garage/src/model/s3/version_table.rs

277 lines
6.7 KiB
Rust
Raw Normal View History

use std::sync::Arc;
2020-04-09 21:45:07 +00:00
Abstract database behind generic interface and implement alternative drivers (#322) - [x] Design interface - [x] Implement Sled backend - [x] Re-implement the SledCountedTree hack ~~on Sled backend~~ on all backends (i.e. over the abstraction) - [x] Convert Garage code to use generic interface - [x] Proof-read converted Garage code - [ ] Test everything well - [x] Implement sqlite backend - [x] Implement LMDB backend - [ ] (Implement Persy backend?) - [ ] (Implement other backends? (like RocksDB, ...)) - [x] Implement backend choice in config file and garage server module - [x] Add CLI for converting between DB formats - Exploit the new interface to put more things in transactions - [x] `.updated()` trigger on Garage tables Fix #284 **Bugs** - [x] When exporting sqlite, trees iterate empty?? - [x] LMDB doesn't work **Known issues for various back-ends** - Sled: - Eats all my RAM and also all my disk space - `.len()` has to traverse the whole table - Is actually quite slow on some operations - And is actually pretty bad code... - Sqlite: - Requires a lock to be taken on all operations. The lock is also taken when iterating on a table with `.iter()`, and the lock isn't released until the iterator is dropped. This means that we must be VERY carefull to not do anything else inside a `.iter()` loop or else we will have a deadlock! Most such cases have been eliminated from the Garage codebase, but there might still be some that remain. If your Garage-over-Sqlite seems to hang/freeze, this is the reason. - (adapter uses a bunch of unsafe code) - Heed (LMDB): - Not suited for 32-bit machines as it has to map the whole DB in memory. - (adpater uses a tiny bit of unsafe code) **My recommendation:** avoid 32-bit machines and use LMDB as much as possible. **Converting databases** is actually quite easy. For example from Sled to LMDB: ```bash cd src/db cargo run --features cli --bin convert -- -i path/to/garage/meta/db -a sled -o path/to/garage/meta/db.lmdb -b lmdb ``` Then, just add this to your `config.toml`: ```toml db_engine = "lmdb" ``` Co-authored-by: Alex Auvolat <alex@adnab.me> Reviewed-on: https://git.deuxfleurs.fr/Deuxfleurs/garage/pulls/322 Co-authored-by: Alex <alex@adnab.me> Co-committed-by: Alex <alex@adnab.me>
2022-06-08 08:01:44 +00:00
use garage_db as db;
2020-04-24 10:10:01 +00:00
use garage_util::data::*;
2020-04-23 17:05:46 +00:00
use garage_table::crdt::*;
2021-03-26 18:41:46 +00:00
use garage_table::replication::TableShardedReplication;
2020-04-24 10:10:01 +00:00
use garage_table::*;
2020-04-09 21:45:07 +00:00
First implementation of K2V (#293) **Specification:** View spec at [this URL](https://git.deuxfleurs.fr/Deuxfleurs/garage/src/branch/k2v/doc/drafts/k2v-spec.md) - [x] Specify the structure of K2V triples - [x] Specify the DVVS format used for causality detection - [x] Specify the K2V index (just a counter of number of values per partition key) - [x] Specify single-item endpoints: ReadItem, InsertItem, DeleteItem - [x] Specify index endpoint: ReadIndex - [x] Specify multi-item endpoints: InsertBatch, ReadBatch, DeleteBatch - [x] Move to JSON objects instead of tuples - [x] Specify endpoints for polling for updates on single values (PollItem) **Implementation:** - [x] Table for K2V items, causal contexts - [x] Indexing mechanism and table for K2V index - [x] Make API handlers a bit more generic - [x] K2V API endpoint - [x] K2V API router - [x] ReadItem - [x] InsertItem - [x] DeleteItem - [x] PollItem - [x] ReadIndex - [x] InsertBatch - [x] ReadBatch - [x] DeleteBatch **Testing:** - [x] Just a simple Python script that does