From e74ae6fc0502d8fc143f300c71d1049c14083f0c Mon Sep 17 00:00:00 2001
From: Haojun Liao <hjliao@taosdata.com>
Date: Fri, 28 Feb 2025 16:05:39 +0800
Subject: [PATCH 01/21] refactor(stream): not generated hb when trying to
 restart tasks.

---
 source/libs/stream/src/streamHb.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/source/libs/stream/src/streamHb.c b/source/libs/stream/src/streamHb.c
index 7c157bb05e..d973d7c05f 100644
--- a/source/libs/stream/src/streamHb.c
+++ b/source/libs/stream/src/streamHb.c
@@ -363,14 +363,18 @@ void streamMetaHbToMnode(void* param, void* tmrId) {
     pMeta->pHbInfo->hbStart = taosGetTimestampMs();
   }
 
-  streamMetaRLock(pMeta);
-  code = streamMetaSendHbHelper(pMeta);
-  if (code) {
-    stError("vgId:%d failed to send hmMsg to mnode, try again in 5s, code:%s", pMeta->vgId, tstrerror(code));
+  // NOTE: stream task in restart procedure. not generate the hb now, try to acquire the lock may cause stuck this timer.
+  if (pMeta->startInfo.startAllTasks == 1) {
+    stWarn("vgId:%d in restart all task procedure, not try to genbrate the hb now, try again in 5s", pMeta->vgId);
+  } else {
+    streamMetaRLock(pMeta);
+    code = streamMetaSendHbHelper(pMeta);
+    if (code) {
+      stError("vgId:%d failed to send hmMsg to mnode, try again in 5s, code:%s", pMeta->vgId, tstrerror(code));
+    }
+    streamMetaRUnLock(pMeta);
   }
 
-  streamMetaRUnLock(pMeta);
-
   streamTmrStart(streamMetaHbToMnode, META_HB_CHECK_INTERVAL, param, streamTimer, &pMeta->pHbInfo->hbTmr, pMeta->vgId,
                  "meta-hb-tmr");
 

From e1c4451f81fb3fae33197187b5f2761655ddd85d Mon Sep 17 00:00:00 2001
From: Haojun Liao <hjliao@taosdata.com>
Date: Fri, 28 Feb 2025 17:46:54 +0800
Subject: [PATCH 02/21] fix(stream): avoid lock in timer to avoid sync
 operation blocks.

---
 include/libs/stream/tstream.h            |  1 +
 source/dnode/vnode/src/tq/tqStreamTask.c | 11 +++++++----
 source/libs/stream/src/streamUtil.c      |  9 +++++++++
 3 files changed, 17 insertions(+), 4 deletions(-)

diff --git a/include/libs/stream/tstream.h b/include/libs/stream/tstream.h
index c4c0aaf742..de9466f736 100644
--- a/include/libs/stream/tstream.h
+++ b/include/libs/stream/tstream.h
@@ -788,6 +788,7 @@ void    streamMetaClearSetUpdateTaskListComplete(SStreamMeta* pMeta);
 bool    streamMetaInitUpdateTaskList(SStreamMeta* pMeta, int32_t transId);
 
 void    streamMetaRLock(SStreamMeta* pMeta);
+int32_t streamMetaTryRlock(SStreamMeta* pMeta);
 void    streamMetaRUnLock(SStreamMeta* pMeta);
 void    streamMetaWLock(SStreamMeta* pMeta);
 void    streamMetaWUnLock(SStreamMeta* pMeta);
diff --git a/source/dnode/vnode/src/tq/tqStreamTask.c b/source/dnode/vnode/src/tq/tqStreamTask.c
index b34ea78f64..77765e94c6 100644
--- a/source/dnode/vnode/src/tq/tqStreamTask.c
+++ b/source/dnode/vnode/src/tq/tqStreamTask.c
@@ -146,9 +146,13 @@ static void doStartScanWal(void* param, void* tmrId) {
     return;
   }
 
-  streamMetaRLock(pMeta);
-  numOfTasks = taosArrayGetSize(pMeta->pTaskList);
-  streamMetaRUnLock(pMeta);
+  code = streamMetaTryRlock(pMeta);
+  if (code == 0) {
+    numOfTasks = taosArrayGetSize(pMeta->pTaskList);
+    streamMetaRUnLock(pMeta);
+  } else {
+    numOfTasks = 0;
+  }
 
   if (numOfTasks == 0) {
     goto _end;
@@ -167,7 +171,6 @@ static void doStartScanWal(void* param, void* tmrId) {
   }
 
 _end:
-
   streamTmrStart(doStartScanWal, SCAN_WAL_IDLE_DURATION, pParam, pTimer, &pMeta->scanInfo.scanTimer, vgId, "scan-wal");
   tqDebug("vgId:%d scan-wal will start in %dms", vgId, SCAN_WAL_IDLE_DURATION*SCAN_WAL_WAIT_COUNT);
 
diff --git a/source/libs/stream/src/streamUtil.c b/source/libs/stream/src/streamUtil.c
index 4c481e6041..11e291c876 100644
--- a/source/libs/stream/src/streamUtil.c
+++ b/source/libs/stream/src/streamUtil.c
@@ -54,6 +54,15 @@ void streamMetaRUnLock(SStreamMeta* pMeta) {
   }
 }
 
+int32_t streamMetaTryRlock(SStreamMeta* pMeta) {
+  int32_t code = taosThreadRwlockTryRdlock(&pMeta->lock);
+  if (code) {
+    stError("vgId:%d try meta-rlock failed, code:%s", pMeta->vgId, tstrerror(code));
+  }
+
+  return code;
+}
+
 void streamMetaWLock(SStreamMeta* pMeta) {
   //  stTrace("vgId:%d meta-wlock", pMeta->vgId);
   int32_t code = taosThreadRwlockWrlock(&pMeta->lock);

From 6e3f738490823f284a6f5fb08ae6282efda33cea Mon Sep 17 00:00:00 2001
From: Haojun Liao <hjliao@taosdata.com>
Date: Fri, 28 Feb 2025 18:49:48 +0800
Subject: [PATCH 03/21] refactor(stream): try to lock during build stream hb.

---
 source/libs/stream/src/streamHb.c | 23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)

diff --git a/source/libs/stream/src/streamHb.c b/source/libs/stream/src/streamHb.c
index 312ed85764..ca5b6630fd 100644
--- a/source/libs/stream/src/streamHb.c
+++ b/source/libs/stream/src/streamHb.c
@@ -364,15 +364,22 @@ void streamMetaHbToMnode(void* param, void* tmrId) {
   }
 
   // NOTE: stream task in restart procedure. not generate the hb now, try to acquire the lock may cause stuck this timer.
-  if (pMeta->startInfo.startAllTasks == 1) {
-    stWarn("vgId:%d in restart all task procedure, not try to genbrate the hb now, try again in 5s", pMeta->vgId);
-  } else {
-    streamMetaRLock(pMeta);
-    code = streamMetaSendHbHelper(pMeta);
-    if (code) {
-      stError("vgId:%d failed to send hmMsg to mnode, try again in 5s, code:%s", pMeta->vgId, tstrerror(code));
+  int32_t count = 30;
+  bool    send = false;
+  while ((--count) >= 0) {
+    int32_t ret = streamMetaTryRlock(pMeta);
+    if (ret != 0) {
+      taosMsleep(10);
+    } else {
+      send = true;
+      code = streamMetaSendHbHelper(pMeta);
+      streamMetaRUnLock(pMeta);
+      break;
     }
-    streamMetaRUnLock(pMeta);
+  }
+
+  if (!send) {
+    stError("vgId:%d failed to send hmMsg to mnode, retry again in 5s, code:%s", pMeta->vgId, tstrerror(code));
   }
 
   streamTmrStart(streamMetaHbToMnode, META_HB_CHECK_INTERVAL, param, streamTimer, &pMeta->pHbInfo->hbTmr, pMeta->vgId,

From 8c1d28219ae4ce37b0615ea93e37ec83a22352dd Mon Sep 17 00:00:00 2001
From: Haojun Liao <hjliao@taosdata.com>
Date: Fri, 28 Feb 2025 19:38:15 +0800
Subject: [PATCH 04/21] fix(stream): fix race condition

---
 source/libs/stream/src/streamCheckpoint.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/source/libs/stream/src/streamCheckpoint.c b/source/libs/stream/src/streamCheckpoint.c
index 387010d0a6..eb8f2c741a 100644
--- a/source/libs/stream/src/streamCheckpoint.c
+++ b/source/libs/stream/src/streamCheckpoint.c
@@ -1585,10 +1585,13 @@ int32_t streamTaskSendNegotiateChkptIdMsg(SStreamTask* pTask) {
     pTask->pBackend = NULL;
   }
 
+  streamMetaWLock(pTask->pMeta);
   if (pTask->exec.pExecutor != NULL) {
     qDestroyTask(pTask->exec.pExecutor);
     pTask->exec.pExecutor = NULL;
   }
+  streamMetaWUnLock(pTask->pMeta);
+
   return 0;
 }
 

From 86e8675fae3f119638300a4be483aee7b3f40e41 Mon Sep 17 00:00:00 2001
From: dmchen <cademfly@hotmail.com>
Date: Wed, 26 Feb 2025 16:55:06 +0800
Subject: [PATCH 05/21] fix/remove-sync-heartbeat-lock

---
 source/libs/sync/inc/syncPipeline.h |  1 +
 source/libs/sync/src/syncPipeline.c | 16 +++++++++++++---
 2 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/source/libs/sync/inc/syncPipeline.h b/source/libs/sync/inc/syncPipeline.h
index eeb24d2f16..147f8a67ae 100644
--- a/source/libs/sync/inc/syncPipeline.h
+++ b/source/libs/sync/inc/syncPipeline.h
@@ -40,6 +40,7 @@ typedef struct SSyncLogReplMgr {
   int32_t       retryBackoff;
   int32_t       peerId;
   int32_t       sendCount;
+  TdThreadMutex mutex;
 } SSyncLogReplMgr;
 
 typedef struct SSyncLogBufEntry {
diff --git a/source/libs/sync/src/syncPipeline.c b/source/libs/sync/src/syncPipeline.c
index 18252db9ee..ceaba1e643 100644
--- a/source/libs/sync/src/syncPipeline.c
+++ b/source/libs/sync/src/syncPipeline.c
@@ -1140,26 +1140,29 @@ int32_t syncLogReplRecover(SSyncLogReplMgr* pMgr, SSyncNode* pNode, SyncAppendEn
 
 int32_t syncLogReplProcessHeartbeatReply(SSyncLogReplMgr* pMgr, SSyncNode* pNode, SyncHeartbeatReply* pMsg) {
   SSyncLogBuffer* pBuf = pNode->pLogBuf;
-  (void)taosThreadMutexLock(&pBuf->mutex);
+  (void)taosThreadMutexLock(&pMgr->mutex);
   if (pMsg->startTime != 0 && pMsg->startTime != pMgr->peerStartTime) {
     sInfo("vgId:%d, reset sync log repl in heartbeat. peer:%" PRIx64 ", start time:%" PRId64 ", old:%" PRId64 "",
           pNode->vgId, pMsg->srcId.addr, pMsg->startTime, pMgr->peerStartTime);
     syncLogReplReset(pMgr);
     pMgr->peerStartTime = pMsg->startTime;
   }
-  (void)taosThreadMutexUnlock(&pBuf->mutex);
+  (void)taosThreadMutexUnlock(&pMgr->mutex);
   return 0;
 }
 
 int32_t syncLogReplProcessReply(SSyncLogReplMgr* pMgr, SSyncNode* pNode, SyncAppendEntriesReply* pMsg) {
   SSyncLogBuffer* pBuf = pNode->pLogBuf;
-  (void)taosThreadMutexLock(&pBuf->mutex);
+  (void)taosThreadMutexLock(&pMgr->mutex);
   if (pMsg->startTime != pMgr->peerStartTime) {
     sInfo("vgId:%d, reset sync log repl in appendlog reply. peer:%" PRIx64 ", start time:%" PRId64 ", old:%" PRId64,
           pNode->vgId, pMsg->srcId.addr, pMsg->startTime, pMgr->peerStartTime);
     syncLogReplReset(pMgr);
     pMgr->peerStartTime = pMsg->startTime;
   }
+  (void)taosThreadMutexUnlock(&pMgr->mutex);
+
+  (void)taosThreadMutexLock(&pBuf->mutex);
 
   int32_t code = 0;
   if (pMgr->restored) {
@@ -1324,6 +1327,12 @@ SSyncLogReplMgr* syncLogReplCreate() {
     return NULL;
   }
 
+  int32_t code = taosThreadMutexInit(&pMgr->mutex, NULL);
+  if (code) {
+    terrno = code;
+    return NULL;
+  }
+
   return pMgr;
 }
 
@@ -1331,6 +1340,7 @@ void syncLogReplDestroy(SSyncLogReplMgr* pMgr) {
   if (pMgr == NULL) {
     return;
   }
+  taosThreadMutexDestroy(&pMgr->mutex);
   taosMemoryFree(pMgr);
   return;
 }

From 21e3ec5a35b449b7503d4686e8d1777562c52a23 Mon Sep 17 00:00:00 2001
From: dmchen <cademfly@hotmail.com>
Date: Wed, 26 Feb 2025 18:41:17 +0800
Subject: [PATCH 06/21] fix/remove-sync-heartbeat-lock-fix-check

---
 source/libs/sync/src/syncPipeline.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/source/libs/sync/src/syncPipeline.c b/source/libs/sync/src/syncPipeline.c
index ceaba1e643..8d81a03344 100644
--- a/source/libs/sync/src/syncPipeline.c
+++ b/source/libs/sync/src/syncPipeline.c
@@ -1340,7 +1340,7 @@ void syncLogReplDestroy(SSyncLogReplMgr* pMgr) {
   if (pMgr == NULL) {
     return;
   }
-  taosThreadMutexDestroy(&pMgr->mutex);
+  (void)taosThreadMutexDestroy(&pMgr->mutex);
   taosMemoryFree(pMgr);
   return;
 }

