Merge remote-tracking branch 'origin/3.0' into feature/3.0_wxy

include/common/tmsgdef.h

@@ -158,6 +158,7 @@ enum {
   TD_DEF_MSG_TYPE(TDMT_MND_DROP_INDEX, "mnode-drop-index", NULL, NULL)
   TD_DEF_MSG_TYPE(TDMT_MND_GET_DB_CFG, "mnode-get-db-cfg", NULL, NULL)
   TD_DEF_MSG_TYPE(TDMT_MND_GET_INDEX, "mnode-get-index", NULL, NULL)
+  TD_DEF_MSG_TYPE(TDMT_MND_APPLY_MSG, "mnode-apply-msg", NULL, NULL)
 
   // Requests handled by VNODE
   TD_NEW_MSG_SEG(TDMT_VND_MSG)

include/dnode/mnode/mnode.h

@@ -89,6 +89,7 @@ int32_t mndGetLoad(SMnode *pMnode, SMnodeLoad *pLoad);
  * @return int32_t 0 for success, -1 for failure.
  */
 int32_t mndProcessMsg(SRpcMsg *pMsg);
+int32_t mndProcessSyncMsg(SRpcMsg *pMsg);
 
 /**
  * @brief Generate machine code

include/dnode/mnode/sdb/sdb.h

@@ -304,13 +304,16 @@ int32_t sdbGetMaxId(SSdb *pSdb, ESdbType type);
 int64_t sdbGetTableVer(SSdb *pSdb, ESdbType type);
 
 /**
- * @brief Update the version of sdb
+ * @brief Update the index of sdb
  *
  * @param pSdb The sdb object.
- * @param val The update value of the version.
- * @return int32_t The current version of sdb
+ * @param index The update value of the apply index.
+ * @return int32_t The current index of sdb
  */
-int64_t sdbUpdateVer(SSdb *pSdb, int32_t val);
+void    sdbSetApplyIndex(SSdb *pSdb, int64_t index);
+int64_t sdbGetApplyIndex(SSdb *pSdb);
+void    sdbSetApplyTerm(SSdb *pSdb, int64_t term);
+int64_t sdbGetApplyTerm(SSdb *pSdb);
 
 SSdbRaw *sdbAllocRaw(ESdbType type, int8_t sver, int32_t dataLen);
 void     sdbFreeRaw(SSdbRaw *pRaw);
@@ -339,6 +342,7 @@ typedef struct SSdb {
   char          *tmpDir;
   int64_t        lastCommitVer;
   int64_t        curVer;
+  int64_t        curTerm;
   int64_t        tableVer[SDB_MAX];
   int64_t        maxId[SDB_MAX];
   EKeyType       keyTypes[SDB_MAX];

include/libs/sync/sync.h

@@ -78,6 +78,8 @@ typedef struct SFsmCbMeta {
   int32_t    code;
   ESyncState state;
   uint64_t   seqNum;
+  SyncTerm   term;
+  SyncTerm   currentTerm;
 } SFsmCbMeta;
 
 typedef struct SSyncFSM {
@@ -85,6 +87,7 @@ typedef struct SSyncFSM {
   void (*FpCommitCb)(struct SSyncFSM* pFsm, const SRpcMsg* pMsg, SFsmCbMeta cbMeta);
   void (*FpPreCommitCb)(struct SSyncFSM* pFsm, const SRpcMsg* pMsg, SFsmCbMeta cbMeta);
   void (*FpRollBackCb)(struct SSyncFSM* pFsm, const SRpcMsg* pMsg, SFsmCbMeta cbMeta);
+  void (*FpRestoreFinish)(struct SSyncFSM* pFsm);
   int32_t (*FpGetSnapshot)(struct SSyncFSM* pFsm, SSnapshot* pSnapshot);
   int32_t (*FpRestoreSnapshot)(struct SSyncFSM* pFsm, const SSnapshot* snapshot);
 } SSyncFSM;
@@ -117,7 +120,6 @@ typedef struct SSyncLogStore {
 
