diff --git a/source/libs/executor/test/joinTests.cpp b/source/libs/executor/test/joinTests.cpp
index 6dd42c3a93..efbe1fcc83 100755
--- a/source/libs/executor/test/joinTests.cpp
+++ b/source/libs/executor/test/joinTests.cpp
@@ -41,6 +41,7 @@
 #include "tvariant.h"
 #include "stub.h"
 #include "querytask.h"
+#include "hashjoin.h"
 
 
 namespace {
@@ -3766,6 +3767,21 @@ TEST(leftWinJoin, noCondProjectionTest) {
 
 #endif
 
+#if 1
+TEST(functionsTest, branch) {
+  struct SOperatorInfo op = {0};
+  SHJoinOperatorInfo join;
+  SBufRowInfo bufrow = {0};
+  SSDataBlock blk = {0};
+
+  op.info = &join;
+  memset(&join, 0, sizeof(join));
+  join.ctx.pBuildRow = &bufrow;
+  blk.info.rows = 1;
+  join.finBlk = &blk;
+  hInnerJoinDo(&op);
+}
+#endif
 
 
 int main(int argc, char** argv) {
diff --git a/source/libs/planner/src/planner.c b/source/libs/planner/src/planner.c
index c3aa95f5b7..ee460e2610 100644
--- a/source/libs/planner/src/planner.c
+++ b/source/libs/planner/src/planner.c
@@ -161,6 +161,10 @@ int32_t qSubPlanToString(const SSubplan* pSubplan, char** pStr, int32_t* pLen) {
 int32_t qStringToSubplan(const char* pStr, SSubplan** pSubplan) { return nodesStringToNode(pStr, (SNode**)pSubplan); }
 
 int32_t qSubPlanToMsg(const SSubplan* pSubplan, char** pStr, int32_t* pLen) {
+  if (NULL == pSubplan) {
+    return terrno = TSDB_CODE_INVALID_PARA;
+  }
+  
   if (SUBPLAN_TYPE_MODIFY == pSubplan->subplanType && NULL == pSubplan->pNode) {
     SDataInserterNode* insert = (SDataInserterNode*)pSubplan->pDataSink;
     *pLen = insert->size;
diff --git a/source/libs/scheduler/inc/schInt.h b/source/libs/scheduler/inc/schInt.h
index 607f43a06f..cdbeb2a2e0 100644
--- a/source/libs/scheduler/inc/schInt.h
+++ b/source/libs/scheduler/inc/schInt.h
@@ -149,11 +149,11 @@ typedef struct SSchedulerMgmt {
   bool          exit;
   int32_t       jobRef;
   int32_t       jobNum;
-  SSchStat      stat;
   void         *timer;
   SRWLatch      hbLock;
   SHashObj     *hbConnections;
   void         *queryMgmt;
+  SSchStat      stat;
 } SSchedulerMgmt;
 
 typedef struct SSchCallbackParamHeader {
diff --git a/source/libs/scheduler/src/schFlowCtrl.c b/source/libs/scheduler/src/schFlowCtrl.c
index d9a0b88e40..c66f1dafe6 100644
--- a/source/libs/scheduler/src/schFlowCtrl.c
+++ b/source/libs/scheduler/src/schFlowCtrl.c
@@ -50,11 +50,6 @@ int32_t schChkJobNeedFlowCtrl(SSchJob *pJob, SSchLevel *pLevel) {
   int32_t taskNum = taosArrayGetSize(pJob->dataSrcTasks);
   for (int32_t i = 0; i < taskNum; ++i) {
     SSchTask *pTask = *(SSchTask **)taosArrayGet(pJob->dataSrcTasks, i);
-    if (NULL == pTask) {
-      SCH_JOB_DLOG("fail to get the %dth task", i);
-      SCH_ERR_RET(TSDB_CODE_SCH_INTERNAL_ERROR);
-    }
-
     sum += pTask->plan->execNodeStat.tableNum;
   }
 
diff --git a/source/libs/scheduler/src/schTask.c b/source/libs/scheduler/src/schTask.c
index b31353e97f..0e76e21aff 100644
--- a/source/libs/scheduler/src/schTask.c
+++ b/source/libs/scheduler/src/schTask.c
@@ -273,18 +273,18 @@ int32_t schProcessOnTaskSuccess(SSchJob *pJob, SSchTask *pTask) {
       if (taskDone < pTask->level->taskNum) {
         SCH_TASK_DLOG("wait all tasks, done:%d, all:%d", taskDone, pTask->level->taskNum);
         return TSDB_CODE_SUCCESS;
-      } else if (taskDone > pTask->level->taskNum) {
-        SCH_TASK_ELOG("taskDone number invalid, done:%d, total:%d", taskDone, pTask->level->taskNum);
       }
+      
+      SCH_TASK_DLOG("taskDone number reach level task number, done:%d, total:%d", taskDone, pTask->level->taskNum);
 