From 48787ba1aace0a14771c09af56a60070ce627ae0 Mon Sep 17 00:00:00 2001
From: dmchen <cademfly@hotmail.com>
Date: Thu, 27 Feb 2025 11:02:18 +0800
Subject: [PATCH 07/21] fix/move-status-to-new-queue

---
 include/common/tmsgcb.h                       |  1 +
 source/dnode/mgmt/mgmt_dnode/src/dmHandle.c   |  7 ++++---
 source/dnode/mgmt/mgmt_mnode/inc/mmInt.h      |  2 ++
 source/dnode/mgmt/mgmt_mnode/src/mmHandle.c   |  2 +-
 source/dnode/mgmt/mgmt_mnode/src/mmWorker.c   | 20 +++++++++++++++++++
 source/dnode/mgmt/mgmt_qnode/src/qmWorker.c   |  1 +
 source/dnode/mnode/impl/test/trans/trans2.cpp |  3 ++-
 source/libs/sync/inc/syncUtil.h               |  4 ++--
 8 files changed, 33 insertions(+), 7 deletions(-)

diff --git a/include/common/tmsgcb.h b/include/common/tmsgcb.h
index 2847f4278a..cdff5aaba2 100644
--- a/include/common/tmsgcb.h
+++ b/include/common/tmsgcb.h
@@ -31,6 +31,7 @@ typedef enum {
   QUERY_QUEUE,
   FETCH_QUEUE,
   READ_QUEUE,
+  STATUS_QUEUE,
   WRITE_QUEUE,
   APPLY_QUEUE,
   SYNC_QUEUE,
diff --git a/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c b/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
index 637713d2f9..ef0106cdbb 100644
--- a/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
+++ b/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
@@ -267,7 +267,7 @@ void dmSendStatusReq(SDnodeMgmt *pMgmt) {
   code =
       rpcSendRecvWithTimeout(pMgmt->msgCb.statusRpc, &epSet, &rpcMsg, &rpcRsp, &epUpdated, tsStatusInterval * 5 * 1000);
   if (code != 0) {
-    dError("failed to send status req since %s", tstrerror(code));
+    dError("failed to SendRecv with timeout %d status req since %s", tsStatusInterval * 5 * 1000, tstrerror(code));
     return;
   }
 
@@ -275,7 +275,8 @@ void dmSendStatusReq(SDnodeMgmt *pMgmt) {
     dmRotateMnodeEpSet(pMgmt->pData);
     char tbuf[512];
     dmEpSetToStr(tbuf, sizeof(tbuf), &epSet);
-    dError("failed to send status req since %s, epSet:%s, inUse:%d", tstrerror(rpcRsp.code), tbuf, epSet.inUse);
+    dInfo("Rotate mnode ep set since failed to SendRecv status req %s, epSet:%s, inUse:%d", tstrerror(rpcRsp.code),
+          tbuf, epSet.inUse);
   } else {
     if (epUpdated == 1) {
       dmSetMnodeEpSet(pMgmt->pData, &epSet);
@@ -403,7 +404,7 @@ void dmSendConfigReq(SDnodeMgmt *pMgmt) {
   code =
       rpcSendRecvWithTimeout(pMgmt->msgCb.statusRpc, &epSet, &rpcMsg, &rpcRsp, &epUpdated, tsStatusInterval * 5 * 1000);
   if (code != 0) {
-    dError("failed to send status req since %s", tstrerror(code));
+    dError("failed to SendRecv config req with timeout %d since %s", tsStatusInterval * 5 * 1000, tstrerror(code));
     return;
   }
   if (rpcRsp.code != 0) {
diff --git a/source/dnode/mgmt/mgmt_mnode/inc/mmInt.h b/source/dnode/mgmt/mgmt_mnode/inc/mmInt.h
index bc2977fe31..2d7aa35c13 100644
--- a/source/dnode/mgmt/mgmt_mnode/inc/mmInt.h
+++ b/source/dnode/mgmt/mgmt_mnode/inc/mmInt.h
@@ -32,6 +32,7 @@ typedef struct SMnodeMgmt {
   SSingleWorker  queryWorker;
   SSingleWorker  fetchWorker;
   SSingleWorker  readWorker;
+  SSingleWorker  statusWorker;
   SSingleWorker  writeWorker;
   SSingleWorker  arbWorker;
   SSingleWorker  syncWorker;
@@ -58,6 +59,7 @@ int32_t mmPutMsgToArbQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg);
 int32_t mmPutMsgToSyncQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg);
 int32_t mmPutMsgToSyncRdQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg);
 int32_t mmPutMsgToReadQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg);
+int32_t mmPutMsgToStatusQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg);
 int32_t mmPutMsgToQueryQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg);
 int32_t mmPutMsgToFetchQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg);
 int32_t mmPutMsgToQueue(SMnodeMgmt *pMgmt, EQueueType qtype, SRpcMsg *pRpc);
diff --git a/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c b/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c
index 80ef0d31de..0031042497 100644
--- a/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c
+++ b/source/dnode/mgmt/mgmt_mnode/src/mmHandle.c
@@ -202,7 +202,7 @@ SArray *mmGetMsgHandles() {
   if (dmSetMgmtHandle(pArray, TDMT_MND_KILL_QUERY, mmPutMsgToWriteQueue, 0) == NULL) goto _OVER;
   if (dmSetMgmtHandle(pArray, TDMT_MND_KILL_CONN, mmPutMsgToWriteQueue, 0) == NULL) goto _OVER;
   if (dmSetMgmtHandle(pArray, TDMT_MND_HEARTBEAT, mmPutMsgToReadQueue, 0) == NULL) goto _OVER;
-  if (dmSetMgmtHandle(pArray, TDMT_MND_STATUS, mmPutMsgToReadQueue, 0) == NULL) goto _OVER;
+  if (dmSetMgmtHandle(pArray, TDMT_MND_STATUS, mmPutMsgToStatusQueue, 0) == NULL) goto _OVER;
   if (dmSetMgmtHandle(pArray, TDMT_MND_CONFIG, mmPutMsgToWriteQueue, 0) == NULL) goto _OVER;
   if (dmSetMgmtHandle(pArray, TDMT_MND_NOTIFY, mmPutMsgToReadQueue, 0) == NULL) goto _OVER;
   if (dmSetMgmtHandle(pArray, TDMT_MND_SYSTABLE_RETRIEVE, mmPutMsgToReadQueue, 0) == NULL) goto _OVER;
diff --git a/source/dnode/mgmt/mgmt_mnode/src/mmWorker.c b/source/dnode/mgmt/mgmt_mnode/src/mmWorker.c
index cfdc09f29e..cb82ddae2e 100644
--- a/source/dnode/mgmt/mgmt_mnode/src/mmWorker.c
+++ b/source/dnode/mgmt/mgmt_mnode/src/mmWorker.c
@@ -133,6 +133,10 @@ int32_t mmPutMsgToReadQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg) {
   return mmPutMsgToWorker(pMgmt, &pMgmt->readWorker, pMsg);
 }
 
+int32_t mmPutMsgToStatusQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg) {
+  return mmPutMsgToWorker(pMgmt, &pMgmt->statusWorker, pMsg);
+}
+
 int32_t mmPutMsgToQueryQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg) {
   int32_t code = 0;
   if (NULL == pMgmt->pMnode) {
@@ -172,6 +176,9 @@ int32_t mmPutMsgToQueue(SMnodeMgmt *pMgmt, EQueueType qtype, SRpcMsg *pRpc) {
     case READ_QUEUE:
       pWorker = &pMgmt->readWorker;
       break;
+    case STATUS_QUEUE:
+      pWorker = &pMgmt->statusWorker;
+      break;
     case ARB_QUEUE:
       pWorker = &pMgmt->arbWorker;
       break;
@@ -246,6 +253,18 @@ int32_t mmStartWorker(SMnodeMgmt *pMgmt) {
     return code;
   }
 
+  SSingleWorkerCfg stautsCfg = {
+      .min = 1,
+      .max = 1,
+      .name = "mnode-status",
+      .fp = (FItem)mmProcessRpcMsg,
+      .param = pMgmt,
+  };
+  if ((code = tSingleWorkerInit(&pMgmt->statusWorker, &stautsCfg)) != 0) {
+    dError("failed to start mnode-status worker since %s", tstrerror(code));
+    return code;
+  }
+
   SSingleWorkerCfg wCfg = {
       .min = 1,
       .max = 1,
@@ -304,6 +323,7 @@ void mmStopWorker(SMnodeMgmt *pMgmt) {
   tSingleWorkerCleanup(&pMgmt->queryWorker);
   tSingleWorkerCleanup(&pMgmt->fetchWorker);
   tSingleWorkerCleanup(&pMgmt->readWorker);
+  tSingleWorkerCleanup(&pMgmt->statusWorker);
   tSingleWorkerCleanup(&pMgmt->writeWorker);
   tSingleWorkerCleanup(&pMgmt->arbWorker);
   tSingleWorkerCleanup(&pMgmt->syncWorker);
diff --git a/source/dnode/mgmt/mgmt_qnode/src/qmWorker.c b/source/dnode/mgmt/mgmt_qnode/src/qmWorker.c
index ab5b70079a..950da46a40 100644
--- a/source/dnode/mgmt/mgmt_qnode/src/qmWorker.c
+++ b/source/dnode/mgmt/mgmt_qnode/src/qmWorker.c
@@ -71,6 +71,7 @@ int32_t qmPutRpcMsgToQueue(SQnodeMgmt *pMgmt, EQueueType qtype, SRpcMsg *pRpc) {
       code = taosWriteQitem(pMgmt->queryWorker.queue, pMsg);
       return code;
     case READ_QUEUE:
+    case STATUS_QUEUE:
     case FETCH_QUEUE:
       dTrace("msg:%p, is created and will put into qnode-fetch queue, len:%d", pMsg, pRpc->contLen);
       code = taosWriteQitem(pMgmt->fetchWorker.queue, pMsg);
diff --git a/source/dnode/mnode/impl/test/trans/trans2.cpp b/source/dnode/mnode/impl/test/trans/trans2.cpp
index b73f07c778..e520fce111 100644
--- a/source/dnode/mnode/impl/test/trans/trans2.cpp
+++ b/source/dnode/mnode/impl/test/trans/trans2.cpp
@@ -65,7 +65,8 @@ class MndTestTrans2 : public ::testing::Test {
     msgCb.sendRspFp = sendRsp;
     msgCb.queueFps[SYNC_QUEUE] = putToQueue;
     msgCb.queueFps[WRITE_QUEUE] = putToQueue;
-     msgCb.queueFps[READ_QUEUE] = putToQueue;
+    msgCb.queueFps[READ_QUEUE] = putToQueue;
+    msgCb.queueFps[STATUS_QUEUE] = putToQueue;
     msgCb.mgmt = (SMgmtWrapper *)(&msgCb);  // hack
     tmsgSetDefault(&msgCb);
 
diff --git a/source/libs/sync/inc/syncUtil.h b/source/libs/sync/inc/syncUtil.h
index 7b71491f47..e7110c0964 100644
--- a/source/libs/sync/inc/syncUtil.h
+++ b/source/libs/sync/inc/syncUtil.h
@@ -50,8 +50,8 @@ extern "C" {
 #define sNError(pNode, ...)  if (sDebugFlag & DEBUG_ERROR) { syncPrintNodeLog("SYN ERROR ", DEBUG_ERROR, 255, true,       pNode,  __VA_ARGS__); }
 #define sNWarn(pNode, ...)   if (sDebugFlag & DEBUG_WARN)  { syncPrintNodeLog("SYN WARN ",  DEBUG_WARN,  255, true,        pNode,  __VA_ARGS__); }
 #define sNInfo(pNode, ...)   if (sDebugFlag & DEBUG_INFO)  { syncPrintNodeLog("SYN ",       DEBUG_INFO,  255, true,        pNode, __VA_ARGS__); }
-#define sNDebug(pNode, ...)  if (sDebugFlag & DEBUG_DEBUG) { syncPrintNodeLog("SYN ",       DEBUG_DEBUG, sDebugFlag, false, pNode,  __VA_ARGS__); }
-#define sNTrace(pNode, ...)  if (sDebugFlag & DEBUG_TRACE) { syncPrintNodeLog("SYN ",       DEBUG_TRACE, sDebugFlag, false, pNode,  __VA_ARGS__); }
+#define sNDebug(pNode, ...)  if (sDebugFlag & DEBUG_DEBUG) { syncPrintNodeLog("SYN ",       DEBUG_DEBUG, sDebugFlag, true, pNode,  __VA_ARGS__); }
+#define sNTrace(pNode, ...)  if (sDebugFlag & DEBUG_TRACE) { syncPrintNodeLog("SYN ",       DEBUG_TRACE, sDebugFlag, true, pNode,  __VA_ARGS__); }
 
 #define sSFatal(pSender, ...)  if (sDebugFlag & DEBUG_FATAL) { syncPrintSnapshotSenderLog("SYN FATAL ", DEBUG_FATAL, 255,        pSender, __VA_ARGS__); }
 #define sSError(pSender, ...)  if (sDebugFlag & DEBUG_ERROR) { syncPrintSnapshotSenderLog("SYN ERROR ", DEBUG_ERROR, 255,        pSender, __VA_ARGS__); }

