From b44d3fc796484a50cd6854f20c9b46e5fddedc9d Mon Sep 17 00:00:00 2001 From: Alex Date: Wed, 8 Jun 2022 10:01:44 +0200 Subject: 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 Reviewed-on: https://git.deuxfleurs.fr/Deuxfleurs/garage/pulls/322 Co-authored-by: Alex Co-committed-by: Alex --- src/model/k2v/item_table.rs | 24 ++++++++++++++++++++---- 1 file changed, 20 insertions(+), 4 deletions(-) (limited to 'src/model/k2v') diff --git a/src/model/k2v/item_table.rs b/src/model/k2v/item_table.rs index 8b7cc08a..991fe66d 100644 --- a/src/model/k2v/item_table.rs +++ b/src/model/k2v/item_table.rs @@ -2,6 +2,7 @@ use serde::{Deserialize, Serialize}; use std::collections::BTreeMap; use std::sync::Arc; +use garage_db as db; use garage_util::data::*; use garage_table::crdt::*; @@ -221,7 +222,12 @@ impl TableSchema for K2VItemTable { type E = K2VItem; type Filter = ItemFilter; - fn updated(&self, old: Option<&Self::E>, new: Option<&Self::E>) { + fn updated( + &self, + tx: &mut db::Transaction, + old: Option<&Self::E>, + new: Option<&Self::E>, + ) -> db::TxOpResult<()> { // 1. Count let (old_entries, old_conflicts, old_values, old_bytes) = match old { None => (0, 0, 0, 0), @@ -239,7 +245,8 @@ impl TableSchema for K2VItemTable { .map(|e| &e.partition.partition_key) .unwrap_or_else(|| &new.unwrap().partition.partition_key); - if let Err(e) = self.counter_table.count( + let counter_res = self.counter_table.count( + tx, &count_pk, count_sk, &[ @@ -248,14 +255,23 @@ impl TableSchema for K2VItemTable { (VALUES, new_values - old_values), (BYTES, new_bytes - old_bytes), ], - ) { - error!("Could not update K2V counter for bucket {:?} partition {}; counts will now be inconsistent. {}", count_pk, count_sk, e); + ); + if let Err(e) = db::unabort(counter_res)? { + // This result can be returned by `counter_table.count()` for instance + // if messagepack serialization or deserialization fails at some step. + // Warn admin but ignore this error for now, that's all we can do. + error!( + "Unable to update K2V item counter for bucket {:?} partition {}: {}. Index values will be wrong!", + count_pk, count_sk, e + ); } // 2. Notify if let Some(new_ent) = new { self.subscriptions.notify(new_ent); } + + Ok(()) } #[allow(clippy::nonminimal_bool)] -- cgit v1.2.3