refactor: retrieve table routes in batches

GreptimeTeam · Jun 24, 2024 · 117ea03 · 117ea03
1 parent e698cc6
commit 117ea03
Showing 1 changed file with 46 additions and 15 deletions.
diff --git a/src/meta-srv/src/region/supervisor.rs b/src/meta-srv/src/region/supervisor.rs
@@ -12,6 +12,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use std::collections::HashSet;
 use std::fmt::Debug;
 use std::sync::{Arc, Mutex};
 use std::time::Duration;
@@ -247,6 +248,39 @@ impl RegionSupervisor {
                 return;
             }
         }
+
+        let table_ids = regions
+            .iter()
+            .map(|(_, _, region_id)| region_id.table_id())
+            .collect::<HashSet<_>>()
+            .into_iter()
+            .collect::<Vec<_>>();
+
+        let table_ids = match self
+            .table_metadata_manager
+            .table_route_manager()
+            .table_route_storage()
+            .batch_get(&table_ids)
+            .await
+            .context(error::TableMetadataManagerSnafu)
+        {
+            Ok(table_routes) => table_ids
+                .into_iter()
+                .zip(table_routes)
+                .flat_map(|(table_id, route)| {
+                    if route.is_some() {
+                        Some(table_id)
+                    } else {
+                        None
+                    }
+                })
+                .collect::<HashSet<_>>(),
+            Err(err) => {
+                error!(err; "Failed to retrieves table routes: {table_ids:?}");
+                return;
+            }
+        };
+
         warn!(
             "Detects region failures: {:?}",
             regions
@@ -255,9 +289,19 @@ impl RegionSupervisor {
                 .collect::<Vec<_>>()
         );
         for (cluster_id, datanode_id, region_id) in regions {
-            if let Err(err) = self.do_failover(cluster_id, datanode_id, region_id).await {
-                error!(err; "Failed to execute region failover for region: {region_id}, datanode: {datanode_id}");
+            if table_ids.contains(&region_id.table_id()) {
+                if let Err(err) = self.do_failover(cluster_id, datanode_id, region_id).await {
+                    error!(err; "Failed to execute region failover for region: {region_id}, datanode: {datanode_id}");
+                } else {
+                    self.failure_detector
+                        .remove(&(cluster_id, datanode_id, region_id));
+                }
             } else {
+                info!(
+                    "Skipping to execute region failover for region: {}, target table:{} is not exists",
+                    region_id,
+                    region_id.table_id()
+                );
                 self.failure_detector
                     .remove(&(cluster_id, datanode_id, region_id));
             }
@@ -277,7 +321,6 @@ impl RegionSupervisor {
         datanode_id: DatanodeId,
         region_id: RegionId,
     ) -> Result<()> {
-        // TODO(weny): skip the failover if meta is under maintenance mode.
         let task = self.region_migration_manager.tracker().get(region_id);
         match task {
             Some(task) => {
@@ -287,18 +330,6 @@ impl RegionSupervisor {
                 );
             }
             None => {
-                if self
-                    .table_metadata_manager
-                    .table_route_manager()
-                    .table_route_storage()
-                    .get(region_id.table_id())
-                    .await
-                    .context(error::TableMetadataManagerSnafu)?
-                    .is_none()
-                {
-                    return Ok(());
-                }
-
                 let mut peers = self
                     .selector
                     .select(