From 774c5121483567936240a342d28f80eb50f69dd4 Mon Sep 17 00:00:00 2001
From: dmchen <cademfly@hotmail.com>
Date: Thu, 27 Feb 2025 15:22:33 +0800
Subject: [PATCH 08/21] fix:add-status-timestamp

---
 include/common/tmsg.h                       |  1 +
 include/common/ttime.h                      |  2 +-
 source/common/src/msg/tmsg.c                |  6 +++++
 source/common/src/ttime.c                   | 28 +++++++++++++++++++++
 source/dnode/mgmt/mgmt_dnode/src/dmHandle.c |  1 +
 source/dnode/mnode/impl/src/mndDnode.c      |  8 ++++--
 source/libs/sync/src/syncUtil.c             |  2 +-
 7 files changed, 44 insertions(+), 4 deletions(-)

diff --git a/include/common/tmsg.h b/include/common/tmsg.h
index 0b6a8b3f1b..38fdc52513 100644
--- a/include/common/tmsg.h
+++ b/include/common/tmsg.h
@@ -1839,6 +1839,7 @@ typedef struct {
   int32_t     statusSeq;
   int64_t     ipWhiteVer;
   int64_t     analVer;
+  int64_t     timestamp;
 } SStatusReq;
 
 int32_t tSerializeSStatusReq(void* buf, int32_t bufLen, SStatusReq* pReq);
diff --git a/include/common/ttime.h b/include/common/ttime.h
index 47875752d5..77a3041b2b 100644
--- a/include/common/ttime.h
+++ b/include/common/ttime.h
@@ -82,7 +82,7 @@ int32_t convertStringToTimestamp(int16_t type, char* inputData, int64_t timePrec
 int32_t getDuration(int64_t val, char unit, int64_t* result, int32_t timePrecision);
 
 int32_t taosFormatUtcTime(char* buf, int32_t bufLen, int64_t ts, int32_t precision);
-
+char*   formatTimestampLocal(char* buf, int64_t val, int precision);
 struct STm {
   struct tm tm;
   int64_t   fsec;  // in NANOSECOND
diff --git a/source/common/src/msg/tmsg.c b/source/common/src/msg/tmsg.c
index f1ebcb3f42..cc635885c0 100644
--- a/source/common/src/msg/tmsg.c
+++ b/source/common/src/msg/tmsg.c
@@ -1471,6 +1471,8 @@ int32_t tSerializeSStatusReq(void *buf, int32_t bufLen, SStatusReq *pReq) {
   TAOS_CHECK_EXIT(tEncodeI64(&encoder, pReq->analVer));
   TAOS_CHECK_EXIT(tSerializeSMonitorParas(&encoder, &pReq->clusterCfg.monitorParas));
 
+  TAOS_CHECK_EXIT(tEncodeI64(&encoder, pReq->timestamp));
+
   tEndEncode(&encoder);
 
 _exit:
@@ -1600,6 +1602,10 @@ int32_t tDeserializeSStatusReq(void *buf, int32_t bufLen, SStatusReq *pReq) {
     TAOS_CHECK_EXIT(tDeserializeSMonitorParas(&decoder, &pReq->clusterCfg.monitorParas));
   }
 
+  if (!tDecodeIsEnd(&decoder)) {
+    TAOS_CHECK_EXIT(tDecodeI64(&decoder, &pReq->timestamp));
+  }
+
   tEndDecode(&decoder);
 
 _exit:
diff --git a/source/common/src/ttime.c b/source/common/src/ttime.c
index 9746fea034..22ab106bb3 100644
--- a/source/common/src/ttime.c
+++ b/source/common/src/ttime.c
@@ -1000,6 +1000,34 @@ int32_t taosFormatUtcTime(char* buf, int32_t bufLen, int64_t t, int32_t precisio
   TAOS_RETURN(TSDB_CODE_SUCCESS);
 }
 
+char* formatTimestampLocal(char* buf, int64_t val, int precision) {
+  time_t tt;
+  if (precision == TSDB_TIME_PRECISION_MICRO) {
+    tt = (time_t)(val / 1000000);
+  }
+  if (precision == TSDB_TIME_PRECISION_NANO) {
+    tt = (time_t)(val / 1000000000);
+  } else {
+    tt = (time_t)(val / 1000);
+  }
+
+  struct tm tm;
+  if (taosLocalTime(&tt, &tm, NULL, 0, NULL) == NULL) {
+    return NULL;
+  }
+  size_t pos = taosStrfTime(buf, 32, "%Y-%m-%d %H:%M:%S", &tm);
+
+  if (precision == TSDB_TIME_PRECISION_MICRO) {
+    sprintf(buf + pos, ".%06d", (int)(val % 1000000));
+  } else if (precision == TSDB_TIME_PRECISION_NANO) {
+    sprintf(buf + pos, ".%09d", (int)(val % 1000000000));
+  } else {
+    sprintf(buf + pos, ".%03d", (int)(val % 1000));
+  }
+
+  return buf;
+}
+
 int32_t taosTs2Tm(int64_t ts, int32_t precision, struct STm* tm, timezone_t tz) {
   tm->fsec = ts % TICK_PER_SECOND[precision] * (TICK_PER_SECOND[TSDB_TIME_PRECISION_NANO] / TICK_PER_SECOND[precision]);
   time_t t = ts / TICK_PER_SECOND[precision];
diff --git a/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c b/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
index ef0106cdbb..653a54592a 100644
--- a/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
+++ b/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
@@ -167,6 +167,7 @@ static void dmProcessStatusRsp(SDnodeMgmt *pMgmt, SRpcMsg *pRsp) {
 void dmSendStatusReq(SDnodeMgmt *pMgmt) {
   int32_t    code = 0;
   SStatusReq req = {0};
+  req.timestamp = taosGetTimestampMs();
 
   dDebug("send status req to mnode, statusSeq:%d, begin to mgnt lock", pMgmt->statusSeq);
   (void)taosThreadRwlockRdlock(&pMgmt->pData->lock);
diff --git a/source/dnode/mnode/impl/src/mndDnode.c b/source/dnode/mnode/impl/src/mndDnode.c
index ca119191eb..dac38dc4db 100644
--- a/source/dnode/mnode/impl/src/mndDnode.c
+++ b/source/dnode/mnode/impl/src/mndDnode.c
@@ -756,8 +756,12 @@ static int32_t mndProcessStatusReq(SRpcMsg *pReq) {
   bool    needCheck = !online || dnodeChanged || reboot || supportVnodesChanged || analVerChanged ||
                    pMnode->ipWhiteVer != statusReq.ipWhiteVer || encryptKeyChanged || enableWhiteListChanged;
   const STraceId *trace = &pReq->info.traceId;
-  mGTrace("dnode:%d, status received, accessTimes:%d check:%d online:%d reboot:%d changed:%d statusSeq:%d", pDnode->id,
-          pDnode->accessTimes, needCheck, online, reboot, dnodeChanged, statusReq.statusSeq);
+  char            timestamp[TD_TIME_STR_LEN] = {0};
+  if (mDebugFlag & DEBUG_TRACE) (void)formatTimestampLocal(timestamp, statusReq.timestamp, TSDB_TIME_PRECISION_MILLI);
+  mGTrace(
+      "dnode:%d, status received, accessTimes:%d check:%d online:%d reboot:%d changed:%d statusSeq:%d "
+      "timestamp:%s",
+      pDnode->id, pDnode->accessTimes, needCheck, online, reboot, dnodeChanged, statusReq.statusSeq, timestamp);
 
   if (reboot) {
     tsGrantHBInterval = GRANT_HEART_BEAT_MIN;
diff --git a/source/libs/sync/src/syncUtil.c b/source/libs/sync/src/syncUtil.c
index 38fd660cfd..5a2571a291 100644
--- a/source/libs/sync/src/syncUtil.c
+++ b/source/libs/sync/src/syncUtil.c
@@ -116,7 +116,7 @@ static void syncPrintTime(bool formatTime, int32_t* len, int64_t tsMs, int32_t i
   if (formatTime) {
     char pBuf[TD_TIME_STR_LEN] = {0};
     if (tsMs > 0) {
-      if (taosFormatUtcTime(pBuf, TD_TIME_STR_LEN, tsMs, TSDB_TIME_PRECISION_MILLI) != 0) {
+      if (formatTimestampLocal(pBuf, tsMs, TSDB_TIME_PRECISION_MILLI) == NULL) {
         pBuf[0] = '\0';
       }
     }

From 3359483836878d82fd7033df7d2874c0d3050b8e Mon Sep 17 00:00:00 2001
From: dmchen <cademfly@hotmail.com>
Date: Thu, 27 Feb 2025 19:05:06 +0800
Subject: [PATCH 09/21] fix:decouple get mload

---
 source/dnode/mgmt/mgmt_dnode/src/dmHandle.c | 23 ++++++++++++++-------
 source/dnode/mnode/impl/src/mndMain.c       |  1 +
 2 files changed, 17 insertions(+), 7 deletions(-)

diff --git a/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c b/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
index 653a54592a..49f634443f 100644
--- a/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
+++ b/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
@@ -25,6 +25,7 @@
 extern SConfig *tsCfg;
 
 SMonVloadInfo tsVinfo = {0};
+SMnodeLoad    tsMLoad = {0};
 
 static void dmUpdateDnodeCfg(SDnodeMgmt *pMgmt, SDnodeCfg *pCfg) {
   int32_t code = 0;
@@ -208,23 +209,22 @@ void dmSendStatusReq(SDnodeMgmt *pMgmt) {
   memcpy(req.clusterCfg.charset, tsCharset, TD_LOCALE_LEN);
   (void)taosThreadRwlockUnlock(&pMgmt->pData->lock);
 
-  dDebug("send status req to mnode, statusSeq:%d, begin to get vnode loads", pMgmt->statusSeq);
+  dDebug("send status req to mnode, statusSeq:%d, begin to get vnode and loads", pMgmt->statusSeq);
   if (taosThreadMutexLock(&pMgmt->pData->statusInfolock) != 0) {
     dError("failed to lock status info lock");
     return;
   }
+
   req.pVloads = tsVinfo.pVloads;
   tsVinfo.pVloads = NULL;
+
+  req.mload = tsMLoad;
+
   if (taosThreadMutexUnlock(&pMgmt->pData->statusInfolock) != 0) {
     dError("failed to unlock status info lock");
     return;
   }
 
-  dDebug("send status req to mnode, statusSeq:%d, begin to get mnode loads", pMgmt->statusSeq);
-  SMonMloadInfo minfo = {0};
-  (*pMgmt->getMnodeLoadsFp)(&minfo);
-  req.mload = minfo.load;
-
   dDebug("send status req to mnode, statusSeq:%d, begin to get qnode loads", pMgmt->statusSeq);
   (*pMgmt->getQnodeLoadsFp)(&req.qload);
 
@@ -418,12 +418,18 @@ void dmSendConfigReq(SDnodeMgmt *pMgmt) {
 void dmUpdateStatusInfo(SDnodeMgmt *pMgmt) {
   SMonVloadInfo vinfo = {0};
   dDebug("begin to get vnode loads");
-  (*pMgmt->getVnodeLoadsFp)(&vinfo);
+  (*pMgmt->getVnodeLoadsFp)(&vinfo);  // dmGetVnodeLoads
+
+  dDebug("begin to get mnode loads");
+  SMonMloadInfo minfo = {0};
+  (*pMgmt->getMnodeLoadsFp)(&minfo);  // dmGetMnodeLoads
+
   dDebug("begin to lock status info");
   if (taosThreadMutexLock(&pMgmt->pData->statusInfolock) != 0) {
     dError("failed to lock status info lock");
     return;
   }
+
   if (tsVinfo.pVloads == NULL) {
     tsVinfo.pVloads = vinfo.pVloads;
     vinfo.pVloads = NULL;
@@ -431,6 +437,9 @@ void dmUpdateStatusInfo(SDnodeMgmt *pMgmt) {
     taosArrayDestroy(vinfo.pVloads);
     vinfo.pVloads = NULL;
   }
+
+  tsMLoad = minfo.load;
+
   if (taosThreadMutexUnlock(&pMgmt->pData->statusInfolock) != 0) {
     dError("failed to unlock status info lock");
     return;
diff --git a/source/dnode/mnode/impl/src/mndMain.c b/source/dnode/mnode/impl/src/mndMain.c
index b241af5adb..09314c9e63 100644
--- a/source/dnode/mnode/impl/src/mndMain.c
+++ b/source/dnode/mnode/impl/src/mndMain.c
@@ -1174,6 +1174,7 @@ int32_t mndGetMonitorInfo(SMnode *pMnode, SMonClusterInfo *pClusterInfo, SMonVgr
 }
 
 int32_t mndGetLoad(SMnode *pMnode, SMnodeLoad *pLoad) {
+  mTrace("mnode get load");
   SSyncState state = syncGetState(pMnode->syncMgmt.sync);
   pLoad->syncState = state.state;
   pLoad->syncRestore = state.restored;

From 6d2925b832c73ce4d15c2a330489a02f921a04d2 Mon Sep 17 00:00:00 2001
From: dmchen <cademfly@hotmail.com>
Date: Fri, 28 Feb 2025 11:52:27 +0800
Subject: [PATCH 10/21] fix/remove-get-queue-size-when-print-heartbeat-log

