Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | block manager: read_block using old layout versions if necessary | Alex Auvolat | 2023-11-27 | 1 | -2/+4 |
| | |||||
* | layou: implement ack locking | Alex Auvolat | 2023-11-15 | 1 | -1/+1 |
| | |||||
* | table: take into account multiple write sets in inserts | Alex Auvolat | 2023-11-14 | 1 | -4/+3 |
| | |||||
* | layout: prepare for write sets | Alex Auvolat | 2023-11-14 | 1 | -1/+2 |
| | |||||
* | wip: split out layout management from System into separate LayoutManager | Alex Auvolat | 2023-11-09 | 1 | -5/+5 |
| | |||||
* | block manager: remove data_dir fieldmultihdd | Alex Auvolat | 2023-09-11 | 1 | -3/+0 |
| | |||||
* | block manager: fix bug where rebalance didn't delete old copies | Alex Auvolat | 2023-09-07 | 1 | -1/+38 |
| | |||||
* | block manager: add rebalance operation to rebalance multi-hdd setups | Alex Auvolat | 2023-09-07 | 1 | -9/+13 |
| | |||||
* | block manager: refactor and fix monitoring/statistics | Alex Auvolat | 2023-09-07 | 1 | -27/+20 |
| | |||||
* | block manager: avoid incorrect data_dir configs and avoid losing files | Alex Auvolat | 2023-09-06 | 1 | -2/+6 |
| | |||||
* | block manager: refactoring & increase max worker count to 8 | Alex Auvolat | 2023-09-06 | 1 | -13/+13 |
| | |||||
* | block manager: refactoring | Alex Auvolat | 2023-09-06 | 1 | -52/+44 |
| | |||||
* | block manager: get rid of check_block_status | Alex Auvolat | 2023-09-06 | 1 | -49/+35 |
| | |||||
* | block manager: move blocks in write_block if necessary | Alex Auvolat | 2023-09-06 | 1 | -12/+26 |
| | |||||
* | block manager: use data paths from layout | Alex Auvolat | 2023-09-06 | 1 | -72/+75 |
| | |||||
* | block manager: multi-directory layout computation | Alex Auvolat | 2023-09-06 | 1 | -3/+7 |
| | |||||
* | block manager: skeleton for multi-hdd support | Alex Auvolat | 2023-09-06 | 1 | -8/+21 |
| | |||||
* | Make fsync an option for meta and data | Alex Auvolat | 2023-06-09 | 1 | -13/+22 |
| | |||||
* | block/manager.rs: In is_block_compressed - check which compression_level | Jonathan Davies | 2023-05-09 | 1 | -5/+25 |
| | | | | | is configured on a node and check for raw block first if compression is disabled (to help reduce syscalls during a scrub). | ||||
* | block/manager.rs: Set defaults for scrub_persister. | Jonathan Davies | 2023-03-09 | 1 | -0/+1 |
| | |||||
* | block/manager.rs: Display scrub-next-run. | Jonathan Davies | 2023-03-06 | 1 | -0/+3 |
| | |||||
* | block/metrics.rs: Added compression_level metric. | Jonathan Davies | 2023-01-10 | 1 | -2/+6 |
| | |||||
* | Merge pull request 'PutObject: better cleanup when request is interrupted in ↵ | Alex | 2023-01-04 | 1 | -4/+32 |
|\ | | | | | | | | | | | the middle' (#462) from interrupted-cleanup into main Reviewed-on: https://git.deuxfleurs.fr/Deuxfleurs/garage/pulls/462 | ||||
| * | When saving block, delete .tmp file if we could not completeinterrupted-cleanup | Alex Auvolat | 2023-01-03 | 1 | -4/+32 |
| | | |||||
* | | Uniform framework for bg variable management | Alex Auvolat | 2023-01-04 | 1 | -2/+29 |
|/ | |||||
* | Refactor background runner and get rid of job worker | Alex Auvolat | 2022-12-14 | 1 | -5/+4 |
| | |||||
* | Spawn all background workers in a separate step | Alex Auvolat | 2022-12-14 | 1 | -12/+19 |
| | |||||
* | Add block.rc_size, table.size and table.merkle_tree_size metrics | Alex Auvolat | 2022-12-13 | 1 | -1/+2 |
