[ConfigNode] Revert procedure-based RegionGroup cleanup and fix RegionMaintainer retries

iotdb-client/service-rpc/src/main/java/org/apache/iotdb/rpc/TSStatusCode.java

@@ -197,6 +197,8 @@ public enum TSStatusCode {
   RECONSTRUCT_REGION_ERROR(908),
   EXTEND_REGION_ERROR(909),
   REMOVE_REGION_PEER_ERROR(910),
+  REGION_ALREADY_EXISTS(911),
+  REGION_NOT_EXIST(912),
 
   // Cluster Manager
   ADD_CONFIGNODE_ERROR(1000),

iotdb-core/confignode/src/main/i18n/en/org/apache/iotdb/confignode/i18n/ConfigNodeMessages.java

@@ -634,10 +634,6 @@ private ConfigNodeMessages() {}
       "Unexpected interruption while waiting for ConfigNode leader load warm-up.";
   public static final String EXCEPTION_PROCEDURE_FILE_ARG_EXCEEDS_THE_LOAD_BUFFER_LIMIT_ARG_ACTUAL_SIZE_ARG_62375B4C =
       "Procedure file %s exceeds the load buffer limit %s, actual size %s";
-  public static final String MESSAGE_DROPPING_LEGACY_REGION_DELETE_TASK_FOR_ARG_WHILE_REPLAYING_OFFER_PLAN_REGION_DELETION_IS_NOW_HANDLED_BY_REMOVEREGIONGROUPPROCEDURE_2A81A649 =
-      "Dropping legacy region-delete task for {} while replaying offer plan; region deletion is now handled by RemoveRegionGroupProcedure.";
-  public static final String MESSAGE_DROPPING_LEGACY_REGION_DELETE_TASK_FOR_ARG_WHILE_LOADING_SNAPSHOT_REGION_DELETION_IS_NOW_HANDLED_BY_REMOVEREGIONGROUPPROCEDURE_A9D409A0 =
-      "Dropping legacy region-delete task for {} while loading snapshot; region deletion is now handled by RemoveRegionGroupProcedure.";
   public static final String MESSAGE_CONFIGNODE_LEADER_IS_WARMING_UP_BEFORE_SERVING_THE_REGISTERING_CONFIGNODE_WILL_WAIT_2E051639 =
       "ConfigNode leader is warming up before serving the registering ConfigNode, will wait";
   public static final String MESSAGE_CONFIGNODE_LEADER_IS_WARMING_UP_BEFORE_SERVING_THE_REGISTERING_CONFIGNODE_WILL_WAIT_AND_RETRY_STATUS_ARG_RETRY_ARG_3C924873 =

iotdb-core/confignode/src/main/i18n/en/org/apache/iotdb/confignode/i18n/ManagerMessages.java

