Merge pull request #5707 from Mubelotix/last_embedder_message

Add last embedder error in batches
This commit is contained in:
Louis Dureuil 2025-06-26 15:21:17 +00:00 committed by GitHub
commit 3aa6c3c750
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
35 changed files with 492 additions and 93 deletions

View file

@ -1,7 +1,10 @@
use std::collections::BTreeMap;
use std::sync::atomic::AtomicUsize;
use meili_snap::{json_string, snapshot};
use reqwest::IntoUrl;
use std::time::Duration;
use tokio::sync::mpsc;
use wiremock::matchers::{method, path};
use wiremock::{Mock, MockServer, Request, ResponseTemplate};
@ -334,6 +337,41 @@ async fn create_mock_raw() -> (MockServer, Value) {
(mock_server, embedder_settings)
}
async fn create_faulty_mock_raw(sender: mpsc::Sender<()>) -> (MockServer, Value) {
let mock_server = MockServer::start().await;
let count = AtomicUsize::new(0);
Mock::given(method("POST"))
.and(path("/"))
.respond_with(move |_req: &Request| {
let count = count.fetch_add(1, std::sync::atomic::Ordering::SeqCst);
if count >= 5 {
let _ = sender.try_send(());
ResponseTemplate::new(500)
.set_delay(Duration::from_secs(u64::MAX)) // Make the response hang forever
.set_body_string("Service Unavailable")
} else {
ResponseTemplate::new(500).set_body_string("Service Unavailable")
}
})
.mount(&mock_server)
.await;
let url = mock_server.uri();
let embedder_settings = json!({
"source": "rest",
"url": url,
"dimensions": 3,
"request": "{{text}}",
"response": "{{embedding}}",
"documentTemplate": "{{doc.name}}"
});
(mock_server, embedder_settings)
}
pub async fn post<T: IntoUrl>(url: T, text: &str) -> reqwest::Result<reqwest::Response> {
reqwest::Client::builder().build()?.post(url).json(&json!(text)).send().await
}
@ -2111,3 +2149,70 @@ async fn searchable_reindex() {
}
"###);
}
#[actix_rt::test]
async fn last_error_stats() {
let (sender, mut receiver) = mpsc::channel(10);
let (_mock, setting) = create_faulty_mock_raw(sender).await;
let server = get_server_vector().await;
let index = server.index("doggo");
let (response, code) = index
.update_settings(json!({
"embedders": {
"rest": setting,
},
}))
.await;
snapshot!(code, @"202 Accepted");
let task = server.wait_task(response.uid()).await;
snapshot!(task["status"], @r###""succeeded""###);
let documents = json!([
{"id": 0, "name": "will_return_500"},
{"id": 1, "name": "will_error"},
{"id": 2, "name": "must_error"},
]);
let (_value, code) = index.add_documents(documents, None).await;
snapshot!(code, @"202 Accepted");
// The task will eventually fail, so let's not wait for it.
// Let's just wait for the server's signal
receiver.recv().await;
let (response, _code) = index.filtered_batches(&[], &[], &[]).await;
snapshot!(json_string!(response["results"][0], {
".progress" => "[ignored]",
".stats.embedderRequests.total" => "[ignored]",
".startedAt" => "[ignored]"
}), @r#"
{
"uid": 1,
"progress": "[ignored]",
"details": {
"receivedDocuments": 3,
"indexedDocuments": null
},
"stats": {
"totalNbTasks": 1,
"status": {
"processing": 1
},
"types": {
"documentAdditionOrUpdate": 1
},
"indexUids": {
"doggo": 1
},
"embedderRequests": {
"total": "[ignored]",
"failed": 5,
"lastError": "runtime error: received internal error HTTP 500 from embedding server\n - server replied with `Service Unavailable`"
}
},
"duration": null,
"startedAt": "[ignored]",
"finishedAt": null,
"batchStrategy": "batched all enqueued tasks"
}
"#);
}