       if (pTask->level->taskFailed > 0) {
         SCH_RET(schHandleJobFailure(pJob, pJob->errCode));
-      } else {
-        SCH_RET(schSwitchJobStatus(pJob, JOB_TASK_STATUS_PART_SUCC, NULL));
       }
-    } else {
-      pJob->resNode = pTask->succeedAddr;
+
+      SCH_RET(schSwitchJobStatus(pJob, JOB_TASK_STATUS_PART_SUCC, NULL));
     }
+    
+    pJob->resNode = pTask->succeedAddr;
 
     pJob->fetchTask = pTask;
 
@@ -600,12 +600,12 @@ int32_t schPushTaskToExecList(SSchJob *pJob, SSchTask *pTask) {
   int32_t code = taosHashPut(pJob->execTasks, &pTask->taskId, sizeof(pTask->taskId), &pTask, POINTER_BYTES);
   if (0 != code) {
     if (HASH_NODE_EXIST(code)) {
-      SCH_TASK_DLOG("task already in execTask list, code:%x", code);
+      SCH_TASK_DLOG("task already in execTask list, code:0x%x", code);
       return TSDB_CODE_SUCCESS;
     }
 
-    SCH_TASK_ELOG("taosHashPut task to execTask list failed, errno:0x%x", errno);
-    SCH_ERR_RET(TSDB_CODE_OUT_OF_MEMORY);
+    SCH_TASK_ELOG("taosHashPut task to execTask list failed, code:0x%x", code);
+    SCH_ERR_RET(code);
   }
 
   SCH_TASK_DLOG("task added to execTask list, numOfTasks:%d", taosHashGetSize(pJob->execTasks));
@@ -800,11 +800,6 @@ int32_t schSetAddrsFromNodeList(SSchJob *pJob, SSchTask *pTask) {
 
     for (int32_t i = 0; i < nodeNum; ++i) {
       SQueryNodeLoad *nload = taosArrayGet(pJob->nodeList, i);
-      if (NULL == nload) {
-        SCH_TASK_ELOG("fail to get the %dth node in nodeList, nodeNum:%d", i, nodeNum);
-        SCH_ERR_RET(TSDB_CODE_SCH_INTERNAL_ERROR);
-      }
-
       SQueryNodeAddr *naddr = &nload->addr;
 
       if (NULL == taosArrayPush(pTask->candidateAddrs, naddr)) {
diff --git a/source/libs/scheduler/test/schedulerTests.cpp b/source/libs/scheduler/test/schedulerTests.cpp
index 50b2527802..135c151743 100644
--- a/source/libs/scheduler/test/schedulerTests.cpp
+++ b/source/libs/scheduler/test/schedulerTests.cpp
@@ -63,6 +63,15 @@ extern "C" int32_t schRescheduleTask(SSchJob *pJob, SSchTask *pTask);
 extern "C" int32_t schValidateRspMsgType(SSchJob *pJob, SSchTask *pTask, int32_t msgType);
 extern "C" int32_t schProcessFetchRsp(SSchJob *pJob, SSchTask *pTask, char *msg, int32_t rspCode);
 extern "C" int32_t schProcessResponseMsg(SSchJob *pJob, SSchTask *pTask, SDataBuf *pMsg, int32_t rspCode);
+extern "C" void schInitTaskRetryTimes(SSchJob *pJob, SSchTask *pTask, SSchLevel *pLevel);
+extern "C" int32_t schRecordTaskSucceedNode(SSchJob *pJob, SSchTask *pTask);
+extern "C" int32_t schDropTaskExecNode(SSchJob *pJob, SSchTask *pTask, void *handle, int32_t execId);
+extern "C" int32_t schPushTaskToExecList(SSchJob *pJob, SSchTask *pTask);
+extern "C" int32_t schSetAddrsFromNodeList(SSchJob *pJob, SSchTask *pTask);
+extern "C" int32_t schNotifyTaskOnExecNode(SSchJob *pJob, SSchTask *pTask, ETaskNotifyType type);
+extern "C" int32_t schLaunchRemoteTask(SSchJob *pJob, SSchTask *pTask);
+extern "C" int32_t schLaunchTaskImpl(void *param);
+extern "C" void schHandleTimerEvent(void *param, void *tmrId);
 