---
 source/dnode/mgmt/mgmt_vnode/inc/vmInt.h    |  2 +-
 source/dnode/mgmt/mgmt_vnode/src/vmFile.c   | 18 +++++------
 source/dnode/mgmt/mgmt_vnode/src/vmHandle.c | 12 ++++----
 source/dnode/mgmt/mgmt_vnode/src/vmInt.c    | 34 ++++++++++-----------
 source/libs/sync/src/syncUtil.c             | 30 +++++++++++-------
 source/libs/transport/src/tmsgcb.c          |  2 +-
 6 files changed, 53 insertions(+), 45 deletions(-)

diff --git a/source/dnode/mgmt/mgmt_vnode/inc/vmInt.h b/source/dnode/mgmt/mgmt_vnode/inc/vmInt.h
index 9b4c11d6ae..e33730130d 100644
--- a/source/dnode/mgmt/mgmt_vnode/inc/vmInt.h
+++ b/source/dnode/mgmt/mgmt_vnode/inc/vmInt.h
@@ -40,7 +40,7 @@ typedef struct SVnodeMgmt {
   SHashObj             *runngingHash;
   SHashObj             *closedHash;
   SHashObj             *creatingHash;
-  TdThreadRwlock        lock;
+  TdThreadRwlock        hashLock;
   TdThreadMutex         mutex;
   SVnodesStat           state;
   STfs                 *pTfs;
diff --git a/source/dnode/mgmt/mgmt_vnode/src/vmFile.c b/source/dnode/mgmt/mgmt_vnode/src/vmFile.c
index dbef048c23..cb14155b1c 100644
--- a/source/dnode/mgmt/mgmt_vnode/src/vmFile.c
+++ b/source/dnode/mgmt/mgmt_vnode/src/vmFile.c
@@ -20,7 +20,7 @@
 #define MAX_CONTENT_LEN 2 * 1024 * 1024
 
 int32_t vmGetAllVnodeListFromHash(SVnodeMgmt *pMgmt, int32_t *numOfVnodes, SVnodeObj ***ppVnodes) {
-  (void)taosThreadRwlockRdlock(&pMgmt->lock);
+  (void)taosThreadRwlockRdlock(&pMgmt->hashLock);
 
   int32_t num = 0;
   int32_t size = taosHashGetSize(pMgmt->runngingHash);
@@ -28,7 +28,7 @@ int32_t vmGetAllVnodeListFromHash(SVnodeMgmt *pMgmt, int32_t *numOfVnodes, SVnod
   size += closedSize;
   SVnodeObj **pVnodes = taosMemoryCalloc(size, sizeof(SVnodeObj *));
   if (pVnodes == NULL) {
-    (void)taosThreadRwlockUnlock(&pMgmt->lock);
+    (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
     return terrno;
   }
 
@@ -60,7 +60,7 @@ int32_t vmGetAllVnodeListFromHash(SVnodeMgmt *pMgmt, int32_t *numOfVnodes, SVnod
     }
   }
 
-  (void)taosThreadRwlockUnlock(&pMgmt->lock);
+  (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
   *numOfVnodes = num;
   *ppVnodes = pVnodes;
 
@@ -68,7 +68,7 @@ int32_t vmGetAllVnodeListFromHash(SVnodeMgmt *pMgmt, int32_t *numOfVnodes, SVnod
 }
 
 int32_t vmGetAllVnodeListFromHashWithCreating(SVnodeMgmt *pMgmt, int32_t *numOfVnodes, SVnodeObj ***ppVnodes) {
-  (void)taosThreadRwlockRdlock(&pMgmt->lock);
+  (void)taosThreadRwlockRdlock(&pMgmt->hashLock);
 
   int32_t num = 0;
   int32_t size = taosHashGetSize(pMgmt->runngingHash);
@@ -76,7 +76,7 @@ int32_t vmGetAllVnodeListFromHashWithCreating(SVnodeMgmt *pMgmt, int32_t *numOfV
   size += creatingSize;
   SVnodeObj **pVnodes = taosMemoryCalloc(size, sizeof(SVnodeObj *));
   if (pVnodes == NULL) {
-    (void)taosThreadRwlockUnlock(&pMgmt->lock);
+    (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
     return terrno;
   }
 
@@ -107,7 +107,7 @@ int32_t vmGetAllVnodeListFromHashWithCreating(SVnodeMgmt *pMgmt, int32_t *numOfV
       taosHashCancelIterate(pMgmt->creatingHash, pIter);
     }
   }
-  (void)taosThreadRwlockUnlock(&pMgmt->lock);
+  (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
 
   *numOfVnodes = num;
   *ppVnodes = pVnodes;
@@ -116,13 +116,13 @@ int32_t vmGetAllVnodeListFromHashWithCreating(SVnodeMgmt *pMgmt, int32_t *numOfV
 }
 
 int32_t vmGetVnodeListFromHash(SVnodeMgmt *pMgmt, int32_t *numOfVnodes, SVnodeObj ***ppVnodes) {
-  (void)taosThreadRwlockRdlock(&pMgmt->lock);
+  (void)taosThreadRwlockRdlock(&pMgmt->hashLock);
 
   int32_t     num = 0;
   int32_t     size = taosHashGetSize(pMgmt->runngingHash);
   SVnodeObj **pVnodes = taosMemoryCalloc(size, sizeof(SVnodeObj *));
   if (pVnodes == NULL) {
-    (void)taosThreadRwlockUnlock(&pMgmt->lock);
+    (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
     return terrno;
   }
 
@@ -140,7 +140,7 @@ int32_t vmGetVnodeListFromHash(SVnodeMgmt *pMgmt, int32_t *numOfVnodes, SVnodeOb
     }
   }
 
-  (void)taosThreadRwlockUnlock(&pMgmt->lock);
+  (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
   *numOfVnodes = num;
   *ppVnodes = pVnodes;
 
diff --git a/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c b/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c
index 1dea7d3cad..5c16349d9a 100644
--- a/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c
+++ b/source/dnode/mgmt/mgmt_vnode/src/vmHandle.c
@@ -25,7 +25,7 @@ void vmGetVnodeLoads(SVnodeMgmt *pMgmt, SMonVloadInfo *pInfo, bool isReset) {
 
   tfsUpdateSize(pMgmt->pTfs);
 
-  (void)taosThreadRwlockRdlock(&pMgmt->lock);
+  (void)taosThreadRwlockRdlock(&pMgmt->hashLock);
 
   void *pIter = taosHashIterate(pMgmt->runngingHash, NULL);
   while (pIter) {
@@ -46,14 +46,14 @@ void vmGetVnodeLoads(SVnodeMgmt *pMgmt, SMonVloadInfo *pInfo, bool isReset) {
     pIter = taosHashIterate(pMgmt->runngingHash, pIter);
   }
 
-  (void)taosThreadRwlockUnlock(&pMgmt->lock);
+  (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
 }
 
 void vmGetVnodeLoadsLite(SVnodeMgmt *pMgmt, SMonVloadInfo *pInfo) {
   pInfo->pVloads = taosArrayInit(pMgmt->state.totalVnodes, sizeof(SVnodeLoadLite));
   if (!pInfo->pVloads) return;
 
-  (void)taosThreadRwlockRdlock(&pMgmt->lock);
+  (void)taosThreadRwlockRdlock(&pMgmt->hashLock);
 
   void *pIter = taosHashIterate(pMgmt->runngingHash, NULL);
   while (pIter) {
@@ -74,7 +74,7 @@ void vmGetVnodeLoadsLite(SVnodeMgmt *pMgmt, SMonVloadInfo *pInfo) {
     pIter = taosHashIterate(pMgmt->runngingHash, pIter);
   }
 
-  (void)taosThreadRwlockUnlock(&pMgmt->lock);
+  (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
 }
 
 void vmGetMonitorInfo(SVnodeMgmt *pMgmt, SMonVmInfo *pInfo) {
@@ -137,7 +137,7 @@ void vmCleanExpriedSamples(SVnodeMgmt *pMgmt) {
     dError("failed to get vgroup ids");
     return;
   }
-  (void)taosThreadRwlockRdlock(&pMgmt->lock);
+  (void)taosThreadRwlockRdlock(&pMgmt->hashLock);
   for (int i = 0; i < list_size; i++) {
     int32_t vgroup_id = vgroup_ids[i];
     void   *vnode = taosHashGet(pMgmt->runngingHash, &vgroup_id, sizeof(int32_t));
@@ -148,7 +148,7 @@ void vmCleanExpriedSamples(SVnodeMgmt *pMgmt) {
       }
     }
   }
-  (void)taosThreadRwlockUnlock(&pMgmt->lock);
+  (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
   if (vgroup_ids) taosMemoryFree(vgroup_ids);
   if (keys) taosMemoryFree(keys);
   return;
diff --git a/source/dnode/mgmt/mgmt_vnode/src/vmInt.c b/source/dnode/mgmt/mgmt_vnode/src/vmInt.c
index 6f30977e10..8871cd575f 100644
--- a/source/dnode/mgmt/mgmt_vnode/src/vmInt.c
+++ b/source/dnode/mgmt/mgmt_vnode/src/vmInt.c
@@ -24,12 +24,12 @@ int32_t vmGetPrimaryDisk(SVnodeMgmt *pMgmt, int32_t vgId) {
   int32_t    diskId = -1;
   SVnodeObj *pVnode = NULL;
 
-  (void)taosThreadRwlockRdlock(&pMgmt->lock);
+  (void)taosThreadRwlockRdlock(&pMgmt->hashLock);
   int32_t r = taosHashGetDup(pMgmt->runngingHash, &vgId, sizeof(int32_t), (void *)&pVnode);
   if (pVnode != NULL) {
     diskId = pVnode->diskPrimary;
   }
-  (void)taosThreadRwlockUnlock(&pMgmt->lock);
+  (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
   return diskId;
 }
 
@@ -62,7 +62,7 @@ static int32_t vmRegisterCreatingState(SVnodeMgmt *pMgmt, int32_t vgId, int32_t
   pCreatingVnode->vgId = vgId;
   pCreatingVnode->diskPrimary = diskId;
 
-  code = taosThreadRwlockWrlock(&pMgmt->lock);
+  code = taosThreadRwlockWrlock(&pMgmt->hashLock);
   if (code != 0) {
     taosMemoryFree(pCreatingVnode);
     return code;
@@ -75,7 +75,7 @@ static int32_t vmRegisterCreatingState(SVnodeMgmt *pMgmt, int32_t vgId, int32_t
     taosMemoryFree(pCreatingVnode);
   }
 
-  int32_t r = taosThreadRwlockUnlock(&pMgmt->lock);
+  int32_t r = taosThreadRwlockUnlock(&pMgmt->hashLock);
   if (r != 0) {
     dError("vgId:%d, failed to unlock since %s", vgId, tstrerror(r));
   }
@@ -86,7 +86,7 @@ static int32_t vmRegisterCreatingState(SVnodeMgmt *pMgmt, int32_t vgId, int32_t
 static void vmUnRegisterCreatingState(SVnodeMgmt *pMgmt, int32_t vgId) {
   SVnodeObj *pOld = NULL;
 
-  (void)taosThreadRwlockWrlock(&pMgmt->lock);
+  (void)taosThreadRwlockWrlock(&pMgmt->hashLock);
   int32_t r = taosHashGetDup(pMgmt->creatingHash, &vgId, sizeof(int32_t), (void *)&pOld);
   if (r != 0) {
     dError("vgId:%d, failed to get vnode from creating Hash", vgId);
@@ -96,7 +96,7 @@ static void vmUnRegisterCreatingState(SVnodeMgmt *pMgmt, int32_t vgId) {
   if (r != 0) {
     dError("vgId:%d, failed to remove vnode from creatingHash", vgId);
   }
-  (void)taosThreadRwlockUnlock(&pMgmt->lock);
+  (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
 
   if (pOld) {
     dTrace("vgId:%d, free vnode pOld:%p", vgId, &pOld);
@@ -205,7 +205,7 @@ void vmCleanPrimaryDisk(SVnodeMgmt *pMgmt, int32_t vgId) { vmUnRegisterCreatingS
 SVnodeObj *vmAcquireVnodeImpl(SVnodeMgmt *pMgmt, int32_t vgId, bool strict) {
   SVnodeObj *pVnode = NULL;
 
-  (void)taosThreadRwlockRdlock(&pMgmt->lock);
+  (void)taosThreadRwlockRdlock(&pMgmt->hashLock);
   int32_t r = taosHashGetDup(pMgmt->runngingHash, &vgId, sizeof(int32_t), (void *)&pVnode);
   if (pVnode == NULL || strict && (pVnode->dropped || pVnode->failed)) {
     terrno = TSDB_CODE_VND_INVALID_VGROUP_ID;
@@ -214,7 +214,7 @@ SVnodeObj *vmAcquireVnodeImpl(SVnodeMgmt *pMgmt, int32_t vgId, bool strict) {
     int32_t refCount = atomic_add_fetch_32(&pVnode->refCount, 1);
     dTrace("vgId:%d, acquire vnode, vnode:%p, ref:%d", pVnode->vgId, pVnode, refCount);
   }
-  (void)taosThreadRwlockUnlock(&pMgmt->lock);
+  (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
 
   return pVnode;
 }
@@ -334,10 +334,10 @@ int32_t vmOpenVnode(SVnodeMgmt *pMgmt, SWrapperCfg *pCfg, SVnode *pImpl) {
     pVnode->failed = 1;
   }
 
-  (void)taosThreadRwlockWrlock(&pMgmt->lock);
+  (void)taosThreadRwlockWrlock(&pMgmt->hashLock);
   int32_t code = vmRegisterRunningState(pMgmt, pVnode);
   vmUnRegisterClosedState(pMgmt, pVnode);
-  (void)taosThreadRwlockUnlock(&pMgmt->lock);
+  (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
 
   return code;
 }
@@ -350,15 +350,15 @@ void vmCloseVnode(SVnodeMgmt *pMgmt, SVnodeObj *pVnode, bool commitAndRemoveWal,
     vnodeProposeCommitOnNeed(pVnode->pImpl, atExit);
   }
 