 } SSyncLogStore;
 
-
 typedef struct SSyncInfo {
   SyncGroupId vgId;
   SSyncCfg    syncCfg;
@@ -144,6 +146,7 @@ int32_t     syncGetVgId(int64_t rid);
 int32_t     syncPropose(int64_t rid, const SRpcMsg* pMsg, bool isWeak);
 bool        syncEnvIsStart();
 const char* syncStr(ESyncState state);
+bool        syncIsRestoreFinish(int64_t rid);
 
 #ifdef __cplusplus
 }

include/libs/transport/trpc.h

@@ -28,7 +28,7 @@ extern "C" {
 #define TAOS_CONN_CLIENT 1
 #define IsReq(pMsg)      (pMsg->msgType & 1U)
 
-extern int tsRpcHeadSize;
+extern int32_t tsRpcHeadSize;
 
 typedef struct {
   uint32_t clientIp;
@@ -69,10 +69,10 @@ typedef struct SRpcInit {
   char     localFqdn[TSDB_FQDN_LEN];
   uint16_t localPort;     // local port
   char *   label;         // for debug purpose
-  int      numOfThreads;  // number of threads to handle connections
-  int      sessions;      // number of sessions allowed
+  int32_t  numOfThreads;  // number of threads to handle connections
+  int32_t  sessions;      // number of sessions allowed
   int8_t   connType;      // TAOS_CONN_UDP, TAOS_CONN_TCPC, TAOS_CONN_TCPS
-  int      idleTime;      // milliseconds, 0 means idle timer is disabled
+  int32_t  idleTime;      // milliseconds, 0 means idle timer is disabled
 
   // the following is for client app ecurity only
   char *user;  // user name
@@ -108,9 +108,9 @@ int32_t rpcInit();
 void    rpcCleanup();
 void *  rpcOpen(const SRpcInit *pRpc);
 void    rpcClose(void *);
-void *  rpcMallocCont(int contLen);
+void *  rpcMallocCont(int32_t contLen);
 void    rpcFreeCont(void *pCont);
-void *  rpcReallocCont(void *ptr, int contLen);
+void *  rpcReallocCont(void *ptr, int32_t contLen);
 
 // Because taosd supports multi-process mode
 // These functions should not be used on the server side
@@ -121,10 +121,10 @@ void rpcRegisterBrokenLinkArg(SRpcMsg *msg);
 void rpcReleaseHandle(void *handle, int8_t type);  // just release client conn to rpc instance, no close sock
 
 // These functions will not be called in the child process
-void rpcSendRedirectRsp(void *pConn, const SEpSet *pEpSet);
-void rpcSendRequestWithCtx(void *thandle, const SEpSet *pEpSet, SRpcMsg *pMsg, int64_t *rid, SRpcCtx *ctx);
-int  rpcGetConnInfo(void *thandle, SRpcConnInfo *pInfo);
-void rpcSendRecv(void *shandle, SEpSet *pEpSet, SRpcMsg *pReq, SRpcMsg *pRsp);
+void    rpcSendRedirectRsp(void *pConn, const SEpSet *pEpSet);
+void    rpcSendRequestWithCtx(void *thandle, const SEpSet *pEpSet, SRpcMsg *pMsg, int64_t *rid, SRpcCtx *ctx);
+int32_t rpcGetConnInfo(void *thandle, SRpcConnInfo *pInfo);
+void    rpcSendRecv(void *shandle, SEpSet *pEpSet, SRpcMsg *pReq, SRpcMsg *pRsp);
 
