Allow many Meilisearch write addrs (#5102)

* Write to many Meilisearch write addrs

* Keep client results ordered

* Attach Read Meilisearch client to actix data

* Load balanced meilisearch Compose profile

* Nginx config (round_robin)

* Fix nginx

* Meilisearch + nginx in same net

* Fix env vars example

* Fix env example again

* Fix env again

* Use try_collect with FuturesOrdered

* maybe fix remove_documents

* Clippy

Author: fetchfern

apps/labrinth/.env.docker-compose

@@ -12,7 +12,8 @@ DATABASE_URL=postgresql://labrinth:labrinth@labrinth-postgres/labrinth
 DATABASE_MIN_CONNECTIONS=0
 DATABASE_MAX_CONNECTIONS=16
 
-MEILISEARCH_ADDR=http://labrinth-meilisearch:7700
+MEILISEARCH_READ_ADDR=http://localhost:7700
+MEILISEARCH_WRITE_ADDRS=http://localhost:7700
 MEILISEARCH_KEY=modrinth
 
 REDIS_URL=redis://labrinth-redis

apps/labrinth/.env.local

@@ -13,7 +13,13 @@ DATABASE_URL=postgresql://labrinth:labrinth@localhost/labrinth
 DATABASE_MIN_CONNECTIONS=0
 DATABASE_MAX_CONNECTIONS=16
 
-MEILISEARCH_ADDR=http://localhost:7700
+MEILISEARCH_READ_ADDR=http://localhost:7700
+MEILISEARCH_WRITE_ADDRS=http://localhost:7700
+
+# # For a sharded Meilisearch setup (sharded-meilisearch docker compose profile)
+# MEILISEARCH_READ_ADDR=http://localhost:7710
+# MEILISEARCH_WRITE_ADDRS=http://localhost:7700,http://localhost:7701
+
 MEILISEARCH_KEY=modrinth
 
 REDIS_URL=redis://localhost

apps/labrinth/nginx/meili-lb.conf

@@ -0,0 +1,14 @@
+upstream meilisearch_upstream {
+    server meilisearch0:7700;
+    server meilisearch1:7700;
+}
+
+server {
+    listen 80;
+
+    location / {
+        proxy_pass http://meilisearch_upstream;
+        proxy_set_header Host $host;
+        proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
+    }
+}

apps/labrinth/src/lib.rs

@@ -19,6 +19,7 @@ use crate::background_task::update_versions;
 use crate::database::ReadOnlyPgPool;
 use crate::queue::billing::{index_billing, index_subscriptions};
 use crate::queue::moderation::AutomatedModerationQueue;
+use crate::search::MeilisearchReadClient;
 use crate::util::anrok;
 use crate::util::archon::ArchonClient;
 use crate::util::env::{parse_strings_from_var, parse_var};
@@ -68,6 +69,7 @@ pub struct LabrinthConfig {
     pub email_queue: web::Data<EmailQueue>,
     pub archon_client: web::Data<ArchonClient>,
     pub gotenberg_client: GotenbergClient,
+    pub search_read_client: web::Data<MeilisearchReadClient>,
 }
 
 #[allow(clippy::too_many_arguments)]
@@ -274,6 +276,11 @@ pub fn app_setup(
         file_host,
         scheduler: Arc::new(scheduler),
         ip_salt,
+        search_read_client: web::Data::new(
+            search_config.make_loadbalanced_read_client().expect(
+                "Failed to make Meilisearch client for read operations",
+            ),
+        ),
         search_config,
         session_queue,
         payouts_queue: web::Data::new(PayoutsQueue::new()),
@@ -325,6 +332,7 @@ pub fn app_config(
     .app_data(labrinth_config.archon_client.clone())
     .app_data(web::Data::new(labrinth_config.stripe_client.clone()))
     .app_data(web::Data::new(labrinth_config.anrok_client.clone()))
+    .app_data(labrinth_config.search_read_client.clone())
     .app_data(labrinth_config.rate_limiter.clone())
     .configure({
         #[cfg(target_os = "linux")]
@@ -373,7 +381,8 @@ pub fn check_env_vars() -> bool {
     failed |= check_var::<String>("LABRINTH_EXTERNAL_NOTIFICATION_KEY");
     failed |= check_var::<String>("RATE_LIMIT_IGNORE_KEY");
     failed |= check_var::<String>("DATABASE_URL");
-    failed |= check_var::<String>("MEILISEARCH_ADDR");
+    failed |= check_var::<String>("MEILISEARCH_READ_ADDR");
+    failed |= check_var::<String>("MEILISEARCH_WRITE_ADDRS");
     failed |= check_var::<String>("MEILISEARCH_KEY");
     failed |= check_var::<String>("REDIS_URL");
     failed |= check_var::<String>("BIND_ADDR");

apps/labrinth/src/routes/v2/projects.rs

@@ -13,7 +13,9 @@ use crate::queue::moderation::AutomatedModerationQueue;
 use crate::queue::session::AuthQueue;
 use crate::routes::v3::projects::ProjectIds;
 use crate::routes::{ApiError, v2_reroute, v3};
-use crate::search::{SearchConfig, SearchError, search_for_project};
+use crate::search::{
+    MeilisearchReadClient, SearchConfig, SearchError, search_for_project,
+};
 use actix_web::{HttpRequest, HttpResponse, delete, get, patch, post, web};
 use serde::{Deserialize, Serialize};
 use sqlx::PgPool;
@@ -54,6 +56,7 @@ pub fn config(cfg: &mut web::ServiceConfig) {
 pub async fn project_search(
     web::Query(info): web::Query<SearchRequest>,
     config: web::Data<SearchConfig>,
+    read_client: web::Data<MeilisearchReadClient>,
 ) -> Result<HttpResponse, SearchError> {
     // Search now uses loader_fields instead of explicit 'client_side' and 'server_side' fields
     // While the backend for this has changed, it doesnt affect much
@@ -99,7 +102,7 @@ pub async fn project_search(
         ..info
     };
 
-    let results = search_for_project(&info, &config).await?;
+    let results = search_for_project(&info, &config, &read_client).await?;
 
     let results = LegacySearchResults::from(results);
 

apps/labrinth/src/routes/v3/projects.rs

@@ -27,7 +27,9 @@ use crate::queue::moderation::AutomatedModerationQueue;
 use crate::queue::session::AuthQueue;
 use crate::routes::ApiError;
 use crate::search::indexing::remove_documents;
-use crate::search::{SearchConfig, SearchError, search_for_project};
+use crate::search::{
+    MeilisearchReadClient, SearchConfig, SearchError, search_for_project,
+};
 use crate::util::img;
 use crate::util::img::{delete_old_images, upload_image_optimized};
 use crate::util::routes::read_limited_from_payload;
@@ -1037,8 +1039,9 @@ pub async fn edit_project_categories(
 pub async fn project_search(
     web::Query(info): web::Query<SearchRequest>,
     config: web::Data<SearchConfig>,
+    read_client: web::Data<MeilisearchReadClient>,
 ) -> Result<HttpResponse, SearchError> {
-    let results = search_for_project(&info, &config).await?;
+    let results = search_for_project(&info, &config, &read_client).await?;
 
     // TODO: add this back
     // let results = ReturnSearchResults {