[fix-buffering] implement block_ram_buffer_max to avoid excessive RAM usage

This commit is contained in:
Alex 2024-03-27 15:26:08 +01:00
parent 95eb8808e8
commit 0d3e285d13
Signed by untrusted user: lx
GPG key ID: 0E496D15096376BE
7 changed files with 148 additions and 13 deletions

View file

@ -20,6 +20,7 @@ metadata_auto_snapshot_interval = "6h"
db_engine = "lmdb" db_engine = "lmdb"
block_size = "1M" block_size = "1M"
block_ram_buffer_max = "256MiB"
sled_cache_capacity = "128MiB" sled_cache_capacity = "128MiB"
sled_flush_every_ms = 2000 sled_flush_every_ms = 2000
@ -88,6 +89,7 @@ The following gives details about each available configuration option.
Top-level configuration options: Top-level configuration options:
[`allow_world_readable_secrets`](#allow_world_readable_secrets), [`allow_world_readable_secrets`](#allow_world_readable_secrets),
[`block_ram_buffer_max`](#block_ram_buffer_max),
[`block_size`](#block_size), [`block_size`](#block_size),
[`bootstrap_peers`](#bootstrap_peers), [`bootstrap_peers`](#bootstrap_peers),
[`compression_level`](#compression_level), [`compression_level`](#compression_level),
@ -420,6 +422,37 @@ files will remain available. This however means that chunks from existing files
will not be deduplicated with chunks from newly uploaded files, meaning you will not be deduplicated with chunks from newly uploaded files, meaning you
might use more storage space that is optimally possible. might use more storage space that is optimally possible.
#### `block_ram_buffer_max` (since v0.9.4) {#block_ram_buffer_max}
A limit on the total size of data blocks kept in RAM by S3 API nodes awaiting
to be sent to storage nodes asynchronously.
Explanation: since Garage wants to tolerate node failures, it uses quorum
writes to send data blocks to storage nodes: try to write the block to three
nodes, and return ok as soon as two writes complete. So even if all three nodes
are online, the third write always completes asynchronously. In general, there
are not many writes to a cluster, and the third asynchronous write can
terminate early enough so as to not cause unbounded RAM growth. However, if
the S3 API node is continuously receiving large quantities of data and the
third node is never able to catch up, many data blocks will be kept buffered in
RAM as they are awaiting transfer to the third node.
The `block_ram_buffer_max` sets a limit to the size of buffers that can be kept
in RAM in this process. When the limit is reached, backpressure is applied
back to the S3 client.
Note that this only counts buffers that have arrived to a certain stage of
processing (received from the client + encrypted and/or compressed as
necessary) and are ready to send to the storage nodes. Many other buffers will
not be counted and this is not a hard limit on RAM consumption. In particular,
if many clients send requests simultaneously with large objects, the RAM
consumption will always grow linearly with the number of concurrent requests,
as each request will use a few buffers of size `block_size` for receiving and
intermediate processing before even trying to send the data to the storage
node.
The default value is 256MiB.
#### `sled_cache_capacity` {#sled_cache_capacity} #### `sled_cache_capacity` {#sled_cache_capacity}
This parameter can be used to tune the capacity of the cache used by This parameter can be used to tune the capacity of the cache used by

View file

@ -225,6 +225,17 @@ block_bytes_read 120586322022
block_bytes_written 3386618077 block_bytes_written 3386618077
``` ```
#### `block_ram_buffer_free_kb` (gauge)
Kibibytes available for buffering blocks that have to be sent to remote nodes.
When clients send too much data to this node and a storage node is not receiving
data fast enough due to slower network conditions, this will decrease down to
zero and backpressure will be applied.
```
block_ram_buffer_free_kb 219829
```
#### `block_compression_level` (counter) #### `block_compression_level` (counter)
Exposes the block compression level configured for the Garage node. Exposes the block compression level configured for the Garage node.