@@ -73,6 +73,7 @@ public final class ManagerMessages {
       "DataRegionGroupExtensionPolicy %s doesn't exist.";
   public static final String DECREASE_REFERENCE_COUNT_FOR_SNAPSHOT_ERROR =
       "Decrease reference count for snapshot {} error.";
+  public static final String DELETING_REGIONS_COSTS_MS = "Deleting regions costs {}ms";
   public static final String DETECTED_HISTORICAL_PIPE_COMPLETION_REPORT_FROM_DATANODE =
       "Detected historical pipe completion report from DataNode {} for pipe {}. remainingEventCount: {}, remainingTime: {}, completedDataNodes: {}";
   public static final String DETECTED_COMPLETION_OF_PIPE_STATIC_META_REMOVE_IT =
@@ -440,6 +441,8 @@ public final class ManagerMessages {
       "Start to create Region: {} on DataNode: {}";
   public static final String START_TO_CREATE_UDF_ON_DATA_NODES_NEEDTOSAVEJAR =
       "Start to create UDF [{}] on Data Nodes, needToSaveJar[{}]";
+  public static final String START_TO_DELETE_REGION_ON_DATANODE =
+      "Start to delete Region: {} on DataNode: {}";
   public static final String START_TRANSFER_OF = "Start transfer of {}";
   public static final String STOP_SUBMITTING_CQ_BECAUSE = "Stop submitting CQ {} because {}";
   public static final String STOP_SUBMITTING_CQ_BECAUSE_CURRENT_NODE_IS_NOT_LEADER_OR =
@@ -522,8 +525,6 @@ public final class ManagerMessages {
       "Unexpected interruption during waiting for configNode leader ready.";
   public static final String UNEXPECTED_INTERRUPTION_DURING_WAITING_FOR_GET_CLUSTER_ID =
       "Unexpected interruption during waiting for get cluster id.";
-  public static final String UNEXPECTED_NON_CREATE_REGION_MAINTAIN_TASK_SKIPPED =
-      "Unexpected non-create task in the RegionMaintainer queue; skipping it (the queue only recreates region replicas now, and region deletion is handled by RemoveRegionGroupProcedure).";
   public static final String UNEXPECTED_NULL_PROCEDURE_PARAMETERS_FOR_WAITINGPROCEDUREFINISHED =
       "Unexpected null procedure parameters for waitingProcedureFinished";
   public static final String UNKNOWN_DATAPARTITION_ALLOCATION_STRATEGY_USING_INHERIT_STRATEGY_BY_DEFAULT =

iotdb-core/confignode/src/main/i18n/en/org/apache/iotdb/confignode/i18n/ProcedureMessages.java

@@ -632,16 +632,6 @@ public final class ProcedureMessages {
   public static final String PID_ADDREGION_STATE_FAILED = "[pid{}][AddRegion] state {} failed";
   public static final String PID_ADDREGION_SUCCESS_HAS_BEEN_ADDED_TO_DATANODE_PROCEDURE_TOOK =
       "[pid{}][AddRegion] success, {} has been added to DataNode {}. Procedure took {} (start at {}).";
-  public static final String PID_REMOVEREGIONGROUP_STARTED_WILL_BE_DELETED =
-      "[pid{}][RemoveRegionGroup] started, region group {} will be deleted from DataNodes {}.";
-  public static final String PID_REMOVEREGIONGROUP_STARTED_REPLICA_WILL_BE_DELETED_FROM_DATANODE =
-      "[pid{}][RemoveRegionGroup] region {} will be deleted from DataNode {}.";
-  public static final String PID_REMOVEREGIONGROUP_STATE_FAILED =
-      "[pid{}][RemoveRegionGroup] state {} failed";
-  public static final String PID_REMOVEREGIONGROUP_DELETE_REPLICA_FAILED =
-      "[pid{}][RemoveRegionGroup] failed to delete a replica of region {} (attempt {}), will keep retrying until it is deleted. reason: {}";
-  public static final String PID_REMOVEREGIONGROUP_SUCCESS_PROCEDURE_TOOK =
-      "[pid{}][RemoveRegionGroup] success, region group {} has been deleted. Procedure took {} (started at {}).";
   public static final String PID_MIGRATEREGION_STARTED_WILL_BE_MIGRATED_FROM_DATANODE_TO =
       "[pid{}][MigrateRegion] started, {} will be migrated from DataNode {} to {}.";
   public static final String PID_MIGRATEREGION_STATE_COMPLETE =
@@ -1399,10 +1389,6 @@ private ProcedureMessages() {}
   public static final String LOG_ARG_8393DD4A = "{}";
   public static final String MESSAGE_HALT_PID_ARG_ACTIVECOUNT_ARG_411F3EBF = "Halt pid={}, activeCount={}";
   public static final String MESSAGE_EXCEPTION_HAPPENED_WHEN_WORKER_ARG_EXECUTE_PROCEDURE_ARG_6E3AD27D = "Exception happened when worker {} execute procedure {}";
-  public static final String EXCEPTION_CANNOT_DERIVE_A_COLLISION_FREE_DELETE_TASKID_PROCID_ARG_DELETETASKSEQ_ARG_EXCEED_THE_71B7046A =
-      "cannot derive a collision-free delete taskId: procId=%d, deleteTaskSeq=%d exceed the ";
-  public static final String EXCEPTION_CANNOT_DERIVE_A_COLLISION_FREE_DELETE_TASKID_PROCID_ARG_DELETETASKSEQ_ARG_EXCEED_THE_ARG_ARG_BIT_BUDGET_015C598D =
-      "cannot derive a collision-free delete taskId: procId=%d, deleteTaskSeq=%d exceed the %d/%d-bit budget";
   public static final String MESSAGE_FAILED_TO_SHOW_DATAPARTITIONTABLE_INTEGRITY_CHECK_PROGRESS_5EE98694 = "Failed to show DataPartitionTable integrity check progress";
   public static final String MESSAGE_ENCOUNTERED_UNEXPECTED_DATAPARTITIONTABLEINTEGRITYCHECKPROCEDURESTATE_ARG_WHEN_SHOWING_PROGRESS_5FA2739F =
       "Encountered unexpected DataPartitionTableIntegrityCheckProcedureState {} when showing progress";