some requests to check visually that things are going right (does not contain parsing of results or assertions on returned values) - [x] Actual tests: - [x] Adapt testing framework - [x] Simple test with InsertItem + ReadItem - [x] Test with several Insert/Read/DeleteItem + ReadIndex - [x] Test all combinations of return formats for ReadItem - [x] Test with ReadBatch, InsertBatch, DeleteBatch - [x] Test with PollItem - [x] Test error codes - [ ] Fix most broken stuff - [x] test PollItem broken randomly - [x] when invalid causality tokens are given, errors should be 4xx not 5xx **Improvements:** - [x] Descending range queries - [x] Specify - [x] Implement - [x] Add test - [x] Batch updates to index counter - [x] Put K2V behind `k2v` feature flag Co-authored-by: Alex Auvolat <alex@adnab.me> Reviewed-on: https://git.deuxfleurs.fr/Deuxfleurs/garage/pulls/293 Co-authored-by: Alex <alex@adnab.me> Co-committed-by: Alex <alex@adnab.me>
2022-05-10 11:16:57 +00:00
use crate::s3::block_ref_table::*;
2023-01-03 13:44:47 +00:00
mod v05 {
use garage_util::crdt;
use garage_util::data::{Hash, Uuid};
use serde::{Deserialize, Serialize};
/// A version of an object
#[derive(PartialEq, Eq, Clone, Debug, Serialize, Deserialize)]
pub struct Version {
/// UUID of the version, used as partition key
pub uuid: Uuid,
// Actual data: the blocks for this version
// In the case of a multipart upload, also store the etags
// of individual parts and check them when doing CompleteMultipartUpload
/// Is this version deleted
pub deleted: crdt::Bool,
/// list of blocks of data composing the version
pub blocks: crdt::Map<VersionBlockKey, VersionBlock>,
/// Etag of each part in case of a multipart upload, empty otherwise
pub parts_etags: crdt::Map<u64, String>,
// Back link to bucket+key so that we can figure if
// this was deleted later on
/// Bucket in which the related object is stored
pub bucket: String,
/// Key in which the related object is stored
pub key: String,
}
#[derive(PartialEq, Eq, Clone, Copy, Debug, Serialize, Deserialize)]
pub struct VersionBlockKey {
/// Number of the part
pub part_number: u64,
/// Offset of this sub-segment in its part
pub offset: u64,
}
/// Informations about a single block
#[derive(PartialEq, Eq, Ord, PartialOrd, Clone, Copy, Debug, Serialize, Deserialize)]
pub struct VersionBlock {
/// Blake2 sum of the block
pub hash: Hash,
/// Size of the block
pub size: u64,
}
impl garage_util::migrate::InitialFormat for Version {}
}
mod v08 {
use garage_util::crdt;
use garage_util::data::Uuid;
use serde::{Deserialize, Serialize};
use super::v05;
2023-04-27 15:57:54 +00:00
pub use v05::{VersionBlock, VersionBlockKey};
2023-01-03 13:44:47 +00:00
/// A version of an object
#[derive(PartialEq, Eq, Clone, Debug, Serialize, Deserialize)]
pub struct Version {
/// UUID of the version, used as partition key
pub uuid: Uuid,
// Actual data: the blocks for this version
// In the case of a multipart upload, also store the etags
// of individual parts and check them when doing CompleteMultipartUpload
/// Is this version deleted
pub deleted: crdt::Bool,
/// list of blocks of data composing the version
pub blocks: crdt::Map<VersionBlockKey, VersionBlock>,
/// Etag of each part in case of a multipart upload, empty otherwise
pub parts_etags: crdt::Map<u64, String>,
// Back link to bucket+key so that we can figure if
// this was deleted later on
/// Bucket in which the related object is stored
pub bucket_id: Uuid,
/// Key in which the related object is stored
pub key: String,
}
impl garage_util::migrate::Migrate for Version {
type Previous = v05::Version;
fn migrate(old: v05::Version) -> Version {
use garage_util::data::blake2sum;
Version {
uuid: old.uuid,
deleted: old.deleted,
blocks: old.blocks,