View file

@ -1,3 +1,4 @@
use std::convert::TryInto;
use std::path::PathBuf; use std::path::PathBuf;
use std::sync::Arc; use std::sync::Arc;
use std::time::Duration; use std::time::Duration;
@ -10,7 +11,7 @@ use serde::{Deserialize, Serialize};
use tokio::fs; use tokio::fs;
use tokio::io::{AsyncReadExt, AsyncWriteExt, BufReader}; use tokio::io::{AsyncReadExt, AsyncWriteExt, BufReader};
use tokio::sync::{mpsc, Mutex, MutexGuard}; use tokio::sync::{mpsc, Mutex, MutexGuard, Semaphore};
use opentelemetry::{ use opentelemetry::{
trace::{FutureExt as OtelFutureExt, TraceContextExt, Tracer}, trace::{FutureExt as OtelFutureExt, TraceContextExt, Tracer},
@ -93,6 +94,7 @@ pub struct BlockManager {
pub(crate) system: Arc<System>, pub(crate) system: Arc<System>,
pub(crate) endpoint: Arc<Endpoint<BlockRpc, Self>>, pub(crate) endpoint: Arc<Endpoint<BlockRpc, Self>>,
buffer_kb_semaphore: Arc<Semaphore>,
pub(crate) metrics: BlockManagerMetrics, pub(crate) metrics: BlockManagerMetrics,
@ -152,11 +154,14 @@ impl BlockManager {
.netapp .netapp
.endpoint("garage_block/manager.rs/Rpc".to_string()); .endpoint("garage_block/manager.rs/Rpc".to_string());
let buffer_kb_semaphore = Arc::new(Semaphore::new(config.block_ram_buffer_max / 1024));
let metrics = BlockManagerMetrics::new( let metrics = BlockManagerMetrics::new(
config.compression_level, config.compression_level,
rc.rc.clone(), rc.rc.clone(),
resync.queue.clone(), resync.queue.clone(),
resync.errors.clone(), resync.errors.clone(),
buffer_kb_semaphore.clone(),
); );
let scrub_persister = PersisterShared::new(&system.metadata_dir, "scrub_info"); let scrub_persister = PersisterShared::new(&system.metadata_dir, "scrub_info");
@ -176,6 +181,7 @@ impl BlockManager {
resync, resync,
system, system,
endpoint, endpoint,
buffer_kb_semaphore,
metrics, metrics,
scrub_persister, scrub_persister,
tx_scrub_command: ArcSwapOption::new(None), tx_scrub_command: ArcSwapOption::new(None),
@ -361,6 +367,14 @@ impl BlockManager {
let (header, bytes) = DataBlock::from_buffer(data, self.compression_level) let (header, bytes) = DataBlock::from_buffer(data, self.compression_level)
.await .await
.into_parts(); .into_parts();
let permit = self
.buffer_kb_semaphore
.clone()
.acquire_many_owned((bytes.len() / 1024).try_into().unwrap())
.await
.ok_or_message("could not reserve space for buffer of data to send to remote nodes")?;
let put_block_rpc = let put_block_rpc =
Req::new(BlockRpc::PutBlock { hash, header })?.with_stream_from_buffer(bytes); Req::new(BlockRpc::PutBlock { hash, header })?.with_stream_from_buffer(bytes);
let put_block_rpc = if let Some(tag) = order_tag { let put_block_rpc = if let Some(tag) = order_tag {
@ -376,6 +390,7 @@ impl BlockManager {
&who[..], &who[..],
put_block_rpc, put_block_rpc,
RequestStrategy::with_priority(PRIO_NORMAL | PRIO_SECONDARY) RequestStrategy::with_priority(PRIO_NORMAL | PRIO_SECONDARY)
.with_drop_on_completion(permit)
.with_quorum(self.replication.write_quorum()), .with_quorum(self.replication.write_quorum()),
) )
.await?; .await?;

View file

@ -1,3 +1,7 @@
use std::sync::Arc;
use tokio::sync::Semaphore;
use opentelemetry::{global, metrics::*}; use opentelemetry::{global, metrics::*};
use garage_db as db; use garage_db as db;
@ -9,6 +13,7 @@ pub struct BlockManagerMetrics {
pub(crate) _rc_size: ValueObserver<u64>, pub(crate) _rc_size: ValueObserver<u64>,
pub(crate) _resync_queue_len: ValueObserver<u64>, pub(crate) _resync_queue_len: ValueObserver<u64>,
