clear();
let ring: Arc<Ring> = table.system.ring.borrow().clone();
for i in 0..ring.ring.len() {
let nodes = ring.walk_ring_from_pos(i, table.param.replication_factor);
let begin = ring.ring[i].location.clone();
if i == ring.ring.len() - 1 {
let end = ring.ring[0].location.clone();
self.add_full_scan_aux(table, begin, [0xffu8; 32].into(), &nodes[..], &my_id);
self.add_full_scan_aux(table, [0u8; 32].into(), end, &nodes[..], &my_id);
} else {
let end = ring.ring[i + 1].location.clone();
self.add_full_scan_aux(table, begin, end, &nodes[..], &my_id);
}
}
}
fn add_full_scan_aux<F: TableSchema>(
&mut self,
table: &Table<F>,
begin: Hash,
end: Hash,
nodes: &[UUID],
my_id: &UUID,
) {
if !nodes.contains(my_id) {
// Check if we have some data to send, otherwise skip
if table
.store
.range(begin.clone()..end.clone())
.next()
.is_none()
{}
}
self.todo.push(Partition { begin, end });
}
fn add_ring_difference<F: TableSchema>(&mut self, table: &Table<F>, old: &Ring, new: &Ring) {
let old_ring = ring_points(old);
let new_ring = ring_points(new);
unimplemented!()
}
fn pop_task(&mut self) -> Option<Partition> {
if self.todo.is_empty() {
return None;
}
let i = rand::thread_rng().gen_range::<usize, _, _>(0, self.todo.len());
if i == self.todo.len() - 1 {
self.todo.pop()
} else {
let replacement = self.todo.pop().unwrap();
let ret = std::mem::replace(&mut self.todo[i], replacement);
Some(ret)
}
}
}
fn ring_points(ring: &Ring) -> BTreeSet<Hash> {
let mut ret = BTreeSet::new();
ret.insert([0u8; 32].into());
ret.insert([0xFFu8; 32].into());
for i in 0..ring.ring.len() {
ret.insert(ring.ring[i].location.clone());
}
ret
}

View File

@ -1,17 +1,17 @@
use std::{fs, io};
use core::task::{Poll, Context};
use core::future::Future;
use core::task::{Context, Poll};
use std::pin::Pin;
use std::sync::Arc;
use core::future::Future;
use std::{fs, io};
use futures_util::future::*;
use tokio::io::{AsyncRead, AsyncWrite};
use rustls::internal::pemfile;
use hyper::client::HttpConnector;
use hyper::client::connect::Connection;
use hyper::client::HttpConnector;
use hyper::service::Service;
use hyper::Uri;
use hyper_rustls::MaybeHttpsStream;
use rustls::internal::pemfile;
use tokio::io::{AsyncRead, AsyncWrite};
use tokio_rustls::TlsConnector;
use webpki::DNSNameRef;
@ -58,7 +58,6 @@ pub fn load_private_key(filename: &str) -> Result<rustls::PrivateKey, Error> {
Ok(keys[0].clone())
}
// ---- AWFUL COPYPASTA FROM HYPER-RUSTLS connector.rs
// ---- ALWAYS USE `garage` AS HOSTNAME FOR TLS VERIFICATION
@ -85,7 +84,7 @@ impl HttpsConnectorFixedDnsname<HttpConnector> {
}
impl<T> Service<Uri> for HttpsConnectorFixedDnsname<T>
where
where
T: Service<Uri>,
T::Response: Connection + AsyncRead + AsyncWrite + Send + Unpin + 'static,
T::Future: Send + 'static,
@ -122,8 +121,8 @@ impl<T> Service<Uri> for HttpsConnectorFixedDnsname<T>
let cfg = self.tls_config.clone();
let connecting_future = self.http.call(dst);
let dnsname = DNSNameRef::try_from_ascii_str(self.fixed_dnsname)
.expect("Invalid fixed dnsname");
let dnsname =
DNSNameRef::try_from_ascii_str(self.fixed_dnsname).expect("Invalid fixed dnsname");
let f = async move {
let tcp = connecting_future.await.map_err(Into::into)?;