parts_etags: old.parts_etags,
bucket_id: blake2sum(old.bucket.as_bytes()),
key: old.key,
}
}
}
2020-04-09 21:45:07 +00:00
}
2023-04-27 15:57:54 +00:00
pub(crate) mod v09 {
use garage_util::crdt;
use garage_util::data::Uuid;
use serde::{Deserialize, Serialize};
use super::v08;
pub use v08::{VersionBlock, VersionBlockKey};
/// A version of an object
#[derive(PartialEq, Eq, Clone, Debug, Serialize, Deserialize)]
pub struct Version {
/// UUID of the version, used as partition key
pub uuid: Uuid,
// Actual data: the blocks for this version
// In the case of a multipart upload, also store the etags
// of individual parts and check them when doing CompleteMultipartUpload
/// Is this version deleted
pub deleted: crdt::Bool,
/// list of blocks of data composing the version
pub blocks: crdt::Map<VersionBlockKey, VersionBlock>,
// Back link to bucket+key so that we can figure if
// this was deleted later on
pub backlink: VersionBacklink,
}
#[derive(PartialEq, Eq, Clone, Debug, Serialize, Deserialize)]
pub enum VersionBacklink {
Object {
/// Bucket in which the related object is stored
bucket_id: Uuid,
/// Key in which the related object is stored
key: String,
},
MultipartUpload {
upload_id: Uuid,
},
}
impl garage_util::migrate::Migrate for Version {
const VERSION_MARKER: &'static [u8] = b"G09s3v";
type Previous = v08::Version;
fn migrate(old: v08::Version) -> Version {
Version {
uuid: old.uuid,
deleted: old.deleted,
blocks: old.blocks,
backlink: VersionBacklink::Object {
bucket_id: old.bucket_id,
key: old.key,
},
}
}
}
}
pub use v09::*;
2023-01-03 13:44:47 +00:00
impl Version {
2023-04-27 15:57:54 +00:00
pub fn new(uuid: Uuid, backlink: VersionBacklink, deleted: bool) -> Self {
Self {
uuid,
deleted: deleted.into(),
blocks: crdt::Map::new(),
2023-04-27 15:57:54 +00:00
backlink,
}
}
2022-01-11 16:31:09 +00:00
pub fn has_part_number(&self, part_number: u64) -> bool {
2023-04-27 15:57:54 +00:00
self.blocks
2022-01-11 16:31:09 +00:00
.items()
.binary_search_by(|(k, _)| k.part_number.cmp(&part_number))
2023-04-27 15:57:54 +00:00
.is_ok()
2022-01-11 16:31:09 +00:00
}
}
impl Ord for VersionBlockKey {
fn cmp(&self, other: &Self) -> std::cmp::Ordering {
self.part_number
.cmp(&other.part_number)
.then(self.offset.cmp(&other.offset))
}
}
impl PartialOrd for VersionBlockKey {
fn partial_cmp(&self, other: &Self) -> Option<std::cmp::Ordering> {
Some(self.cmp(other))
}
}
2021-05-02 21:13:08 +00:00
impl AutoCrdt for VersionBlock {
const WARN_IF_DIFFERENT: bool = true;
2020-04-09 21:45:07 +00:00
}
2021-12-14 12:55:11 +00:00
impl Entry<Uuid, EmptyKey> for Version {
fn partition_key(&self) -> &Uuid {
2020-04-10 21:11:52 +00:00
&self.uuid
2020-04-09 21:45:07 +00:00
}
fn sort_key(&self) -> &EmptyKey {
&EmptyKey
2020-04-09 21:45:07 +00:00
}
fn is_tombstone(&self) -> bool {
self.deleted.get()
}
}
2020-04-09 21:45:07 +00:00
2021-05-02 21:13:08 +00:00
impl Crdt for Version {
2020-04-09 21:45:07 +00:00
fn merge(&mut self, other: &Self) {
self.deleted.merge(&other.deleted);
if self.deleted.get() {
2020-04-09 21:45:07 +00:00
self.blocks.clear();
} else {
self.blocks.merge(&other.blocks);
2020-04-09 21:45:07 +00:00
}
}
}
pub struct VersionTable {
pub block_ref_table: Arc<Table<BlockRefTable, TableShardedReplication>>,
2020-04-09 21:45:07 +00:00
}
2020-04-12 20:24:53 +00:00
impl TableSchema for VersionTable {
2021-12-14 11:34:01 +00:00
const TABLE_NAME: &'static str = "version";
2021-12-14 12:55:11 +00:00
type P = Uuid;
type S = EmptyKey;
2020-04-09 21:45:07 +00:00
type E = Version;
type Filter = DeletedFilter;
2020-04-09 21:45:07 +00:00