pub(crate) _resync_errored_blocks: ValueObserver<u64>, pub(crate) _resync_errored_blocks: ValueObserver<u64>,
pub(crate) _buffer_free_kb: ValueObserver<u64>,
pub(crate) resync_counter: BoundCounter<u64>, pub(crate) resync_counter: BoundCounter<u64>,
pub(crate) resync_error_counter: BoundCounter<u64>, pub(crate) resync_error_counter: BoundCounter<u64>,
@ -31,6 +36,7 @@ impl BlockManagerMetrics {
rc_tree: db::Tree, rc_tree: db::Tree,
resync_queue: CountedTree, resync_queue: CountedTree,
resync_errors: CountedTree, resync_errors: CountedTree,
buffer_semaphore: Arc<Semaphore>,
) -> Self { ) -> Self {
let meter = global::meter("garage_model/block"); let meter = global::meter("garage_model/block");
Self { Self {
@ -66,6 +72,15 @@ impl BlockManagerMetrics {
.with_description("Number of block hashes whose last resync resulted in an error") .with_description("Number of block hashes whose last resync resulted in an error")
.init(), .init(),
_buffer_free_kb: meter
.u64_value_observer("block.ram_buffer_free_kb", move |observer| {
observer.observe(buffer_semaphore.available_permits() as u64, &[])
})
.with_description(
"Available RAM in KiB to use for buffering data blocks to be written to remote nodes",
)
.init(),
resync_counter: meter resync_counter: meter
.u64_counter("block.resync_counter") .u64_counter("block.resync_counter")
.with_description("Number of calls to resync_block") .with_description("Number of calls to resync_block")

View file

@ -300,7 +300,11 @@ impl K2VRpcHandler {
let rs = RequestStrategy::with_priority(PRIO_NORMAL).without_timeout(); let rs = RequestStrategy::with_priority(PRIO_NORMAL).without_timeout();
let mut requests = nodes let mut requests = nodes
.iter() .iter()
.map(|node| self.system.rpc.call(&self.endpoint, *node, msg.clone(), rs)) .map(|node| {
self.system
.rpc
.call(&self.endpoint, *node, msg.clone(), rs.clone())
})
.collect::<FuturesUnordered<_>>(); .collect::<FuturesUnordered<_>>();
// Fetch responses. This procedure stops fetching responses when any of the following // Fetch responses. This procedure stops fetching responses when any of the following

View file

@ -33,8 +33,7 @@ use crate::ring::Ring;
const DEFAULT_TIMEOUT: Duration = Duration::from_secs(300); const DEFAULT_TIMEOUT: Duration = Duration::from_secs(300);
/// Strategy to apply when making RPC /// Strategy to apply when making RPC
#[derive(Copy, Clone)] pub struct RequestStrategy<T> {
pub struct RequestStrategy {
/// Min number of response to consider the request successful /// Min number of response to consider the request successful
pub rs_quorum: Option<usize>, pub rs_quorum: Option<usize>,
/// Should requests be dropped after enough response are received /// Should requests be dropped after enough response are received
@ -43,6 +42,8 @@ pub struct RequestStrategy {
pub rs_priority: RequestPriority, pub rs_priority: RequestPriority,
/// Custom timeout for this request /// Custom timeout for this request
rs_timeout: Timeout, rs_timeout: Timeout,
/// Data to drop when everything completes
rs_drop_on_complete: T,
} }
#[derive(Copy, Clone)] #[derive(Copy, Clone)]
@ -52,7 +53,19 @@ enum Timeout {
Custom(Duration), Custom(Duration),
} }
impl RequestStrategy { impl Clone for RequestStrategy<()> {
fn clone(&self) -> Self {
RequestStrategy {
rs_quorum: self.rs_quorum,
rs_interrupt_after_quorum: self.rs_interrupt_after_quorum,
rs_priority: self.rs_priority,
rs_timeout: self.rs_timeout,
rs_drop_on_complete: (),
