Commit Graph

188 Commits

Author SHA1 Message Date
Louis Dureuil
732c52093d
Processing time without autobatching implementation 2023-05-03 17:41:48 +02:00
bors[bot]
654a3a9e19
Merge #3688
3688: Following release v1.1.1: bring back changes into `main` r=curquiza a=curquiza

`@meilisearch/engine-team` ensure the changes we bring to `main` are the ones you want

Co-authored-by: Louis Dureuil <louis@meilisearch.com>
Co-authored-by: bors[bot] <26634292+bors[bot]@users.noreply.github.com>
Co-authored-by: Tamo <tamo@meilisearch.com>
Co-authored-by: dureuill <dureuill@users.noreply.github.com>
2023-04-24 11:38:23 +00:00
Louis Dureuil
fd583501d7
Use non_free_pages_size instead of real_disk_size to check task db space taken 2023-04-13 17:07:44 +02:00
bors[bot]
f9960be115
Merge #3659
3659: stops receiving tasks once the task queue is full r=Kerollmops a=irevoire

Give 20GiB to the task queue + once 50% of the task queue is used, it blocks itself and only receives task deletion requests to ensure we never get in a state where we can’t do anything.

Also, create a new error message when we reach this case:
```
Meilisearch cannot receive write operations because the size limit of the tasks database has been reached. Please delete tasks to continue performing write operations.
```

Co-authored-by: Tamo <tamo@meilisearch.com>
2023-04-13 09:11:12 +00:00
Tamo
b4fabce36d
update the error message + update the task db size to 20GiB with a limit at 50% 2023-04-12 18:54:11 +02:00
Tamo
be69ab320d
stops receiving tasks once the task queue is full 2023-04-12 18:54:11 +02:00
Louis Dureuil
a94e78ffb0
Disable autobatching of additions and deletions 2023-04-12 10:53:00 +02:00
Tamo
4d308d5237 Improve the health route by ensuring lmdb is not down
And refactorize slightly the auth controller.
2023-04-06 15:31:42 +02:00
Tamo
597d57bf1d Merge branch 'main' into bring-back-changes-v1.1.0 2023-04-05 11:32:14 +02:00
Tamo
3fb67f94f7 Reduce the time to import a dump by caching some datas
With this commit, for a dump containing 1M tasks we went form 1m02 to 6s
2023-03-29 14:44:15 +02:00
Tamo
cf5145b542 Reduce the time to import a dump
With this commit, for a dump containing 1M tasks we went from 3m36s to import the task queue down to 1m02s
2023-03-29 14:27:40 +02:00
Tamo
a2b151e877 ensure that the task queue is correctly imported
reduce the size of the snapshots file
2023-03-21 14:41:46 +01:00
Louis Dureuil
076a3d371c Eagerly compute stats as fallback to the cache.
- Refactor all around to avoid spawning indexes more times than necessary
2023-03-06 16:57:31 +01:00
Tamo
fd5c48941a Add cache on the indexes stats 2023-03-06 16:57:31 +01:00
Louis Dureuil
0202ff8ab4
Attempt to use default budget for faster startup 2023-02-28 10:55:43 +01:00
Louis Dureuil
3db613ff77
Don't iterate all indexes manually 2023-02-23 11:29:09 +01:00
Louis Dureuil
5822764be9
Skip computing index budget in tests 2023-02-23 11:23:39 +01:00
Louis Dureuil
a529bf160c
Compute budget 2023-02-23 11:23:39 +01:00
Louis Dureuil
80b060f920
Use LRU cache 2023-02-23 11:23:39 +01:00
Louis Dureuil
fdf043580c
Add LruMap 2023-02-23 11:23:38 +01:00
Louis Dureuil
42577403d8
Authentication: Directly pass the authfilter to the index scheduler 2023-02-22 16:35:52 +01:00
bors[bot]
b08a49a16e
Merge #3319 #3470
3319: Transparently resize indexes on MaxDatabaseSizeReached errors r=Kerollmops a=dureuill

# Pull Request

## Related issue
Related to https://github.com/meilisearch/meilisearch/discussions/3280, depends on https://github.com/meilisearch/milli/pull/760

## What does this PR do?

### User standpoint

- Meilisearch no longer fails tasks that encounter the `milli::UserError(MaxDatabaseSizeReached)` error.
- Instead, these tasks are retried after increasing the maximum size allocated to the index where the failure occurred.