Abstract database behind generic interface and implement alternative drivers (#322) - [x] Design interface - [x] Implement Sled backend - [x] Re-implement the SledCountedTree hack ~~on Sled backend~~ on all backends (i.e. over the abstraction) - [x] Convert Garage code to use generic interface - [x] Proof-read converted Garage code - [ ] Test everything well - [x] Implement sqlite backend - [x] Implement LMDB backend - [ ] (Implement Persy backend?) - [ ] (Implement other backends? (like RocksDB, ...)) - [x] Implement backend choice in config file and garage server module - [x] Add CLI for converting between DB formats - Exploit the new interface to put more things in transactions - [x] `.updated()` trigger on Garage tables Fix #284 **Bugs** - [x] When exporting sqlite, trees iterate empty?? - [x] LMDB doesn't work **Known issues for various back-ends** - Sled: - Eats all my RAM and also all my disk space - `.len()` has to traverse the whole table - Is actually quite slow on some operations - And is actually pretty bad code... - Sqlite: - Requires a lock to be taken on all operations. The lock is also taken when iterating on a table with `.iter()`, and the lock isn't released until the iterator is dropped. This means that we must be VERY carefull to not do anything else inside a `.iter()` loop or else we will have a deadlock! Most such cases have been eliminated from the Garage codebase, but there might still be some that remain. If your Garage-over-Sqlite seems to hang/freeze, this is the reason. - (adapter uses a bunch of unsafe code) - Heed (LMDB): - Not suited for 32-bit machines as it has to map the whole DB in memory. - (adpater uses a tiny bit of unsafe code) **My recommendation:** avoid 32-bit machines and use LMDB as much as possible. **Converting databases** is actually quite easy. For example from Sled to LMDB: ```bash cd src/db cargo run --features cli --bin convert -- -i path/to/garage/meta/db -a sled -o path/to/garage/meta/db.lmdb -b lmdb ``` Then, just add this to your `config.toml`: ```toml db_engine = "lmdb" ``` Co-authored-by: Alex Auvolat <alex@adnab.me> Reviewed-on: https://git.deuxfleurs.fr/Deuxfleurs/garage/pulls/322 Co-authored-by: Alex <alex@adnab.me> Co-committed-by: Alex <alex@adnab.me>
2022-06-08 08:01:44 +00:00
fn updated(
&self,
tx: &mut db::Transaction,
Abstract database behind generic interface and implement alternative drivers (#322) - [x] Design interface - [x] Implement Sled backend - [x] Re-implement the SledCountedTree hack ~~on Sled backend~~ on all backends (i.e. over the abstraction) - [x] Convert Garage code to use generic interface - [x] Proof-read converted Garage code - [ ] Test everything well - [x] Implement sqlite backend - [x] Implement LMDB backend - [ ] (Implement Persy backend?) - [ ] (Implement other backends? (like RocksDB, ...)) - [x] Implement backend choice in config file and garage server module - [x] Add CLI for converting between DB formats - Exploit the new interface to put more things in transactions - [x] `.updated()` trigger on Garage tables Fix #284 **Bugs** - [x] When exporting sqlite, trees iterate empty?? - [x] LMDB doesn't work **Known issues for various back-ends** - Sled: - Eats all my RAM and also all my disk space - `.len()` has to traverse the whole table - Is actually quite slow on some operations - And is actually pretty bad code... - Sqlite: - Requires a lock to be taken on all operations. The lock is also taken when iterating on a table with `.iter()`, and the lock isn't released until the iterator is dropped. This means that we must be VERY carefull to not do anything else inside a `.iter()` loop or else we will have a deadlock! Most such cases have been eliminated from the Garage codebase, but there might still be some that remain. If your Garage-over-Sqlite seems to hang/freeze, this is the reason. - (adapter uses a bunch of unsafe code) - Heed (LMDB): - Not