iotdb-core/confignode/src/main/i18n/zh/org/apache/iotdb/confignode/i18n/ConfigNodeMessages.java

@@ -671,12 +671,6 @@ private ConfigNodeMessages() {}
   public static final String
       EXCEPTION_PROCEDURE_FILE_ARG_EXCEEDS_THE_LOAD_BUFFER_LIMIT_ARG_ACTUAL_SIZE_ARG_62375B4C =
           "Procedure 文件 %s 超过了加载缓冲区限制 %s，实际大小为 %s";
-  public static final String
-      MESSAGE_DROPPING_LEGACY_REGION_DELETE_TASK_FOR_ARG_WHILE_REPLAYING_OFFER_PLAN_REGION_DELETION_IS_NOW_HANDLED_BY_REMOVEREGIONGROUPPROCEDURE_2A81A649 =
-          "重放 offer plan 时丢弃 {} 的遗留 region-delete 任务；region 删除现已由 RemoveRegionGroupProcedure 处理。";
-  public static final String
-      MESSAGE_DROPPING_LEGACY_REGION_DELETE_TASK_FOR_ARG_WHILE_LOADING_SNAPSHOT_REGION_DELETION_IS_NOW_HANDLED_BY_REMOVEREGIONGROUPPROCEDURE_A9D409A0 =
-          "加载快照时丢弃 {} 的遗留 region-delete 任务；region 删除现已由 RemoveRegionGroupProcedure 处理。";
   public static final String
       MESSAGE_CONFIGNODE_LEADER_IS_WARMING_UP_BEFORE_SERVING_THE_REGISTERING_CONFIGNODE_WILL_WAIT_2E051639 =
           "ConfigNode leader 在服务注册中的 ConfigNode 前正在进行预热，将等待";

iotdb-core/confignode/src/main/i18n/zh/org/apache/iotdb/confignode/i18n/ManagerMessages.java