### Implementation standpoint

- Add `Batch::index_uid` to get the `index_uid` of a batch of task if there is one
- `IndexMapper::create_or_open_index` now takes an additional `size` argument that allows to (re)open indexes with a size different from the base `IndexScheduler::index_size` field
- `IndexScheduler::tick` now returns a `Result<TickOutcome>` instead of a `Result<usize>`. This offers more explicit control over what the behavior should be wrt the next tick.
- Add `IndexStatus::BeingResized` that contains a handle that a thread can use to await for the resize operation to complete and the index to be available again.
- Add `IndexMapper::resize_index` to increase the size of an index.
- In `IndexScheduler::tick`, intercept task batches that failed due to `MaxDatabaseSizeReached` and resize the index that caused the error, then request a new tick that will eventually handle the still enqueued task.

## Testing the PR

The following diff can be applied to this branch to make testing the PR easier:

<details>


```diff
diff --git a/index-scheduler/src/index_mapper.rs b/index-scheduler/src/index_mapper.rs
index 553ab45a..022b2f00 100644
--- a/index-scheduler/src/index_mapper.rs
+++ b/index-scheduler/src/index_mapper.rs
`@@` -228,13 +228,15 `@@` impl IndexMapper {
 
         drop(lock);
 
+        std:🧵:sleep_ms(2000);
+
         let current_size = index.map_size()?;
         let closing_event = index.prepare_for_closing();
-        log::info!("Resizing index {} from {} to {} bytes", name, current_size, current_size * 2);
+        log::error!("Resizing index {} from {} to {} bytes", name, current_size, current_size * 2);
 
         closing_event.wait();
 
-        log::info!("Resized index {} from {} to {} bytes", name, current_size, current_size * 2);
+        log::error!("Resized index {} from {} to {} bytes", name, current_size, current_size * 2);
 
         let index_path = self.base_path.join(uuid.to_string());
         let index = self.create_or_open_index(&index_path, None, 2 * current_size)?;
`@@` -268,8 +270,10 `@@` impl IndexMapper {
             match index {
                 Some(Available(index)) => break index,
                 Some(BeingResized(ref resize_operation)) => {
+                    log::error!("waiting for resize end");
                     // Deadlock: no lock taken while doing this operation.
                     resize_operation.wait();
+                    log::error!("trying our luck again!");
                     continue;
                 }
                 Some(BeingDeleted) => return Err(Error::IndexNotFound(name.to_string())),
diff --git a/index-scheduler/src/lib.rs b/index-scheduler/src/lib.rs
index 11b17d05..242dc095 100644
--- a/index-scheduler/src/lib.rs
+++ b/index-scheduler/src/lib.rs
`@@` -908,6 +908,7 `@@` impl IndexScheduler {
     ///
     /// Returns the number of processed tasks.
     fn tick(&self) -> Result<TickOutcome> {
+        log::error!("ticking!");
         #[cfg(test)]
         {
             *self.run_loop_iteration.write().unwrap() += 1;
diff --git a/meilisearch/src/main.rs b/meilisearch/src/main.rs
index 050c825a..63f312f6 100644
--- a/meilisearch/src/main.rs
+++ b/meilisearch/src/main.rs
`@@` -25,7 +25,7 `@@` fn setup(opt: &Opt) -> anyhow::Result<()> {
 
 #[actix_web::main]
 async fn main() -> anyhow::Result<()> {
-    let (opt, config_read_from) = Opt::try_build()?;
+    let (mut opt, config_read_from) = Opt::try_build()?;
 
     setup(&opt)?;
 
`@@` -56,6 +56,8 `@@` We generated a secure master key for you (you can safely copy this token):
         _ => (),
     }
 
+    opt.max_index_size = byte_unit::Byte::from_str("1MB").unwrap();
+
     let (index_scheduler, auth_controller) = setup_meilisearch(&opt)?;
 
     #[cfg(all(not(debug_assertions), feature = "analytics"))]
```
</details>

Mainly, these debug changes do the following:

- Set the default index size to 1MiB so that index resizes are initially frequent
- Turn some logs from info to error so that they can be displayed with `--log-level ERROR` (hiding the other infos)
- Add a long sleep between the beginning and the end of the resize so that we can observe the `BeingResized` index status (otherwise it would never come up in my tests)

## Open questions

- Is the growth factor of x2 the correct solution? For a `Vec` in memory it makes sense, but here we're manipulating quantities that are potentially in the order of 500GiBs. For bigger indexes it may make more sense to add at most e.g. 100GiB on each resize operation, avoiding big steps like 500GiB -> 1TiB.