-  (void)taosThreadRwlockWrlock(&pMgmt->lock);
+  (void)taosThreadRwlockWrlock(&pMgmt->hashLock);
   vmUnRegisterRunningState(pMgmt, pVnode->vgId);
   if (keepClosed) {
     if (vmRegisterClosedState(pMgmt, pVnode) != 0) {
-      (void)taosThreadRwlockUnlock(&pMgmt->lock);
+      (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
       return;
     };
   }
-  (void)taosThreadRwlockUnlock(&pMgmt->lock);
+  (void)taosThreadRwlockUnlock(&pMgmt->hashLock);
 
   vmReleaseVnode(pMgmt, pVnode);
 
@@ -454,14 +454,14 @@ _closed:
 
 void vmCloseFailedVnode(SVnodeMgmt *pMgmt, int32_t vgId) {
   int32_t r = 0;
-  r = taosThreadRwlockWrlock(&pMgmt->lock);
+  r = taosThreadRwlockWrlock(&pMgmt->hashLock);
   if (r != 0) {
     dError("vgId:%d, failed to lock since %s", vgId, tstrerror(r));
   }
   if (r == 0) {
     vmUnRegisterRunningState(pMgmt, vgId);
   }
-  r = taosThreadRwlockUnlock(&pMgmt->lock);
+  r = taosThreadRwlockUnlock(&pMgmt->hashLock);
   if (r != 0) {
     dError("vgId:%d, failed to unlock since %s", vgId, tstrerror(r));
   }
@@ -796,7 +796,7 @@ static void vmCleanup(SVnodeMgmt *pMgmt) {
   vmCloseVnodes(pMgmt);
   vmStopWorker(pMgmt);
   vnodeCleanup();
-  (void)taosThreadRwlockDestroy(&pMgmt->lock);
+  (void)taosThreadRwlockDestroy(&pMgmt->hashLock);
   (void)taosThreadMutexDestroy(&pMgmt->mutex);
   (void)taosThreadMutexDestroy(&pMgmt->fileLock);
   taosMemoryFree(pMgmt);
@@ -884,7 +884,7 @@ static int32_t vmInit(SMgmtInputOpt *pInput, SMgmtOutputOpt *pOutput) {
   pMgmt->msgCb.qsizeFp = (GetQueueSizeFp)vmGetQueueSize;
   pMgmt->msgCb.mgmt = pMgmt;
 
-  code = taosThreadRwlockInit(&pMgmt->lock, NULL);
+  code = taosThreadRwlockInit(&pMgmt->hashLock, NULL);
   if (code != 0) {
     code = TAOS_SYSTEM_ERROR(errno);
     goto _OVER;
diff --git a/source/libs/sync/src/syncUtil.c b/source/libs/sync/src/syncUtil.c
index 5a2571a291..44e8d97aca 100644
--- a/source/libs/sync/src/syncUtil.c
+++ b/source/libs/sync/src/syncUtil.c
@@ -215,7 +215,7 @@ void syncPrintNodeLog(const char* flags, ELogLevel level, int32_t dflag, bool fo
 
   SSnapshot snapshot = {.data = NULL, .lastApplyIndex = -1, .lastApplyTerm = 0};
   if (pNode->pFsm != NULL && pNode->pFsm->FpGetSnapshotInfo != NULL) {
-    (void)pNode->pFsm->FpGetSnapshotInfo(pNode->pFsm, &snapshot);
+    (void)pNode->pFsm->FpGetSnapshotInfo(pNode->pFsm, &snapshot);  // vnodeSyncGetSnapshotInfo
   }
 
   SyncIndex logLastIndex = SYNC_INDEX_INVALID;
@@ -253,13 +253,15 @@ void syncPrintNodeLog(const char* flags, ELogLevel level, int32_t dflag, bool fo
   va_end(argpointer);
 
   int32_t aqItems = 0;
+  /*
   if (pNode != NULL && pNode->pFsm != NULL && pNode->pFsm->FpApplyQueueItems != NULL) {
-    aqItems = pNode->pFsm->FpApplyQueueItems(pNode->pFsm);
+    aqItems = pNode->pFsm->FpApplyQueueItems(pNode->pFsm);  // vnodeApplyQueueItems
   }
+  */
 
   // restore error code
   terrno = errCode;
-  SyncIndex appliedIndex = pNode->pFsm->FpAppliedIndexCb(pNode->pFsm);
+  SyncIndex appliedIndex = pNode->pFsm->FpAppliedIndexCb(pNode->pFsm);  // vnodeSyncAppliedIndex
 
   if (pNode != NULL) {
     taosPrintLog(
@@ -426,19 +428,25 @@ void syncLogSendHeartbeat(SSyncNode* pSyncNode, const SyncHeartbeat* pMsg, bool
 }
 
 void syncLogRecvHeartbeat(SSyncNode* pSyncNode, const SyncHeartbeat* pMsg, int64_t timeDiff, const char* s) {
+  char pBuf[TD_TIME_STR_LEN] = {0};
+  if (pMsg->timeStamp > 0) {
+    if (formatTimestampLocal(pBuf, pMsg->timeStamp, TSDB_TIME_PRECISION_MILLI) == NULL) {
+      pBuf[0] = '\0';
+    }
+  }
   if (timeDiff > SYNC_HEARTBEAT_SLOW_MS) {
     pSyncNode->hbSlowNum++;
 
     sNTrace(pSyncNode,
-            "recv sync-heartbeat from dnode:%d slow {term:%" PRId64 ", commit-index:%" PRId64 ", min-match:%" PRId64
-            ", ts:%" PRId64 "}, QID:%s, net elapsed:%" PRId64,
-            DID(&pMsg->srcId), pMsg->term, pMsg->commitIndex, pMsg->minMatchIndex, pMsg->timeStamp, s, timeDiff);
+            "recv sync-heartbeat from dnode:%d slow(%d ms) {term:%" PRId64 ", commit-index:%" PRId64
+            ", min-match:%" PRId64 ", ts:%s}, QID:%s, net elapsed:%" PRId64 "ms",
+            DID(&pMsg->srcId), SYNC_HEARTBEAT_SLOW_MS, pMsg->term, pMsg->commitIndex, pMsg->minMatchIndex, pBuf, s, timeDiff);
+  } else {
+    sNTrace(pSyncNode,
+            "recv sync-heartbeat from dnode:%d {term:%" PRId64 ", commit-index:%" PRId64 ", min-match:%" PRId64
+            ", ts:%s}, QID:%s, net elapsed:%" PRId64 "ms",
+            DID(&pMsg->srcId), pMsg->term, pMsg->commitIndex, pMsg->minMatchIndex, pBuf, s, timeDiff);
   }
-
-  sNTrace(pSyncNode,
-          "recv sync-heartbeat from dnode:%d {term:%" PRId64 ", commit-index:%" PRId64 ", min-match:%" PRId64
-          ", ts:%" PRId64 "}, QID:%s, net elapsed:%" PRId64,
-          DID(&pMsg->srcId), pMsg->term, pMsg->commitIndex, pMsg->minMatchIndex, pMsg->timeStamp, s, timeDiff);
 }
 
 void syncLogSendHeartbeatReply(SSyncNode* pSyncNode, const SyncHeartbeatReply* pMsg, const char* s) {
diff --git a/source/libs/transport/src/tmsgcb.c b/source/libs/transport/src/tmsgcb.c
index e87011f097..0632e21d20 100644
--- a/source/libs/transport/src/tmsgcb.c
+++ b/source/libs/transport/src/tmsgcb.c
@@ -37,7 +37,7 @@ int32_t tmsgPutToQueue(const SMsgCb* msgcb, EQueueType qtype, SRpcMsg* pMsg) {
 }
 
 int32_t tmsgGetQueueSize(const SMsgCb* msgcb, int32_t vgId, EQueueType qtype) {
-  return (*msgcb->qsizeFp)(msgcb->mgmt, vgId, qtype);
+  return (*msgcb->qsizeFp)(msgcb->mgmt, vgId, qtype);  // vmGetQueueSize
 }
 
 int32_t tmsgSendReq(const SEpSet* epSet, SRpcMsg* pMsg) {

From e257f69c61d61f4b75547d05905cec4f239befac Mon Sep 17 00:00:00 2001
From: dmchen <cademfly@hotmail.com>
Date: Fri, 28 Feb 2025 16:02:18 +0800
Subject: [PATCH 11/21] fix:remove-lock-from-heartbeat-log

---
 source/libs/sync/inc/syncUtil.h |   9 +++
 source/libs/sync/src/syncUtil.c | 133 +++++++++++++++++++++++++-------
 2 files changed, 115 insertions(+), 27 deletions(-)

diff --git a/source/libs/sync/inc/syncUtil.h b/source/libs/sync/inc/syncUtil.h
index e7110c0964..95ea75731c 100644
--- a/source/libs/sync/inc/syncUtil.h
+++ b/source/libs/sync/inc/syncUtil.h
@@ -53,6 +53,13 @@ extern "C" {
 #define sNDebug(pNode, ...)  if (sDebugFlag & DEBUG_DEBUG) { syncPrintNodeLog("SYN ",       DEBUG_DEBUG, sDebugFlag, true, pNode,  __VA_ARGS__); }
 #define sNTrace(pNode, ...)  if (sDebugFlag & DEBUG_TRACE) { syncPrintNodeLog("SYN ",       DEBUG_TRACE, sDebugFlag, true, pNode,  __VA_ARGS__); }
 
+#define sHFatal(pNode, ...)  if (sDebugFlag & DEBUG_FATAL) { syncPrintHbLog("SYN FATAL ", DEBUG_FATAL, 255, true,        pNode,  __VA_ARGS__); }
+#define sHError(pNode, ...)  if (sDebugFlag & DEBUG_ERROR) { syncPrintHbLog("SYN ERROR ", DEBUG_ERROR, 255, true,       pNode,  __VA_ARGS__); }
+#define sHWarn(pNode, ...)   if (sDebugFlag & DEBUG_WARN)  { syncPrintHbLog("SYN WARN ",  DEBUG_WARN,  255, true,        pNode,  __VA_ARGS__); }
+#define sHInfo(pNode, ...)   if (sDebugFlag & DEBUG_INFO)  { syncPrintHbLog("SYN ",       DEBUG_INFO,  255, true,        pNode, __VA_ARGS__); }
+#define sHDebug(pNode, ...)  if (sDebugFlag & DEBUG_DEBUG) { syncPrintHbLog("SYN ",       DEBUG_DEBUG, sDebugFlag, true, pNode,  __VA_ARGS__); }
+#define sHTrace(pNode, ...)  if (sDebugFlag & DEBUG_TRACE) { syncPrintHbLog("SYN ",       DEBUG_TRACE, sDebugFlag, true, pNode,  __VA_ARGS__); }
+
 #define sSFatal(pSender, ...)  if (sDebugFlag & DEBUG_FATAL) { syncPrintSnapshotSenderLog("SYN FATAL ", DEBUG_FATAL, 255,        pSender, __VA_ARGS__); }
 #define sSError(pSender, ...)  if (sDebugFlag & DEBUG_ERROR) { syncPrintSnapshotSenderLog("SYN ERROR ", DEBUG_ERROR, 255,        pSender, __VA_ARGS__); }
 #define sSWarn(pSender, ...)   if (sDebugFlag & DEBUG_WARN)  { syncPrintSnapshotSenderLog("SYN WARN ",  DEBUG_WARN,  255,        pSender, __VA_ARGS__); }
@@ -87,6 +94,8 @@ void syncUtilGenerateArbToken(int32_t nodeId, int32_t groupId, char* buf);
 
 void syncPrintNodeLog(const char* flags, ELogLevel level, int32_t dflag, bool formatTime, SSyncNode* pNode,
                       const char* format, ...);
+void syncPrintHbLog(const char* flags, ELogLevel level, int32_t dflag, bool formatTime, SSyncNode* pNode,
+                    const char* format, ...);
 void syncPrintSnapshotSenderLog(const char* flags, ELogLevel level, int32_t dflag, SSyncSnapshotSender* pSender,
                                 const char* format, ...);
 void syncPrintSnapshotReceiverLog(const char* flags, ELogLevel level, int32_t dflag, SSyncSnapshotReceiver* pReceiver,
diff --git a/source/libs/sync/src/syncUtil.c b/source/libs/sync/src/syncUtil.c
index 44e8d97aca..435586041a 100644
--- a/source/libs/sync/src/syncUtil.c
+++ b/source/libs/sync/src/syncUtil.c
@@ -253,11 +253,9 @@ void syncPrintNodeLog(const char* flags, ELogLevel level, int32_t dflag, bool fo
   va_end(argpointer);
 
   int32_t aqItems = 0;
-  /*
   if (pNode != NULL && pNode->pFsm != NULL && pNode->pFsm->FpApplyQueueItems != NULL) {
     aqItems = pNode->pFsm->FpApplyQueueItems(pNode->pFsm);  // vnodeApplyQueueItems
   }
-  */
 