@@ -73,6 +73,7 @@ public final class ManagerMessages {
       "DataRegionGroupExtensionPolicy %s 不存在。";
   public static final String DECREASE_REFERENCE_COUNT_FOR_SNAPSHOT_ERROR =
       "减少快照 {} 的引用计数失败。";
+  public static final String DELETING_REGIONS_COSTS_MS = "删除 region 耗时 {}ms";
   public static final String DETECTED_HISTORICAL_PIPE_COMPLETION_REPORT_FROM_DATANODE =
       "检测到来自 DataNode {} 的历史 pipe 完成上报，pipe {}。remainingEventCount: {}, remainingTime: {}, completedDataNodes: {}";
   public static final String DETECTED_COMPLETION_OF_PIPE_STATIC_META_REMOVE_IT =
@@ -435,6 +436,8 @@ public final class ManagerMessages {
       "开始在 DataNode: {} 上创建 Region：{}";
   public static final String START_TO_CREATE_UDF_ON_DATA_NODES_NEEDTOSAVEJAR =
       "开始在 Data Nodes 上创建 UDF [{}]，needToSaveJar[{}]";
+  public static final String START_TO_DELETE_REGION_ON_DATANODE =
+      "开始在 DataNode: {} 上删除 Region：{}";
   public static final String START_TRANSFER_OF = "开始传输 {}";
   public static final String STOP_SUBMITTING_CQ_BECAUSE = "停止提交 CQ {}，原因：{}";
   public static final String STOP_SUBMITTING_CQ_BECAUSE_CURRENT_NODE_IS_NOT_LEADER_OR =
@@ -517,8 +520,6 @@ public final class ManagerMessages {
       "等待 configNode leader 就绪过程中发生意外中断。";
   public static final String UNEXPECTED_INTERRUPTION_DURING_WAITING_FOR_GET_CLUSTER_ID =
       "等待获取 cluster id 过程中发生意外中断。";
-  public static final String UNEXPECTED_NON_CREATE_REGION_MAINTAIN_TASK_SKIPPED =
-      "RegionMaintainer 队列中出现意外的非 create 任务；跳过处理（该队列目前仅用于重建 region 副本，region 删除由 RemoveRegionGroupProcedure 处理）。";
   public static final String UNEXPECTED_NULL_PROCEDURE_PARAMETERS_FOR_WAITINGPROCEDUREFINISHED =
       "waitingProcedureFinished 的 procedure 参数为空";
   public static final String UNKNOWN_DATAPARTITION_ALLOCATION_STRATEGY_USING_INHERIT_STRATEGY_BY_DEFAULT =

iotdb-core/confignode/src/main/i18n/zh/org/apache/iotdb/confignode/i18n/ProcedureMessages.java

@@ -598,16 +598,6 @@ public final class ProcedureMessages {
   public static final String PID_ADDREGION_STATE_FAILED = "[pid{}][AddRegion] 状态 {} 失败";
   public static final String PID_ADDREGION_SUCCESS_HAS_BEEN_ADDED_TO_DATANODE_PROCEDURE_TOOK =
       "[pid{}][AddRegion] 成功，{} 已添加到 DataNode {}。Procedure 耗时 {}（开始于 {}）。";
-  public static final String PID_REMOVEREGIONGROUP_STARTED_WILL_BE_DELETED =
-      "[pid{}][RemoveRegionGroup] 开始，region group {} 将从 DataNode {} 上删除。";
-  public static final String PID_REMOVEREGIONGROUP_STARTED_REPLICA_WILL_BE_DELETED_FROM_DATANODE =
-      "[pid{}][RemoveRegionGroup] region {} 将从 DataNode {} 上删除。";
-  public static final String PID_REMOVEREGIONGROUP_STATE_FAILED =
-      "[pid{}][RemoveRegionGroup] 状态 {} 失败";
-  public static final String PID_REMOVEREGIONGROUP_DELETE_REPLICA_FAILED =
-      "[pid{}][RemoveRegionGroup] 删除 region {} 的一个副本失败（第 {} 次尝试），将持续重试直到删除成功。原因：{}";
-  public static final String PID_REMOVEREGIONGROUP_SUCCESS_PROCEDURE_TOOK =
-      "[pid{}][RemoveRegionGroup] 成功，region group {} 已删除。过程耗时 {}（开始于 {}）。";
   public static final String PID_MIGRATEREGION_STARTED_WILL_BE_MIGRATED_FROM_DATANODE_TO =
       "[pid{}][MigrateRegion] 开始，{} 将从 DataNode {} 迁移到 {}。";
   public static final String PID_MIGRATEREGION_STATE_COMPLETE = "[pid{}][MigrateRegion] 状态 {} 完成";
@@ -1466,12 +1456,6 @@ private ProcedureMessages() {}
   public static final String
       MESSAGE_EXCEPTION_HAPPENED_WHEN_WORKER_ARG_EXECUTE_PROCEDURE_ARG_6E3AD27D =
           "worker {} 执行 procedure {} 时发生异常";
-  public static final String
-      EXCEPTION_CANNOT_DERIVE_A_COLLISION_FREE_DELETE_TASKID_PROCID_ARG_DELETETASKSEQ_ARG_EXCEED_THE_71B7046A =
-          "无法推导出无冲突的 delete taskId：procId=%d，deleteTaskSeq=%d 超出了 ";
-  public static final String
-      EXCEPTION_CANNOT_DERIVE_A_COLLISION_FREE_DELETE_TASKID_PROCID_ARG_DELETETASKSEQ_ARG_EXCEED_THE_ARG_ARG_BIT_BUDGET_015C598D =
-          "无法推导出无冲突的 delete taskId：procId=%d，deleteTaskSeq=%d 超出了 %d/%d 位的预算";
   public static final String
       MESSAGE_FAILED_TO_SHOW_DATAPARTITIONTABLE_INTEGRITY_CHECK_PROGRESS_5EE98694 =
           "显示 DataPartitionTable 完整性检查进度失败";