}
}
}
impl RequestStrategy<()> {
/// Create a RequestStrategy with default timeout and not interrupting when quorum reached /// Create a RequestStrategy with default timeout and not interrupting when quorum reached
pub fn with_priority(prio: RequestPriority) -> Self { pub fn with_priority(prio: RequestPriority) -> Self {
RequestStrategy { RequestStrategy {
@ -60,8 +73,22 @@ impl RequestStrategy {
rs_interrupt_after_quorum: false, rs_interrupt_after_quorum: false,
rs_priority: prio, rs_priority: prio,
rs_timeout: Timeout::Default, rs_timeout: Timeout::Default,
rs_drop_on_complete: (),
} }
} }
/// Add an item to be dropped on completion
pub fn with_drop_on_completion<T>(self, drop_on_complete: T) -> RequestStrategy<T> {
RequestStrategy {
rs_quorum: self.rs_quorum,
rs_interrupt_after_quorum: self.rs_interrupt_after_quorum,
rs_priority: self.rs_priority,
rs_timeout: self.rs_timeout,
rs_drop_on_complete: drop_on_complete,
}
}
}
impl<T> RequestStrategy<T> {
/// Set quorum to be reached for request /// Set quorum to be reached for request
pub fn with_quorum(mut self, quorum: usize) -> Self { pub fn with_quorum(mut self, quorum: usize) -> Self {
self.rs_quorum = Some(quorum); self.rs_quorum = Some(quorum);
@ -83,6 +110,19 @@ impl RequestStrategy {
self.rs_timeout = Timeout::Custom(timeout); self.rs_timeout = Timeout::Custom(timeout);
self self
} }
/// Extract drop_on_complete item
fn extract_drop_on_complete(self) -> (RequestStrategy<()>, T) {
(
RequestStrategy {
rs_quorum: self.rs_quorum,
rs_interrupt_after_quorum: self.rs_interrupt_after_quorum,
rs_priority: self.rs_priority,
rs_timeout: self.rs_timeout,
rs_drop_on_complete: (),
},
self.rs_drop_on_complete,
)
}
} }
#[derive(Clone)] #[derive(Clone)]
@ -123,7 +163,7 @@ impl RpcHelper {
endpoint: &Endpoint<M, H>, endpoint: &Endpoint<M, H>,
to: Uuid, to: Uuid,
msg: N, msg: N,
strat: RequestStrategy, strat: RequestStrategy<()>,
) -> Result<S, Error> ) -> Result<S, Error>
where where
M: Rpc<Response = Result<S, Error>>, M: Rpc<Response = Result<S, Error>>,
@ -176,7 +216,7 @@ impl RpcHelper {
endpoint: &Endpoint<M, H>, endpoint: &Endpoint<M, H>,
to: &[Uuid], to: &[Uuid],
msg: N, msg: N,
strat: RequestStrategy, strat: RequestStrategy<()>,
) -> Result<Vec<(Uuid, Result<S, Error>)>, Error> ) -> Result<Vec<(Uuid, Result<S, Error>)>, Error>
where where
M: Rpc<Response = Result<S, Error>>, M: Rpc<Response = Result<S, Error>>,
@ -187,7 +227,7 @@ impl RpcHelper {
let resps = join_all( let resps = join_all(
to.iter() to.iter()
.map(|to| self.call(endpoint, *to, msg.clone(), strat)), .map(|to| self.call(endpoint, *to, msg.clone(), strat.clone())),
) )
.await; .await;
Ok(to Ok(to
@ -201,7 +241,7 @@ impl RpcHelper {
&self, &self,
endpoint: &Endpoint<M, H>, endpoint: &Endpoint<M, H>,
msg: N, msg: N,
strat: RequestStrategy, strat: RequestStrategy<()>,
) -> Result<Vec<(Uuid, Result<S, Error>)>, Error> ) -> Result<Vec<(Uuid, Result<S, Error>)>, Error>
where where
M: Rpc<Response = Result<S, Error>>, M: Rpc<Response = Result<S, Error>>,
@ -220,18 +260,19 @@ impl RpcHelper {
/// Make a RPC call to multiple servers, returning either a Vec of responses, /// Make a RPC call to multiple servers, returning either a Vec of responses,
/// or an error if quorum could not be reached due to too many errors /// or an error if quorum could not be reached due to too many errors
pub async fn try_call_many<M, N, H, S>( pub async fn try_call_many<M, N, H, S, T>(
&self, &self,