suited for 32-bit machines as it has to map the whole DB in memory. - (adpater uses a tiny bit of unsafe code) **My recommendation:** avoid 32-bit machines and use LMDB as much as possible. **Converting databases** is actually quite easy. For example from Sled to LMDB: ```bash cd src/db cargo run --features cli --bin convert -- -i path/to/garage/meta/db -a sled -o path/to/garage/meta/db.lmdb -b lmdb ``` Then, just add this to your `config.toml`: ```toml db_engine = "lmdb" ``` Co-authored-by: Alex Auvolat <alex@adnab.me> Reviewed-on: https://git.deuxfleurs.fr/Deuxfleurs/garage/pulls/322 Co-authored-by: Alex <alex@adnab.me> Co-committed-by: Alex <alex@adnab.me>
2022-06-08 08:01:44 +00:00
old: Option<&Self::E>,
new: Option<&Self::E>,
) -> db::TxOpResult<()> {
if let (Some(old_v), Some(new_v)) = (old, new) {
// Propagate deletion of version blocks
if new_v.deleted.get() && !old_v.deleted.get() {
let deleted_block_refs = old_v.blocks.items().iter().map(|(_k, vb)| BlockRef {
block: vb.hash,
version: old_v.uuid,
deleted: true.into(),
});
for block_ref in deleted_block_refs {
let res = self.block_ref_table.queue_insert(tx, &block_ref);
if let Err(e) = db::unabort(res)? {
error!("Unable to enqueue block ref deletion propagation: {}. A repair will be needed.", e);
}
}
}
}
Abstract database behind generic interface and implement alternative drivers (#322) - [x] Design interface - [x] Implement Sled backend - [x] Re-implement the SledCountedTree hack ~~on Sled backend~~ on all backends (i.e. over the abstraction) - [x] Convert Garage code to use generic interface - [x] Proof-read converted Garage code - [ ] Test everything well - [x] Implement sqlite backend - [x] Implement LMDB backend - [ ] (Implement Persy backend?) - [ ] (Implement other backends? (like RocksDB, ...)) - [x] Implement backend choice in config file and garage server module - [x] Add CLI for converting between DB formats - Exploit the new interface to put more things in transactions - [x] `.updated()` trigger on Garage tables Fix #284 **Bugs** - [x] When exporting sqlite, trees iterate empty?? - [x] LMDB doesn't work **Known issues for various back-ends** - Sled: - Eats all my RAM and also all my disk space - `.len()` has to traverse the whole table - Is actually quite slow on some operations - And is actually pretty bad code... - Sqlite: - Requires a lock to be taken on all operations. The lock is also taken when iterating on a table with `.iter()`, and the lock isn't released until the iterator is dropped. This means that we must be VERY carefull to not do anything else inside a `.iter()` loop or else we will have a deadlock! Most such cases have been eliminated from the Garage codebase, but there might still be some that remain. If your Garage-over-Sqlite seems to hang/freeze, this is the reason. - (adapter uses a bunch of unsafe code) - Heed (LMDB): - Not suited for 32-bit machines as it has to map the whole DB in memory. - (adpater uses a tiny bit of unsafe code) **My recommendation:** avoid 32-bit machines and use LMDB as much as possible. **Converting databases** is actually quite easy. For example from Sled to LMDB: ```bash cd src/db cargo run --features cli --bin convert -- -i path/to/garage/meta/db -a sled -o path/to/garage/meta/db.lmdb -b lmdb ``` Then, just add this to your `config.toml`: ```toml db_engine = "lmdb" ``` Co-authored-by: Alex Auvolat <alex@adnab.me> Reviewed-on: https://git.deuxfleurs.fr/Deuxfleurs/garage/pulls/322 Co-authored-by: Alex <alex@adnab.me> Co-committed-by: Alex <alex@adnab.me>
2022-06-08 08:01:44 +00:00
Ok(())
2020-04-09 21:45:07 +00:00
}
fn matches_filter(entry: &Self::E, filter: &Self::Filter) -> bool {
filter.apply(entry.deleted.get())
}
2020-04-09 21:45:07 +00:00
}