| | |||||
* | cli: prettier table in garage stats | Alex Auvolat | 2022-12-13 | 1 | -0/+5 |
| | |||||
* | Implement `block list-errors` and `block info` | Alex Auvolat | 2022-12-13 | 1 | -0/+34 |
| | |||||
* | RPC performance changes | Alex Auvolat | 2022-09-19 | 1 | -9/+9 |
| | | | | | | - configurable ping timeout - single, much higher, configurable RPC timeout - no more concurrency semaphore | ||||
* | improvements in block manager | Alex Auvolat | 2022-09-12 | 1 | -13/+16 |
| | |||||
* | Merge branch 'lx-perf-improvements' into netapp-stream-body | Alex Auvolat | 2022-09-06 | 1 | -514/+53 |
|\ | |||||
| * | Merge branch 'main' into lx-perf-improvements | Alex Auvolat | 2022-09-05 | 1 | -505/+57 |
| |\ | |||||
| | * | Ability to have up to 4 concurrently working resync workers | Alex Auvolat | 2022-09-02 | 1 | -7/+5 |
| | | | |||||
| | * | Make BlockManagerLocked fully private again | Alex Auvolat | 2022-09-02 | 1 | -13/+22 |
| | | | |||||
| | * | block manager: refactor: split resync into separate file | Alex Auvolat | 2022-09-02 | 1 | -550/+48 |
| | | | |||||
| | * | Ability to dynamically set resync tranquility | Alex Auvolat | 2022-09-02 | 1 | -11/+51 |
| | | | |||||
* | | | Reenable node ordering | Alex Auvolat | 2022-09-06 | 1 | -6/+4 |
| | | | |||||
* | | | Apply PRIO_SECONDARY to block data transfers | Alex Auvolat | 2022-09-01 | 1 | -3/+3 |
| | | | |||||
* | | | Less strict timeouts | Alex Auvolat | 2022-09-01 | 1 | -2/+6 |
| | | | |||||
* | | | Update to Netapp with OrderTag support and exploit OrderTags | Alex Auvolat | 2022-09-01 | 1 | -16/+39 |
| | | | |||||
* | | | Fix bytes_read counter | Alex Auvolat | 2022-08-31 | 1 | -4/+9 |
| | | | |||||
* | | | Factor out node request order selection logic & use in manager | Alex Auvolat | 2022-07-29 | 1 | -0/+2 |
| | | | |||||
* | | | Use streaming in block manager | Alex Auvolat | 2022-07-29 | 1 | -55/+194 |
| | | | |||||
* | | | First adaptation to WIP netapp with streaming body | Alex Auvolat | 2022-07-29 | 1 | -10/+9 |
|/ / | |||||
* | | Performance improvements: | Alex Auvolat | 2022-07-29 | 1 | -12/+15 |
| | | | | | | | | | | - reduce contention on mutation_lock by having 256 of them - better lmdb defaults | ||||
* | | Compute hashes on dedicated threads | Alex Auvolat | 2022-07-29 | 1 | -2/+4 |
|/ | |||||
* | Background task manager (#332) | Alex | 2022-07-08 | 1 | -211/+118 |
| | | | | | | | | | | | | | | | | | | | | | - [x] New background worker trait - [x] Adapt all current workers to use new API - [x] Command to list currently running workers, and whether they are active, idle, or dead - [x] Error reporting - Optimizations - [x] Merkle updater: several items per iteration - [ ] Use `tokio::task::spawn_blocking` where appropriate so that CPU-intensive tasks don't block other things going on - scrub: - [x] have only one worker with a channel to start/pause/cancel - [x] automatic scrub - [x] ability to view and change tranquility from CLI - [x] persistence of a few info - [ ] Testing Co-authored-by: Alex Auvolat <alex@adnab.me> Reviewed-on: https://git.deuxfleurs.fr/Deuxfleurs/garage/pulls/332 Co-authored-by: Alex <alex@adnab.me> Co-committed-by: Alex <alex@adnab.me> | ||||
* | Abstract database behind generic interface and implement alternative drivers ↵ | Alex | 2022-06-08 | 1 | -35/+92 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | (#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> |