   // restore error code
   terrno = errCode;
@@ -272,15 +270,71 @@ void syncPrintNodeLog(const char* flags, ELogLevel level, int32_t dflag, bool fo
         ", elect-times:%d, as-leader-times:%d, as-assigned-leader-times:%d, cfg-ch-times:%d, hb-slow:%d, hbr-slow:%d, "
         "aq-items:%d, snaping:%" PRId64 ", replicas:%d, last-cfg:%" PRId64
         ", chging:%d, restore:%d, quorum:%d, elect-lc-timer:%" PRId64 ", hb:%" PRId64
-        ", buffer:%s, repl-mgrs:%s, members:%s, send hb:%s, recv hb:%s, recv hb-reply:%s, arb-token:%s, msg[sent:%d, recv:%d, slow-recev:%d]",
+        ", buffer:%s, repl-mgrs:%s, members:%s, send hb:%s, recv hb:%s, recv hb-reply:%s, arb-token:%s, msg[sent:%d, "
+        "recv:%d, slow-recv:%d]",
         pNode->vgId, eventLog, syncStr(pNode->state), currentTerm, pNode->commitIndex, pNode->assignedCommitIndex,
         appliedIndex, logBeginIndex, logLastIndex, pNode->minMatchIndex, snapshot.lastApplyIndex,
         snapshot.lastApplyTerm, pNode->electNum, pNode->becomeLeaderNum, pNode->becomeAssignedLeaderNum,
         pNode->configChangeNum, pNode->hbSlowNum, pNode->hbrSlowNum, aqItems, pNode->snapshottingIndex,
         pNode->replicaNum, pNode->raftCfg.lastConfigIndex, pNode->changing, pNode->restoreFinish,
         syncNodeDynamicQuorum(pNode), pNode->electTimerLogicClock, pNode->heartbeatTimerLogicClockUser, bufferStatesStr,
-        replMgrStatesStr, cfgStr, sentHbTimeStr, hbTimeStr, hbrTimeStr, pNode->arbToken, pNode->sendCount, pNode->recvCount,
-        pNode->slowCount);
+        replMgrStatesStr, cfgStr, sentHbTimeStr, hbTimeStr, hbrTimeStr, pNode->arbToken, pNode->sendCount,
+        pNode->recvCount, pNode->slowCount);
+  }
+}
+
+void syncPrintHbLog(const char* flags, ELogLevel level, int32_t dflag, bool formatTime, SSyncNode* pNode,
+                    const char* format, ...) {
+  if (pNode == NULL || pNode->pLogStore == NULL) return;
+  int64_t currentTerm = raftStoreGetTerm(pNode);
+
+  // save error code, otherwise it will be overwritten
+  int32_t errCode = terrno;
+
+  int32_t cacheHit = pNode->pLogStore->cacheHit;
+  int32_t cacheMiss = pNode->pLogStore->cacheMiss;
+
+  char cfgStr[1024] = "";
+  syncCfg2SimpleStr(&pNode->raftCfg.cfg, cfgStr, sizeof(cfgStr));
+
+  char replMgrStatesStr[1024] = "";
+  syncLogReplStates2Str(pNode, replMgrStatesStr, sizeof(replMgrStatesStr));
+
+  char bufferStatesStr[256] = "";
+  syncLogBufferStates2Str(pNode, bufferStatesStr, sizeof(bufferStatesStr));
+
+  char hbrTimeStr[256] = "";
+  syncHearbeatReplyTime2Str(pNode, hbrTimeStr, sizeof(hbrTimeStr), formatTime);
+
+  char hbTimeStr[256] = "";
+  syncHearbeatTime2Str(pNode, hbTimeStr, sizeof(hbTimeStr), formatTime);
+
+  char sentHbTimeStr[512] = "";
+  syncSentHearbeatTime2Str(pNode, sentHbTimeStr, sizeof(sentHbTimeStr), formatTime);
+
+  char    eventLog[512];  // {0};
+  va_list argpointer;
+  va_start(argpointer, format);
+  int32_t writeLen = vsnprintf(eventLog, sizeof(eventLog), format, argpointer);
+  va_end(argpointer);
+
+  terrno = errCode;
+
+  if (pNode != NULL) {
+    taosPrintLog(
+        flags, level, dflag,
+        "vgId:%d, %s, sync:%s, term:%" PRIu64 ", commit-index:%" PRId64 ", assigned-index:%" PRId64 ", min:%" PRId64
+        ", elect-times:%d, as-leader-times:%d, as-assigned-leader-times:%d, cfg-ch-times:%d, hb-slow:%d, hbr-slow:%d, "
+        ", snaping:%" PRId64 ", replicas:%d, last-cfg:%" PRId64
+        ", chging:%d, restore:%d, quorum:%d, elect-lc-timer:%" PRId64 ", hb:%" PRId64
+        ", buffer:%s, repl-mgrs:%s, members:%s, send hb:%s, recv hb:%s, recv hb-reply:%s, arb-token:%s, msg[sent:%d, "
+        "recv:%d, slow-recv:%d]",
+        pNode->vgId, eventLog, syncStr(pNode->state), currentTerm, pNode->commitIndex, pNode->assignedCommitIndex,
+        pNode->minMatchIndex, pNode->electNum, pNode->becomeLeaderNum, pNode->becomeAssignedLeaderNum,
+        pNode->configChangeNum, pNode->hbSlowNum, pNode->hbrSlowNum, pNode->snapshottingIndex, pNode->replicaNum,
+        pNode->raftCfg.lastConfigIndex, pNode->changing, pNode->restoreFinish, syncNodeDynamicQuorum(pNode),
+        pNode->electTimerLogicClock, pNode->heartbeatTimerLogicClockUser, bufferStatesStr, replMgrStatesStr, cfgStr,
+        sentHbTimeStr, hbTimeStr, hbrTimeStr, pNode->arbToken, pNode->sendCount, pNode->recvCount, pNode->slowCount);
   }
 }
 
@@ -414,12 +468,12 @@ void syncLogRecvAppendEntriesReply(SSyncNode* pSyncNode, const SyncAppendEntries
 void syncLogSendHeartbeat(SSyncNode* pSyncNode, const SyncHeartbeat* pMsg, bool printX, int64_t timerElapsed,
                           int64_t execTime) {
   if (printX) {
-    sNTrace(pSyncNode,
+    sHTrace(pSyncNode,
             "send sync-heartbeat to dnode:%d {term:%" PRId64 ", commit-index:%" PRId64 ", min-match:%" PRId64
             ", ts:%" PRId64 "}, x",
             DID(&pMsg->destId), pMsg->term, pMsg->commitIndex, pMsg->minMatchIndex, pMsg->timeStamp);
   } else {
-    sNTrace(pSyncNode,
+    sHTrace(pSyncNode,
             "send sync-heartbeat to dnode:%d {term:%" PRId64 ", commit-index:%" PRId64 ", min-match:%" PRId64
             ", ts:%" PRId64 "}, timer-elapsed:%" PRId64 ", next-exec:%" PRId64,
             DID(&pMsg->destId), pMsg->term, pMsg->commitIndex, pMsg->minMatchIndex, pMsg->timeStamp, timerElapsed,
@@ -428,29 +482,39 @@ void syncLogSendHeartbeat(SSyncNode* pSyncNode, const SyncHeartbeat* pMsg, bool
 }
 
 void syncLogRecvHeartbeat(SSyncNode* pSyncNode, const SyncHeartbeat* pMsg, int64_t timeDiff, const char* s) {
-  char pBuf[TD_TIME_STR_LEN] = {0};
-  if (pMsg->timeStamp > 0) {
-    if (formatTimestampLocal(pBuf, pMsg->timeStamp, TSDB_TIME_PRECISION_MILLI) == NULL) {
-      pBuf[0] = '\0';
-    }
-  }
   if (timeDiff > SYNC_HEARTBEAT_SLOW_MS) {
     pSyncNode->hbSlowNum++;
 
-    sNTrace(pSyncNode,
+    char pBuf[TD_TIME_STR_LEN] = {0};
+    if (pMsg->timeStamp > 0) {
+      if (formatTimestampLocal(pBuf, pMsg->timeStamp, TSDB_TIME_PRECISION_MILLI) == NULL) {
+        pBuf[0] = '\0';
+      }
+    }
+
+    sHError(pSyncNode,
             "recv sync-heartbeat from dnode:%d slow(%d ms) {term:%" PRId64 ", commit-index:%" PRId64
             ", min-match:%" PRId64 ", ts:%s}, QID:%s, net elapsed:%" PRId64 "ms",
-            DID(&pMsg->srcId), SYNC_HEARTBEAT_SLOW_MS, pMsg->term, pMsg->commitIndex, pMsg->minMatchIndex, pBuf, s, timeDiff);
+            DID(&pMsg->srcId), SYNC_HEARTBEAT_SLOW_MS, pMsg->term, pMsg->commitIndex, pMsg->minMatchIndex, pBuf, s,
+            timeDiff);
   } else {
-    sNTrace(pSyncNode,
-            "recv sync-heartbeat from dnode:%d {term:%" PRId64 ", commit-index:%" PRId64 ", min-match:%" PRId64
-            ", ts:%s}, QID:%s, net elapsed:%" PRId64 "ms",
-            DID(&pMsg->srcId), pMsg->term, pMsg->commitIndex, pMsg->minMatchIndex, pBuf, s, timeDiff);
+    if (sDebugFlag & DEBUG_TRACE) {
+      char pBuf[TD_TIME_STR_LEN] = {0};
+      if (pMsg->timeStamp > 0) {
+        if (formatTimestampLocal(pBuf, pMsg->timeStamp, TSDB_TIME_PRECISION_MILLI) == NULL) {
+          pBuf[0] = '\0';
+        }
+      }
+      sHTrace(pSyncNode,
+              "recv sync-heartbeat from dnode:%d {term:%" PRId64 ", commit-index:%" PRId64 ", min-match:%" PRId64
+              ", ts:%s}, QID:%s, net elapsed:%" PRId64 "ms",
+              DID(&pMsg->srcId), pMsg->term, pMsg->commitIndex, pMsg->minMatchIndex, pBuf, s, timeDiff);
+    }
   }
 }
 
 void syncLogSendHeartbeatReply(SSyncNode* pSyncNode, const SyncHeartbeatReply* pMsg, const char* s) {
-  sNTrace(pSyncNode, "send sync-heartbeat-reply from dnode:%d {term:%" PRId64 ", ts:%" PRId64 "}, %s",
+  sHTrace(pSyncNode, "send sync-heartbeat-reply from dnode:%d {term:%" PRId64 ", ts:%" PRId64 "}, %s",
           DID(&pMsg->destId), pMsg->term, pMsg->timeStamp, s);
 }
 
@@ -458,14 +522,29 @@ void syncLogRecvHeartbeatReply(SSyncNode* pSyncNode, const SyncHeartbeatReply* p
   if (timeDiff > SYNC_HEARTBEAT_REPLY_SLOW_MS) {
     pSyncNode->hbrSlowNum++;
 
-    sNTrace(pSyncNode,
-            "recv sync-heartbeat-reply from dnode:%d slow {term:%" PRId64 ", ts:%" PRId64 "}, %s, net elapsed:%" PRId64,
-            DID(&pMsg->srcId), pMsg->term, pMsg->timeStamp, s, timeDiff);
-  }
+    char pBuf[TD_TIME_STR_LEN] = {0};
+    if (pMsg->timeStamp > 0) {
+      if (formatTimestampLocal(pBuf, pMsg->timeStamp, TSDB_TIME_PRECISION_MILLI) == NULL) {
+        pBuf[0] = '\0';
+      }
+    }
 
-  sNTrace(pSyncNode,
-          "recv sync-heartbeat-reply from dnode:%d {term:%" PRId64 ", ts:%" PRId64 "}, %s, net elapsed:%" PRId64,
-          DID(&pMsg->srcId), pMsg->term, pMsg->timeStamp, s, timeDiff);
+    sHError(pSyncNode,
+            "recv sync-heartbeat-reply from dnode:%d slow(%d ms) {term:%" PRId64 ", ts:%s}, %s, net elapsed:%" PRId64,
+            DID(&pMsg->srcId), SYNC_HEARTBEAT_REPLY_SLOW_MS, pMsg->term, pBuf, s, timeDiff);
+  } else {
+    if (sDebugFlag & DEBUG_TRACE) {
+      char pBuf[TD_TIME_STR_LEN] = {0};
+      if (pMsg->timeStamp > 0) {
+        if (formatTimestampLocal(pBuf, pMsg->timeStamp, TSDB_TIME_PRECISION_MILLI) == NULL) {
+          pBuf[0] = '\0';
+        }
+      }
+      sHTrace(pSyncNode,
+              "recv sync-heartbeat-reply from dnode:%d {term:%" PRId64 ", ts:%" PRId64 "}, %s, net elapsed:%" PRId64,
+              DID(&pMsg->srcId), pMsg->term, pMsg->timeStamp, s, timeDiff);
+    }
+  }
 }
 
 void syncLogSendSyncSnapshotSend(SSyncNode* pSyncNode, const SyncSnapshotSend* pMsg, const char* s) {

From 65773866ca212d26ffaabe07bad781361ce699ea Mon Sep 17 00:00:00 2001
From: dmchen <cademfly@hotmail.com>
Date: Sat, 1 Mar 2025 00:52:11 +0800
Subject: [PATCH 12/21] fix:remove-send-status-lock

---
 source/dnode/mgmt/mgmt_dnode/src/dmHandle.c | 49 ++++++++++++++-------
 1 file changed, 34 insertions(+), 15 deletions(-)

diff --git a/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c b/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
index 49f634443f..4d0135c07f 100644
--- a/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
+++ b/source/dnode/mgmt/mgmt_dnode/src/dmHandle.c
@@ -26,6 +26,7 @@ extern SConfig *tsCfg;
 
 SMonVloadInfo tsVinfo = {0};
 SMnodeLoad    tsMLoad = {0};
+SDnodeData    tsDnodeData = {0};
 
 static void dmUpdateDnodeCfg(SDnodeMgmt *pMgmt, SDnodeCfg *pCfg) {
   int32_t code = 0;
@@ -170,22 +171,27 @@ void dmSendStatusReq(SDnodeMgmt *pMgmt) {
   SStatusReq req = {0};
   req.timestamp = taosGetTimestampMs();
 
