fix(cluster): improve shard reconstruction validation and start background healing service

rust/src/server.rs

@@ -24,8 +24,8 @@ use crate::config::SmartStorageConfig;
 use crate::policy::{self, PolicyDecision, PolicyStore};
 use crate::error::StorageError;
 use crate::cluster::coordinator::DistributedStore;
-use crate::cluster::config::ErasureConfig;
 use crate::cluster::drive_manager::DriveManager;
+use crate::cluster::healing::HealingService;
 use crate::cluster::membership::MembershipManager;
 use crate::cluster::placement;
 use crate::cluster::protocol::NodeInfo;
@@ -237,9 +237,15 @@ impl StorageServer {
             .join_cluster(&cluster_config.seed_nodes)
             .await?;
 
+        // Build local shard stores (one per drive) for shared use
+        let local_shard_stores: Vec<Arc<ShardStore>> = drive_paths
+            .iter()
+            .map(|p| Arc::new(ShardStore::new(p.clone())))
+            .collect();
+
         // Start QUIC accept loop for incoming connections
-        let shard_store_for_accept = Arc::new(ShardStore::new(drive_paths[0].clone()));
-        let (quic_shutdown_tx, quic_shutdown_rx) = watch::channel(false);
+        let shard_store_for_accept = local_shard_stores[0].clone();
+        let (_quic_shutdown_tx, quic_shutdown_rx) = watch::channel(false);
         let transport_clone = transport.clone();
         tokio::spawn(async move {
             transport_clone
@@ -249,11 +255,24 @@ impl StorageServer {
 
         // Start heartbeat loop
         let membership_clone = membership.clone();
-        let (hb_shutdown_tx, hb_shutdown_rx) = watch::channel(false);
+        let (_hb_shutdown_tx, hb_shutdown_rx) = watch::channel(false);
         tokio::spawn(async move {
             membership_clone.heartbeat_loop(hb_shutdown_rx).await;
         });
 
+        // Start healing service
+        let healing_service = HealingService::new(
+            cluster_state.clone(),
+            &erasure_config,
+            local_shard_stores.clone(),
+            manifest_dir.clone(),
+            24, // scan every 24 hours
+        )?;
+        let (_heal_shutdown_tx, heal_shutdown_rx) = watch::channel(false);
+        tokio::spawn(async move {
+            healing_service.run(heal_shutdown_rx).await;
+        });
+
         // Create distributed store
         let distributed_store = DistributedStore::new(
             cluster_state,