## PR checklist
Please check if your PR fulfills the following requirements:
- [ ] Does this PR fix an existing issue, or have you listed the changes applied in the PR description (and why they are needed)?
- [ ] Have you read the contributing guidelines?
- [ ] Have you made sure that the title is accurate and descriptive of the changes?

Thank you so much for contributing to Meilisearch!


3470: Autobatch addition and deletion r=irevoire a=irevoire

This PR adds the capability to meilisearch to batch document addition and deletion together.

Fix https://github.com/meilisearch/meilisearch/issues/3440

--------------

Things to check before merging;

- [x] What happens if we delete multiple time the same documents -> add a test
- [x] If a documentDeletion gets batched with a documentAddition but the index doesn't exist yet? It should not work

Co-authored-by: Louis Dureuil <louis@meilisearch.com>
Co-authored-by: Tamo <tamo@meilisearch.com>
2023-02-20 15:00:19 +00:00
Louis Dureuil
1116788475
Resize indexes when they're full 2023-02-20 13:55:32 +01:00
Louis Dureuil
6cc3797aa1
IndexScheduler::tick returns a TickOutcome 2023-02-20 13:55:31 +01:00
Clément Renault
4570d5bf3a
Merge remote-tracking branch 'origin/main' into temp-wildcard 2023-02-09 13:14:05 +01:00
Tamo
eaad84bd1d fix the test to handle the document deletion correctly 2023-02-09 11:29:13 +01:00
Tamo
860c993ef7
Handle the autobatching of deletion and addition in the scheduler 2023-02-08 20:53:19 +01:00
Kerollmops
a36b1dbd70
Fix the tasks with the new patterns 2023-02-01 18:21:45 +01:00
Tamo
a858531574 apply review comments 2023-01-25 14:51:36 +01:00
Tamo
bf94f89035 Update index-scheduler/src/lib.rs
Co-authored-by: Louis Dureuil <louis@meilisearch.com>
2023-01-25 11:31:50 +01:00
Tamo
c92948b143 Compute the size of the auth-controller, index-scheduler and all update files in the global stats 2023-01-25 11:25:02 +01:00
Tamo
767cb725a5
reimplement the batching of task with or without primary key in the autobatcher 2023-01-23 20:18:22 +01:00
Tamo
5672118bfa
When adding documents, trying to update the primary-key now throw an error
While updating the test suite I also noticed an issue with the indexed_documents value of failed task and had to update it.
I also named a bunch of snapshots that had no name sorry 😬
2023-01-23 17:32:13 +01:00
Kerollmops
e89973f1bf
Do not delete all tasks when no canceled-by matches 2023-01-18 15:50:46 +01:00
Louis Dureuil
233372abea
Remove --max-index-size and --max-task-db-size 2023-01-04 17:20:01 +01:00
amab8901
9a39c4e40d Get date from IndexMetaData 2022-12-22 11:46:17 +01:00
amab8901
0893b175dc Merge branch 'main' into 2983-forward-date-to-milli 2022-12-21 14:31:19 +01:00
amab8901
d5978d11e1 Refactor 2022-12-21 14:28:00 +01:00
amab8901
aa03e02fdc Apply Rustfmt 2022-12-19 19:24:56 +01:00
Louis Dureuil
869d331680
Clippy fixes after updating Rust to v1.66 2022-12-19 14:17:12 +01:00
amab8901
5a0a0468df Combine created and added into date 2022-12-16 08:11:12 +01:00
amab8901
d3eb8d2d5c Enable create_raw_index(...) to specify time 2022-12-14 10:44:25 +01:00
Kerollmops
7b2f2a4f9c
Do only one convertion to u64 2022-12-13 15:31:55 +01:00
jiangbo212
717dd36547 Merge branch 'fix-3037' of github.com:jiangbo212/meilisearch into fix-3037 2022-12-07 22:54:16 +08:00
Kerollmops
f1de3aa75a Make the tests use MB to trigger page size issues 2022-12-06 20:10:10 +01:00
Kerollmops
e4e4370a3c Clamp the databases size to the page size 2022-12-06 20:09:49 +01:00
jiangbo212
5a770ffe47 test fail fix 2022-12-03 22:48:38 +08:00
jiangbo212
bf96b6df93 clippy fix change 2022-11-30 17:59:06 +08:00
jiangbo212
9c28632498 Merge branch 'main' into fix-3037 2022-11-30 09:38:01 +08:00
jiangbo212
38982d13fe fix issue 3037 2022-11-30 00:03:22 +08:00