endpoint: &Arc<Endpoint<M, H>>, endpoint: &Arc<Endpoint<M, H>>,
to: &[Uuid], to: &[Uuid],
msg: N, msg: N,
strategy: RequestStrategy, strategy: RequestStrategy<T>,
) -> Result<Vec<S>, Error> ) -> Result<Vec<S>, Error>
where where
M: Rpc<Response = Result<S, Error>> + 'static, M: Rpc<Response = Result<S, Error>> + 'static,
N: IntoReq<M>, N: IntoReq<M>,
H: StreamingEndpointHandler<M> + 'static, H: StreamingEndpointHandler<M> + 'static,
S: Send + 'static, S: Send + 'static,
T: Send + 'static,
{ {
let quorum = strategy.rs_quorum.unwrap_or(to.len()); let quorum = strategy.rs_quorum.unwrap_or(to.len());
@ -260,12 +301,12 @@ impl RpcHelper {
.await .await
} }
async fn try_call_many_internal<M, N, H, S>( async fn try_call_many_internal<M, N, H, S, T>(
&self, &self,
endpoint: &Arc<Endpoint<M, H>>, endpoint: &Arc<Endpoint<M, H>>,
to: &[Uuid], to: &[Uuid],
msg: N, msg: N,
strategy: RequestStrategy, strategy: RequestStrategy<T>,
quorum: usize, quorum: usize,
) -> Result<Vec<S>, Error> ) -> Result<Vec<S>, Error>
where where
@ -273,9 +314,12 @@ impl RpcHelper {
N: IntoReq<M>, N: IntoReq<M>,
H: StreamingEndpointHandler<M> + 'static, H: StreamingEndpointHandler<M> + 'static,
S: Send + 'static, S: Send + 'static,
T: Send + 'static,
{ {
let msg = msg.into_req().map_err(garage_net::error::Error::from)?; let msg = msg.into_req().map_err(garage_net::error::Error::from)?;
let (strategy, drop_on_complete) = strategy.extract_drop_on_complete();
// Build future for each request // Build future for each request
// They are not started now: they are added below in a FuturesUnordered // They are not started now: they are added below in a FuturesUnordered
// object that will take care of polling them (see below) // object that will take care of polling them (see below)
@ -283,6 +327,7 @@ impl RpcHelper {
let self2 = self.clone(); let self2 = self.clone();
let msg = msg.clone(); let msg = msg.clone();
let endpoint2 = endpoint.clone(); let endpoint2 = endpoint.clone();
let strategy = strategy.clone();
(to, async move { (to, async move {
self2.call(&endpoint2, to, msg, strategy).await self2.call(&endpoint2, to, msg, strategy).await
}) })
@ -377,6 +422,7 @@ impl RpcHelper {
// they have to be put in a proper queue that is persisted to disk. // they have to be put in a proper queue that is persisted to disk.
tokio::spawn(async move { tokio::spawn(async move {
resp_stream.collect::<Vec<Result<_, _>>>().await; resp_stream.collect::<Vec<Result<_, _>>>().await;
drop(drop_on_complete);
}); });
} }
} }

View file

@ -52,6 +52,14 @@ pub struct Config {
)] )]
pub compression_level: Option<i32>, pub compression_level: Option<i32>,
/// Maximum amount of block data to buffer in RAM for sending to
/// remote nodes when these nodes are on slower links
#[serde(
deserialize_with = "deserialize_capacity",
default = "default_block_ram_buffer_max"
)]
pub block_ram_buffer_max: usize,
/// Skip the permission check of secret files. Useful when /// Skip the permission check of secret files. Useful when
/// POSIX ACLs (or more complex chmods) are used. /// POSIX ACLs (or more complex chmods) are used.
#[serde(default)] #[serde(default)]
@ -255,6 +263,9 @@ fn default_sled_flush_every_ms() -> u64 {
fn default_block_size() -> usize { fn default_block_size() -> usize {
1048576 1048576
} }
fn default_block_ram_buffer_max() -> usize {
256 * 1024 * 1024
}
fn default_compression() -> Option<i32> { fn default_compression() -> Option<i32> {
Some(1) Some(1)