 
 int64_t insertJobRefId = 0;
@@ -1480,6 +1489,8 @@ TEST(otherTest, function) {
 TEST(otherTest, branch) {
   SSchJob job = {0};
   SSchTask task = {0};
+  memset(&schMgmt, 0, sizeof(schMgmt));
+  
   schValidateRspMsgType(&job, &task, TDMT_SCH_MERGE_FETCH_RSP);
 
   task.lastMsgType = TDMT_SCH_MERGE_FETCH_RSP - 1;
@@ -1532,6 +1543,7 @@ TEST(otherTest, branch) {
   databuf.pData = taosMemoryMalloc(0);
   job.status = JOB_TASK_STATUS_FAIL;
   schProcessResponseMsg(&job, &task, &databuf, 0);
+  job.status = 0;
 
   job.attr.explainMode = EXPLAIN_MODE_ANALYZE;
   databuf.msgType = TDMT_SCH_EXPLAIN_RSP;
@@ -1570,7 +1582,163 @@ TEST(otherTest, branch) {
   schBuildAndSendHbMsg(&ep, NULL);
 
   schBuildAndSendMsg(&job, &task, NULL, 0, NULL);
-  
+
+  SSchLevel level = {0};
+  SSubplan subplan;
+  memset(&subplan, 0, sizeof(subplan));
+  job.attr.queryJob = true;
+  schMgmt.cfg.schPolicy = SCH_ALL;
+  task.plan = &subplan;
+  schInitTaskRetryTimes(&job, &task, &level);
+  job.attr.queryJob = false;
+  memset(&schMgmt.cfg, 0, sizeof(schMgmt.cfg));
+  memset(&level, 0, sizeof(level));
+
+  schRecordTaskSucceedNode(&job, &task);
+
+  schDropTaskExecNode(&job, &task, NULL, 0);
+
+  task.execNodes = taosHashInit(1, taosGetDefaultHashFunction(TSDB_DATA_TYPE_INT), true, HASH_NO_LOCK);
+  schDropTaskExecNode(&job, &task, NULL, 0);
+
+  int32_t execId = 0;
+  task.execId = 1;
+  (void)taosHashPut(task.execNodes, &execId, sizeof(execId), &execId, sizeof(execId));
+  schDropTaskExecNode(&job, &task, NULL, execId);
+  task.execId = 0;
+  taosHashCleanup(task.execNodes);
+  task.execNodes = NULL;
+
+  job.status = JOB_TASK_STATUS_FAIL;
+  schProcessOnTaskFailure(&job, &task, 0);
+  job.status = 0;
+
+  task.status = JOB_TASK_STATUS_FAIL;
+  schProcessOnTaskFailure(&job, &task, 0);
+  task.status = 0;
+
+  task.level = &level;
+  schProcessOnTaskFailure(&job, &task, TSDB_CODE_SCH_TIMEOUT_ERROR);
+  memset(&level, 0, sizeof(level));
+  task.level = NULL;
+
+  subplan.subplanType = SUBPLAN_TYPE_SCAN;
+  task.plan = &subplan;
+  SEpSet epset = {0};
+  epset.numOfEps = 127;
+  schChkUpdateRedirectCtx(&job, &task, &epset, 0);
+
+  schChkUpdateRedirectCtx(&job, &task, NULL, 0);
+  task.plan = NULL;
+
+  schPushTaskToExecList(&job, &task);
+
+  job.execTasks = taosHashInit(1, taosGetDefaultHashFunction(TSDB_DATA_TYPE_UBIGINT), false, HASH_ENTRY_LOCK);
+  taosHashPut(job.execTasks, &task.taskId, sizeof(task.taskId), &task, POINTER_BYTES);
+  schPushTaskToExecList(&job, &task);
+  taosHashCleanup(job.execTasks);
+  job.execTasks = NULL;
+
+  bool needRetry = false;
+  task.timeoutUsec = SCH_MAX_TASK_TIMEOUT_USEC / 2 + 1;
+  task.retryTimes = 0;
+  task.maxRetryTimes = 0;
+  schTaskCheckSetRetry(&job, &task, TSDB_CODE_SCH_TIMEOUT_ERROR, &needRetry);
+
+  task.execId = 0;