iotdb-core/confignode/src/main/java/org/apache/iotdb/confignode/client/async/CnToDnAsyncRequestType.java

@@ -39,6 +39,7 @@ public enum CnToDnAsyncRequestType {
   // Region Maintenance
   CREATE_DATA_REGION,
   CREATE_SCHEMA_REGION,
+  DELETE_REGION,
   RESET_PEER_LIST,
   NOTIFY_REGION_MIGRATION,
   UPDATE_REGION_ROUTE_MAP,

iotdb-core/confignode/src/main/java/org/apache/iotdb/confignode/client/async/CnToDnInternalServiceAsyncRequestManager.java

@@ -19,6 +19,7 @@
 
 package org.apache.iotdb.confignode.client.async;
 
+import org.apache.iotdb.common.rpc.thrift.TConsensusGroupId;
 import org.apache.iotdb.common.rpc.thrift.TDataNodeLocation;
 import org.apache.iotdb.common.rpc.thrift.TFlushReq;
 import org.apache.iotdb.common.rpc.thrift.TNodeLocations;
@@ -142,6 +143,10 @@ protected void initActionMapBuilder() {
         (req, client, handler) ->
             client.createDataRegion(
                 (TCreateDataRegionReq) req, (DataNodeTSStatusRPCHandler) handler));
+    actionMapBuilder.put(
+        CnToDnAsyncRequestType.DELETE_REGION,
+        (req, client, handler) ->
+            client.deleteRegion((TConsensusGroupId) req, (DataNodeTSStatusRPCHandler) handler));
     actionMapBuilder.put(
         CnToDnAsyncRequestType.CREATE_SCHEMA_REGION,
         (req, client, handler) ->

iotdb-core/confignode/src/main/java/org/apache/iotdb/confignode/manager/ProcedureManager.java

@@ -59,6 +59,7 @@
 import org.apache.iotdb.confignode.consensus.request.write.procedure.UpdateProcedurePlan;
 import org.apache.iotdb.confignode.consensus.request.write.region.CreateRegionGroupsPlan;
 import org.apache.iotdb.confignode.i18n.ManagerMessages;
+import org.apache.iotdb.confignode.manager.partition.PartitionManager;
 import org.apache.iotdb.confignode.persistence.ProcedureInfo;
 import org.apache.iotdb.confignode.procedure.PartitionTableAutoCleaner;
 import org.apache.iotdb.confignode.procedure.Procedure;
@@ -337,6 +338,9 @@ public TSStatus deleteDatabases(
     }
     List<TSStatus> results = new ArrayList<>(procedures.size());
     procedures.forEach(procedure -> results.add(waitingProcedureFinished(procedure)));
+    // Clear the previously deleted regions
+    final PartitionManager partitionManager = getConfigManager().getPartitionManager();
+    partitionManager.getRegionMaintainer().submit(partitionManager::maintainRegionReplicas);
     if (results.stream()
         .allMatch(result -> result.getCode() == TSStatusCode.SUCCESS_STATUS.getStatusCode())) {
       return StatusUtils.OK;