-  dDebug("send status req to mnode, statusSeq:%d, begin to mgnt lock", pMgmt->statusSeq);
-  (void)taosThreadRwlockRdlock(&pMgmt->pData->lock);
+  dDebug("send status req to mnode, statusSeq:%d, begin to mgnt statusInfolock", pMgmt->statusSeq);
+  if (taosThreadMutexLock(&pMgmt->pData->statusInfolock) != 0) {
+    dError("failed to lock status info lock");
+    return;
+  }
+
+  dDebug("send status req to mnode, statusSeq:%d, begin to get dnode info", pMgmt->statusSeq);
   req.sver = tsVersion;
-  req.dnodeVer = pMgmt->pData->dnodeVer;
-  req.dnodeId = pMgmt->pData->dnodeId;
-  req.clusterId = pMgmt->pData->clusterId;
+  req.dnodeVer = tsDnodeData.dnodeVer;
+  req.dnodeId = tsDnodeData.dnodeId;
+  req.clusterId = tsDnodeData.clusterId;
   if (req.clusterId == 0) req.dnodeId = 0;
-  req.rebootTime = pMgmt->pData->rebootTime;
-  req.updateTime = pMgmt->pData->updateTime;
+  req.rebootTime = tsDnodeData.rebootTime;
+  req.updateTime = tsDnodeData.updateTime;
   req.numOfCores = tsNumOfCores;
   req.numOfSupportVnodes = tsNumOfSupportVnodes;
   req.numOfDiskCfg = tsDiskCfgNum;
   req.memTotal = tsTotalMemoryKB * 1024;
   req.memAvail = req.memTotal - tsQueueMemoryAllowed - tsApplyMemoryAllowed - 16 * 1024 * 1024;
   tstrncpy(req.dnodeEp, tsLocalEp, TSDB_EP_LEN);
-  tstrncpy(req.machineId, pMgmt->pData->machineId, TSDB_MACHINE_ID_LEN + 1);
+  tstrncpy(req.machineId, tsDnodeData.machineId, TSDB_MACHINE_ID_LEN + 1);
 
   req.clusterCfg.statusInterval = tsStatusInterval;
   req.clusterCfg.checkTime = 0;
@@ -207,17 +213,13 @@ void dmSendStatusReq(SDnodeMgmt *pMgmt) {
   memcpy(req.clusterCfg.timezone, tsTimezoneStr, TD_TIMEZONE_LEN);
   memcpy(req.clusterCfg.locale, tsLocale, TD_LOCALE_LEN);
   memcpy(req.clusterCfg.charset, tsCharset, TD_LOCALE_LEN);
-  (void)taosThreadRwlockUnlock(&pMgmt->pData->lock);
 
-  dDebug("send status req to mnode, statusSeq:%d, begin to get vnode and loads", pMgmt->statusSeq);
-  if (taosThreadMutexLock(&pMgmt->pData->statusInfolock) != 0) {
-    dError("failed to lock status info lock");
-    return;
-  }
+  dDebug("send status req to mnode, statusSeq:%d, begin to get vnode loads", pMgmt->statusSeq);
 
   req.pVloads = tsVinfo.pVloads;
   tsVinfo.pVloads = NULL;
 
+  dDebug("send status req to mnode, statusSeq:%d, begin to get mnode loads", pMgmt->statusSeq);
   req.mload = tsMLoad;
 
   if (taosThreadMutexUnlock(&pMgmt->pData->statusInfolock) != 0) {
@@ -416,8 +418,19 @@ void dmSendConfigReq(SDnodeMgmt *pMgmt) {
 }
 
 void dmUpdateStatusInfo(SDnodeMgmt *pMgmt) {
-  SMonVloadInfo vinfo = {0};
+  dDebug("begin to get dnode info");
+  SDnodeData dnodeData = {0};
+  (void)taosThreadRwlockRdlock(&pMgmt->pData->lock);
+  dnodeData.dnodeVer = pMgmt->pData->dnodeVer;
+  dnodeData.dnodeId = pMgmt->pData->dnodeId;
+  dnodeData.clusterId = pMgmt->pData->clusterId;
+  dnodeData.rebootTime = pMgmt->pData->rebootTime;
+  dnodeData.updateTime = pMgmt->pData->updateTime;
+  tstrncpy(dnodeData.machineId, pMgmt->pData->machineId, TSDB_MACHINE_ID_LEN + 1);
+  (void)taosThreadRwlockUnlock(&pMgmt->pData->lock);
+
   dDebug("begin to get vnode loads");
+  SMonVloadInfo vinfo = {0};
   (*pMgmt->getVnodeLoadsFp)(&vinfo);  // dmGetVnodeLoads
 
   dDebug("begin to get mnode loads");
@@ -429,6 +442,12 @@ void dmUpdateStatusInfo(SDnodeMgmt *pMgmt) {
     dError("failed to lock status info lock");
     return;
   }
+  tsDnodeData.dnodeVer = dnodeData.dnodeVer;
+  tsDnodeData.dnodeId = dnodeData.dnodeId;
+  tsDnodeData.clusterId = dnodeData.clusterId;
+  tsDnodeData.rebootTime = dnodeData.rebootTime;
+  tsDnodeData.updateTime = dnodeData.updateTime;
+  tstrncpy(tsDnodeData.machineId, dnodeData.machineId, TSDB_MACHINE_ID_LEN + 1);
 
   if (tsVinfo.pVloads == NULL) {
     tsVinfo.pVloads = vinfo.pVloads;

From f8cc9648f6c6af840b4c01aa30898678a0e4e7ce Mon Sep 17 00:00:00 2001
From: Haojun Liao <hjliao@taosdata.com>
Date: Sat, 1 Mar 2025 15:33:31 +0800
Subject: [PATCH 13/21] fix(stream): fix memory leak.

---
 source/dnode/mnode/impl/src/mndConfig.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/source/dnode/mnode/impl/src/mndConfig.c b/source/dnode/mnode/impl/src/mndConfig.c
index 099fff7aee..1e69ae2b5a 100644
--- a/source/dnode/mnode/impl/src/mndConfig.c
+++ b/source/dnode/mnode/impl/src/mndConfig.c
@@ -299,6 +299,8 @@ _OVER:
   }
   sdbRelease(pMnode->pSdb, vObj);
   cfgArrayCleanUp(array);
+
+  tFreeSConfigReq(&configReq);
   return code;
 }
 

From 14ffa60e6ebad4d341c3db07bae03940426958f0 Mon Sep 17 00:00:00 2001
From: Haojun Liao <hjliao@taosdata.com>
Date: Sat, 1 Mar 2025 15:44:02 +0800
Subject: [PATCH 14/21] fix(stream): remove invalid error check.

---
 source/libs/stream/src/streamQueue.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/source/libs/stream/src/streamQueue.c b/source/libs/stream/src/streamQueue.c
index 954e41f288..4139a5ccb7 100644
--- a/source/libs/stream/src/streamQueue.c
+++ b/source/libs/stream/src/streamQueue.c
@@ -144,21 +144,11 @@ void streamQueueNextItemInSourceQ(SStreamQueue* pQueue, SStreamQueueItem** pItem
 
   // let's try the ordinary input q
   pQueue->qItem = NULL;
-  int32_t code = taosGetQitem(pQueue->qall, &pQueue->qItem);
-  if (code) {
-    stError("s-task:%s failed to get item in inputq, code:%s", id, tstrerror(code));
-  }
+  int32_t num = taosGetQitem(pQueue->qall, &pQueue->qItem);
 
   if (pQueue->qItem == NULL) {
-    code = taosReadAllQitems(pQueue->pQueue, pQueue->qall);
-    if (code) {
-      stError("s-task:%s failed to get all items in inputq, code:%s", id, tstrerror(code));
-    }
-
-    code = taosGetQitem(pQueue->qall, &pQueue->qItem);
-    if (code) {
-      stError("s-task:%s failed to get item in inputq, code:%s", id, tstrerror(code));
-    }
+    num = taosReadAllQitems(pQueue->pQueue, pQueue->qall);
+    num = taosGetQitem(pQueue->qall, &pQueue->qItem);
   }
 
   *pItem = streamQueueCurItem(pQueue);

From d2d7279f0d56c203db53ebcb6fd1b3976978e6d0 Mon Sep 17 00:00:00 2001
From: Haojun Liao <hjliao@taosdata.com>
Date: Sat, 1 Mar 2025 19:22:26 +0800
Subject: [PATCH 15/21] fix(stream): fix memory leak if error occurs.

---
 source/dnode/vnode/src/tq/tqSink.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/source/dnode/vnode/src/tq/tqSink.c b/source/dnode/vnode/src/tq/tqSink.c
index 4c2d0d7531..239a8e65b5 100644
--- a/source/dnode/vnode/src/tq/tqSink.c
+++ b/source/dnode/vnode/src/tq/tqSink.c
@@ -1418,6 +1418,7 @@ int32_t handleResultBlockMsg(SStreamTask* pTask, SSDataBlock* pDataBlock, int32_
   code = setDstTableDataUid(pVnode, pTask, pDataBlock, stbFullName, &tbData);
   if (code != TSDB_CODE_SUCCESS) {
     tqError("vgId:%d s-task:%s dst-table not exist, stb:%s discard stream results", vgId, id, stbFullName);
+    tDestroySubmitReq(&submitReq, TSDB_MSG_FLG_ENCODE);
     return code;
   }
 
@@ -1429,12 +1430,14 @@ int32_t handleResultBlockMsg(SStreamTask* pTask, SSDataBlock* pDataBlock, int32_
       tbData.pCreateTbReq = NULL;
     }
 
+    tDestroySubmitReq(&submitReq, TSDB_MSG_FLG_ENCODE);
     return code;
   }
 
   void* p = taosArrayPush(submitReq.aSubmitTbData, &tbData);
   if (p == NULL) {
     tqDebug("vgId:%d, s-task:%s failed to build submit msg, code:%s, data lost", vgId, id, tstrerror(terrno));
+    tDestroySubmitReq(&submitReq, TSDB_MSG_FLG_ENCODE);
     return terrno;
   }
 

From 4540b0f1ec8ce2fac23bc7a9ac64143e1b6f2bd4 Mon Sep 17 00:00:00 2001
From: zyyang <zyyang@taosdata.com>
Date: Mon, 3 Mar 2025 06:44:33 +0800
Subject: [PATCH 16/21] docs(backup): support s3#TS-5806

---
 docs/zh/08-operation/09-backup.md | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/docs/zh/08-operation/09-backup.md b/docs/zh/08-operation/09-backup.md
index fbc2e612e1..babd23c2ba 100644
--- a/docs/zh/08-operation/09-backup.md
+++ b/docs/zh/08-operation/09-backup.md
@@ -76,6 +76,17 @@ taosExplorer 服务页面中，进入“系统管理 - 备份”页面，在“
 8. 备份文件大小：备份文件的大小限制。当备份文件大小达到此限制时，会自动创建新的备份文件。
 9. 文件压缩等级：备份文件的压缩等级。支持：最快速度、最佳压缩比、兼具速度和压缩比。
 
+用户可以通过开启 S3 转储，将备份文件上传至 S3 存储服务上。开启 S3 转储，需要填写以下信息：
+
+1. S3 节点：S3 节点的地址。
+2. 访问密钥 ID：访问密钥 ID。
+3. 访问密钥：访问密钥。
+4. 存储桶：存储桶名称。
+5. 区域：存储桶所在的区域。
+6. 对象前缀：备份文件的对象前缀，类似于 S3 上的目录。
+7. 本地备份文件的保留时长：本地备份的保留时间，所有早于`当前时间 - backup_retention_period`的文件都需要上传到 S3。
+8. 本地备份文件的保留个数：本地备份文件的保留个数，本地只保留最新的`backup_retention_size`个备份文件。
+
 创建成功后，备份计划会开始按照配置的参数运行。在“备份计划”下的列表中，可以查看已创建的备份计划。
 
 备份计划支持以下操作：

From 79ca1c4bcfa064dd65649cf9b76cd51f4ff77f65 Mon Sep 17 00:00:00 2001
From: haoranchen <haoran920c@163.com>
Date: Mon, 3 Mar 2025 16:41:21 +0800
Subject: [PATCH 17/21] fix path error in packaging Dockerfile

---
 packaging/docker/Dockerfile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/packaging/docker/Dockerfile b/packaging/docker/Dockerfile
index c0aadac376..f1d5f7df23 100644
--- a/packaging/docker/Dockerfile
+++ b/packaging/docker/Dockerfile
@@ -15,7 +15,7 @@ ENV DEBIAN_FRONTEND=noninteractive
 ADD https://github.com/krallin/tini/releases/download/${TINI_VERSION}/tini-${cpuType} /tini
 RUN chmod +x /tini
 
-RUN tar -zxf ${pkgFile} && \
+RUN tar -zxf /root/${pkgFile} && \
     /bin/bash /root/${dirName}/install.sh -e no && \
     rm /root/${pkgFile} && \
     rm -rf /root/${dirName} && \
@@ -46,4 +46,4 @@ COPY ./bin/* /usr/bin/
 ENTRYPOINT ["/tini", "--", "/usr/bin/entrypoint.sh"]
 CMD ["taosd"]
 
-VOLUME [ "/var/lib/taos", "/var/log/taos", "/corefile" ]
\ No newline at end of file
+VOLUME [ "/var/lib/taos", "/var/log/taos", "/corefile" ]

From ffc9e0d27f6f9318cb73a2a556ed44bb7d6f784d Mon Sep 17 00:00:00 2001
From: haoranchen <haoran920c@163.com>
Date: Mon, 3 Mar 2025 16:44:38 +0800
Subject: [PATCH 18/21] fix: set dir in packaging dockerfile

---
 packaging/docker/Dockerfile | 1 +
 1 file changed, 1 insertion(+)

diff --git a/packaging/docker/Dockerfile b/packaging/docker/Dockerfile
index f1d5f7df23..010823ff08 100644
--- a/packaging/docker/Dockerfile
+++ b/packaging/docker/Dockerfile
@@ -15,6 +15,7 @@ ENV DEBIAN_FRONTEND=noninteractive
 ADD https://github.com/krallin/tini/releases/download/${TINI_VERSION}/tini-${cpuType} /tini
 RUN chmod +x /tini
 
+WORKDIR /root
 RUN tar -zxf /root/${pkgFile} && \
     /bin/bash /root/${dirName}/install.sh -e no && \
     rm /root/${pkgFile} && \

From 5ddc42874a565d46278665113cccd59b85a07ded Mon Sep 17 00:00:00 2001
From: menshibin <men_shi_bin@163.com>
Date: Mon, 3 Mar 2025 17:16:26 +0800
Subject: [PATCH 19/21] docs: add tableau configuration odbc instructions

---
 docs/en/10-third-party/05-bi/11-superset.md | 2 +-
 docs/en/10-third-party/05-bi/12-tableau.md  | 6 +++++-
 docs/en/10-third-party/05-bi/13-excel.md    | 2 +-
 docs/zh/10-third-party/05-bi/12-tableau.md  | 8 ++++++--
 docs/zh/10-third-party/05-bi/13-excel.md    | 2 +-
 5 files changed, 14 insertions(+), 6 deletions(-)

diff --git a/docs/en/10-third-party/05-bi/11-superset.md b/docs/en/10-third-party/05-bi/11-superset.md
index d8dc200fd6..2ec59a2172 100644
--- a/docs/en/10-third-party/05-bi/11-superset.md
+++ b/docs/en/10-third-party/05-bi/11-superset.md
@@ -7,7 +7,7 @@ Apache Superset is a modern enterprise level business intelligence (BI) web appl
 It is supported by the Apache Software Foundation and is an open source project with an active community and rich ecosystem.
 Apache Superset provides an intuitive user interface that makes creating, sharing, and visualizing data simple, while supporting multiple data sources and rich visualization options.  
 