+  task.retryTimes = 0;
+  task.maxRetryTimes = 100;
+  task.maxExecTimes = 1;
+  schTaskCheckSetRetry(&job, &task, TSDB_CODE_SCH_TIMEOUT_ERROR, &needRetry);
+
+
+  task.execId = 0;
+  task.retryTimes = 0;
+  task.maxRetryTimes = 100;
+  task.maxExecTimes = 100;
+  task.lastMsgType = TDMT_SCH_LINK_BROKEN;
+  schTaskCheckSetRetry(&job, &task, TSDB_CODE_SCH_TIMEOUT_ERROR, &needRetry);
+
+  schSetAddrsFromNodeList(&job, &task);
+
+  schSwitchTaskCandidateAddr(&job, &task);
+
+
+  task.candidateAddrs = taosArrayInit(SCHEDULE_DEFAULT_MAX_NODE_NUM, sizeof(SQueryNodeAddr));
+  SQueryNodeAddr addr = {0};
+  taosArrayPush(task.candidateAddrs, &addr);
+  taosArrayPush(task.candidateAddrs, &addr);
+  schMgmt.cfg.schPolicy = SCH_LOAD_SEQ;
+  task.candidateIdx = 1;
+  schSwitchTaskCandidateAddr(&job, &task);
+
+  schMgmt.cfg.schPolicy = SCH_RANDOM;
+  schSwitchTaskCandidateAddr(&job, &task);
+  taosArrayDestroy(task.candidateAddrs);
+  task.candidateAddrs = NULL;
+  memset(&schMgmt.cfg, 0, sizeof(schMgmt.cfg));
+  task.candidateIdx = 0;
+
+  schDropTaskOnExecNode(&job, &task);
+
+  schNotifyTaskOnExecNode(&job, &task, TASK_NOTIFY_FINISHED);
+
+  schLaunchRemoteTask(&job, &task);
+
+  SSchTaskCtx* pCtx = (SSchTaskCtx*)taosMemoryCalloc(1, sizeof(SSchTaskCtx));
+  pCtx->jobRid = -1;
+  schLaunchTaskImpl((void*)pCtx);
+
+  task.plan = &subplan;
+  subplan.subplanType = SUBPLAN_TYPE_SCAN;
+  job.attr.needFlowCtrl = true;
+  level.taskNum = 1000;
+  task.level = &level;
+  schLaunchTask(&job, &task);
+  task.plan = NULL;
+  task.level = NULL;
+  job.attr.needFlowCtrl = false;
+
+  SSchTimerParam param = {0};
+  param.rId = -1;
+  schHandleTimerEvent(&param, NULL);
+
+  job.execTasks = taosHashInit(1, taosGetDefaultHashFunction(TSDB_DATA_TYPE_UBIGINT), false, HASH_ENTRY_LOCK);
+  task.delayExecMs = 1;
+  schMgmt.timer = NULL;
+  schDelayLaunchTask(&job, &task);
+  task.delayExecMs = 0;
+  taosHashCleanup(job.execTasks);
+  job.execTasks = NULL;
+
+  job.fetchRes = (void*)0x1;
+  schLaunchFetchTask(&job);
+  job.fetchRes = NULL;
+
+  job.fetchTask = &task;
+  job.attr.localExec = true;
+  job.attr.queryJob = true;
+  subplan.subplanType = SUBPLAN_TYPE_MERGE;
+  task.plan = &subplan;
+  void* p = taosMemoryCalloc(1, 1024);
+  schMgmt.queryMgmt = p;
+  schLaunchFetchTask(&job);
+  memset(&job, 0, sizeof(job));
+  memset(&subplan, 0, sizeof(subplan));
+  task.plan = NULL;
+  taosMemoryFreeClear(schMgmt.queryMgmt);
+
+  // flow ctrl
+
+  job.flowCtrl = taosHashInit(1, taosGetDefaultHashFunction(TSDB_DATA_TYPE_BINARY), false, HASH_ENTRY_LOCK);
+  SEp sep = {0};
+  SSchFlowControl nctrl = {0};
+  nctrl.taskList = taosArrayInit(1, POINTER_BYTES);
+  taosHashPut(job.flowCtrl, &sep, sizeof(SEp), &nctrl, sizeof(nctrl));
+  schFreeFlowCtrl(&job);
+
+
+
   schMgmt.jobRef = -1;
 }