-Through the Python connector of TDengine, Superset can support TDengine data sources and provide functions such as data presentation and analysis  
+Through the Python connector of TDengine, Superset can support TDengine data sources and provide functions such as data presentation and analysis.  
 
 ## Prerequisites
 
diff --git a/docs/en/10-third-party/05-bi/12-tableau.md b/docs/en/10-third-party/05-bi/12-tableau.md
index 31292644b2..7a3416d93f 100644
--- a/docs/en/10-third-party/05-bi/12-tableau.md
+++ b/docs/en/10-third-party/05-bi/12-tableau.md
@@ -10,7 +10,7 @@ Tableau is a well-known business intelligence tool that supports multiple data s
 
 Prepare the following environment:
 
-- TDengine 3.3.5.4 and above version is installed and running normally (both Enterprise and Community versions are available)
+- TDengine 3.3.5.8 and above version is installed and running normally (both Enterprise and Community versions are available)
 - taosAdapter is running normally, refer to [taosAdapter Reference](../../../tdengine-reference/components/taosadapter/)
 - Install and run Tableau Desktop (if not installed, please download and install Windows operating system 64-bit [Download Tableau Desktop](https://www.tableau.com/products/desktop/download)). Install Tableau please refer to [Tableau Desktop](https://www.tableau.com).
 - Download the latest Windows operating system X64 client driver from the TDengine official website and install it, refer to [Install ODBC Driver](../../../tdengine-reference/client-libraries/odbc/#Installation).
@@ -19,6 +19,10 @@ Prepare the following environment:
 
 **Step 1**, Search and open the "ODBC Data Source (64 bit)" management tool in the Start menu of the Windows operating system and configure it, refer to [Install ODBC Driver](../../../tdengine-reference/client-libraries/odbc/#Installation).
 
+:::tip
+It should be noted that when configuring the ODBC data source for Tableau, the 【Database】 configuration item on the TDengine ODBC data source configuration page is required. You need to select a database that can be successfully connected.
+:::
+
 **Step 2**, Start Tableau in the Windows system environment, then search for "ODBC" on its connection page and select "Other Databases (ODBC)".
 
 **Step 3**, Click the `DSN` radio button, then select the configured data source (MyTDengine), and click the `Connect` button. After the connection is successful, delete the content of the string attachment, and finally click the `Sign In` button.  
diff --git a/docs/en/10-third-party/05-bi/13-excel.md b/docs/en/10-third-party/05-bi/13-excel.md
index aff4b7adf6..114954e81f 100644
--- a/docs/en/10-third-party/05-bi/13-excel.md
+++ b/docs/en/10-third-party/05-bi/13-excel.md
@@ -10,7 +10,7 @@ toc_max_heading_level: 4
 
 Prepare the following environment:
 
-- TDengine 3.3.5.7 and above version is installed and running normally (both Enterprise and Community versions are available).
+- TDengine 3.3.5.8 and above version is installed and running normally (both Enterprise and Community versions are available).
 - taosAdapter is running normally, refer to [taosAdapter Reference](../../../tdengine-reference/components/taosadapter/).
 - Install and run Excel. If not installed, please download and install it. For specific instructions, please refer to Microsoft's official documentation.
 - Download the latest Windows operating system X64 client driver from the TDengine official website and install it, refer to [Install ODBC Driver](../../../tdengine-reference/client-libraries/odbc/#Installation).
diff --git a/docs/zh/10-third-party/05-bi/12-tableau.md b/docs/zh/10-third-party/05-bi/12-tableau.md
index 9fbdc53d45..ace6370dcb 100644
--- a/docs/zh/10-third-party/05-bi/12-tableau.md
+++ b/docs/zh/10-third-party/05-bi/12-tableau.md
@@ -8,7 +8,7 @@ Tableau 是一款知名的商业智能工具，它支持多种数据源，可方
 ## 前置条件
 
 准备以下环境：
-- TDengine 3.3.5.4 以上版本集群已部署并正常运行（企业及社区版均可）
+- TDengine 3.3.5.8 以上版本集群已部署并正常运行（企业及社区版均可）
 - taosAdapter 能够正常运行。详细参考 [taosAdapter 参考手册](../../../reference/components/taosadapter)
 - Tableau 桌面版安装并运行（如未安装，请下载并安装 Windows 操作系统 64 位 [Tableau 桌面版](https://www.tableau.com/products/desktop/download) ）。安装 Tableau 桌面版请参考 [官方文档](https://www.tableau.com)。
 - 从 TDengine 官网下载最新的 Windows 操作系统 X64 客户端驱动程序，并进行安装。详细参考 [安装 ODBC 驱动](../../../reference/connector/odbc/#安装)。
@@ -18,7 +18,11 @@ Tableau 是一款知名的商业智能工具，它支持多种数据源，可方
 
 **第 1 步**，在Windows操作系统的开始菜单中搜索并打开“ODBC数据源（64位）”管理工具并进行配置。详细参考[配置ODBC数据源](../../../reference/connector/odbc/#配置数据源)。
 
-**第 2 步**，在 Windows 系统环境下启动 Tableau，之后在其连接页面中搜索 “ODBC”，并选择 “其他数据库 (ODBC)”。 
+:::tip
+需要注意的是，在为 Tableau 配置 ODBC 数据源时，TDengine ODBC 数据源配置页面中的【数据库】配置项为必填项，需选择一个可成功连接的数据库。
+:::
+
+**第 2 步**，在 Windows 系统环境下启动 Tableau，之后在其连接页面中搜索 “ODBC”，并选择 “其他数据库 (ODBC)”。 对于 Tableau 的使用的ODBC数据源，在其 TDengine ODBC 数据源配置页面的【数据库】的配置项为必填，需要选择可以连接的数据库。
  
 **第 3 步**，点击 `DSN` 单选框，接着选择已配置好的数据源(MyTDengine)，然后点击`连接`按钮。待连接成功后，删除字符串附加部分的内容，最后点击`登录`按钮即可。
 
diff --git a/docs/zh/10-third-party/05-bi/13-excel.md b/docs/zh/10-third-party/05-bi/13-excel.md
index 442d0175b4..bfa6b5b832 100644
--- a/docs/zh/10-third-party/05-bi/13-excel.md
+++ b/docs/zh/10-third-party/05-bi/13-excel.md
@@ -8,7 +8,7 @@ title: 与 Excel 集成
 ## 前置条件
 
 准备以下环境：
-- TDengine 3.3.5.7 以上版本集群已部署并正常运行（企业及社区版均可）。
+- TDengine 3.3.5.8 以上版本集群已部署并正常运行（企业及社区版均可）。
 - taosAdapter 能够正常运行，详细参考 [taosAdapter 参考手册](../../../reference/components/taosadapter)。
 - Excel 安装并运行, 如未安装，请下载并安装, 具体操作请参考 Microsoft 官方文档。
 - 从 TDengine 官网下载最新的 Windows 操作系统 X64 客户端驱动程序并进行安装，详细参考 [安装 ODBC 驱动](../../../reference/connector/odbc/#安装)。

From 3fe9ffe6c87cf46f8c21c019b18d1806459d9b5c Mon Sep 17 00:00:00 2001
From: menshibin <men_shi_bin@163.com>
Date: Mon, 3 Mar 2025 17:26:42 +0800
Subject: [PATCH 20/21] docs: modify tableau configuration instructions error

---
 docs/en/10-third-party/05-bi/12-tableau.md | 6 +++---
 docs/zh/10-third-party/05-bi/12-tableau.md | 4 ++--
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/docs/en/10-third-party/05-bi/12-tableau.md b/docs/en/10-third-party/05-bi/12-tableau.md
index 7a3416d93f..db8a569926 100644
--- a/docs/en/10-third-party/05-bi/12-tableau.md
+++ b/docs/en/10-third-party/05-bi/12-tableau.md
@@ -10,8 +10,8 @@ Tableau is a well-known business intelligence tool that supports multiple data s
 
 Prepare the following environment:
 
-- TDengine 3.3.5.8 and above version is installed and running normally (both Enterprise and Community versions are available)
-- taosAdapter is running normally, refer to [taosAdapter Reference](../../../tdengine-reference/components/taosadapter/)
+- TDengine 3.3.5.8 and above version is installed and running normally (both Enterprise and Community versions are available).
+- taosAdapter is running normally, refer to [taosAdapter Reference](../../../tdengine-reference/components/taosadapter/).
 - Install and run Tableau Desktop (if not installed, please download and install Windows operating system 64-bit [Download Tableau Desktop](https://www.tableau.com/products/desktop/download)). Install Tableau please refer to [Tableau Desktop](https://www.tableau.com).
 - Download the latest Windows operating system X64 client driver from the TDengine official website and install it, refer to [Install ODBC Driver](../../../tdengine-reference/client-libraries/odbc/#Installation).
 
@@ -20,7 +20,7 @@ Prepare the following environment:
 **Step 1**, Search and open the "ODBC Data Source (64 bit)" management tool in the Start menu of the Windows operating system and configure it, refer to [Install ODBC Driver](../../../tdengine-reference/client-libraries/odbc/#Installation).
 
 :::tip
-It should be noted that when configuring the ODBC data source for Tableau, the 【Database】 configuration item on the TDengine ODBC data source configuration page is required. You need to select a database that can be successfully connected.
+It should be noted that when configuring the ODBC data source for Tableau, the [Database] configuration item on the TDengine ODBC data source configuration page is required. You need to select a database that can be successfully connected.
 :::
 
 **Step 2**, Start Tableau in the Windows system environment, then search for "ODBC" on its connection page and select "Other Databases (ODBC)".
diff --git a/docs/zh/10-third-party/05-bi/12-tableau.md b/docs/zh/10-third-party/05-bi/12-tableau.md
index ace6370dcb..ab4f86549b 100644
--- a/docs/zh/10-third-party/05-bi/12-tableau.md
+++ b/docs/zh/10-third-party/05-bi/12-tableau.md
@@ -8,8 +8,8 @@ Tableau 是一款知名的商业智能工具，它支持多种数据源，可方
 ## 前置条件
 
 准备以下环境：
-- TDengine 3.3.5.8 以上版本集群已部署并正常运行（企业及社区版均可）
-- taosAdapter 能够正常运行。详细参考 [taosAdapter 参考手册](../../../reference/components/taosadapter)
+- TDengine 3.3.5.8 以上版本集群已部署并正常运行（企业及社区版均可）。
+- taosAdapter 能够正常运行。详细参考 [taosAdapter 参考手册](../../../reference/components/taosadapter)。
 - Tableau 桌面版安装并运行（如未安装，请下载并安装 Windows 操作系统 64 位 [Tableau 桌面版](https://www.tableau.com/products/desktop/download) ）。安装 Tableau 桌面版请参考 [官方文档](https://www.tableau.com)。
 - 从 TDengine 官网下载最新的 Windows 操作系统 X64 客户端驱动程序，并进行安装。详细参考 [安装 ODBC 驱动](../../../reference/connector/odbc/#安装)。
 

From 8dd668349035e962d90067b95eb144448d9d4357 Mon Sep 17 00:00:00 2001
From: haoranchen <haoran920c@163.com>
Date: Mon, 3 Mar 2025 17:30:44 +0800
Subject: [PATCH 21/21] fix: set dir in packaging dockerfile

---
 packaging/docker/Dockerfile | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/packaging/docker/Dockerfile b/packaging/docker/Dockerfile
index 010823ff08..2f1f5a1787 100644
--- a/packaging/docker/Dockerfile
+++ b/packaging/docker/Dockerfile
@@ -15,9 +15,10 @@ ENV DEBIAN_FRONTEND=noninteractive
 ADD https://github.com/krallin/tini/releases/download/${TINI_VERSION}/tini-${cpuType} /tini
 RUN chmod +x /tini
 
-WORKDIR /root
 RUN tar -zxf /root/${pkgFile} && \
+    cd /root/${dirName}/ && \
     /bin/bash /root/${dirName}/install.sh -e no && \
+    cd /root/ && \
     rm /root/${pkgFile} && \
     rm -rf /root/${dirName} && \
     apt-get update && \