 #ifdef __cplusplus
 }

source/dnode/mgmt/mgmt_mnode/inc/mmInt.h

@@ -24,19 +24,22 @@ extern "C" {
 #endif
 
 typedef struct SMnodeMgmt {
-  SDnodeData   *pData;
-  SMnode       *pMnode;
-  SMsgCb        msgCb;
-  const char   *path;
-  const char   *name;
-  SSingleWorker queryWorker;
-  SSingleWorker readWorker;
-  SSingleWorker writeWorker;
-  SSingleWorker syncWorker;
-  SSingleWorker monitorWorker;
-  SReplica      replicas[TSDB_MAX_REPLICA];
-  int8_t        replica;
-  int8_t        selfIndex;
+  SDnodeData    *pData;
+  SMnode        *pMnode;
+  SMsgCb         msgCb;
+  const char    *path;
+  const char    *name;
+  SSingleWorker  queryWorker;
+  SSingleWorker  readWorker;
+  SSingleWorker  writeWorker;
+  SSingleWorker  syncWorker;
+  SSingleWorker  monitorWorker;
+  SReplica       replicas[TSDB_MAX_REPLICA];
+  int8_t         replica;
+  int8_t         selfIndex;
+  bool           stopped;
+  int32_t        refCount;
+  TdThreadRwlock lock;
 } SMnodeMgmt;
 
 // mmFile.c
@@ -45,6 +48,8 @@ int32_t mmWriteFile(SMnodeMgmt *pMgmt, SDCreateMnodeReq *pMsg, bool deployed);
 
 // mmInt.c
 int32_t mmAlter(SMnodeMgmt *pMgmt, SDAlterMnodeReq *pMsg);
+int32_t mmAcquire(SMnodeMgmt *pMgmt);
+void    mmRelease(SMnodeMgmt *pMgmt);
 
 // mmHandle.c
 SArray *mmGetMsgHandles();

source/dnode/mgmt/mgmt_mnode/src/mmHandle.c

@@ -237,6 +237,16 @@ SArray *mmGetMsgHandles() {
   if (dmSetMgmtHandle(pArray, TDMT_VND_SYNC_VNODE_RSP, mmPutNodeMsgToWriteQueue, 0) == NULL) goto _OVER;
   if (dmSetMgmtHandle(pArray, TDMT_VND_COMPACT_VNODE_RSP, mmPutNodeMsgToWriteQueue, 0) == NULL) goto _OVER;
 
+  if (dmSetMgmtHandle(pArray, TDMT_VND_SYNC_TIMEOUT, mmPutNodeMsgToSyncQueue, 1) == NULL) goto _OVER;
+  if (dmSetMgmtHandle(pArray, TDMT_VND_SYNC_PING, mmPutNodeMsgToSyncQueue, 1) == NULL) goto _OVER;
+  if (dmSetMgmtHandle(pArray, TDMT_VND_SYNC_PING_REPLY, mmPutNodeMsgToSyncQueue, 1) == NULL) goto _OVER;
+  if (dmSetMgmtHandle(pArray, TDMT_VND_SYNC_CLIENT_REQUEST, mmPutNodeMsgToSyncQueue, 1) == NULL) goto _OVER;
+  if (dmSetMgmtHandle(pArray, TDMT_VND_SYNC_CLIENT_REQUEST_REPLY, mmPutNodeMsgToSyncQueue, 1) == NULL) goto _OVER;
+  if (dmSetMgmtHandle(pArray, TDMT_VND_SYNC_REQUEST_VOTE, mmPutNodeMsgToSyncQueue, 1) == NULL) goto _OVER;
+  if (dmSetMgmtHandle(pArray, TDMT_VND_SYNC_REQUEST_VOTE_REPLY, mmPutNodeMsgToSyncQueue, 1) == NULL) goto _OVER;
+  if (dmSetMgmtHandle(pArray, TDMT_VND_SYNC_APPEND_ENTRIES, mmPutNodeMsgToSyncQueue, 1) == NULL) goto _OVER;
+  if (dmSetMgmtHandle(pArray, TDMT_VND_SYNC_APPEND_ENTRIES_REPLY, mmPutNodeMsgToSyncQueue, 1) == NULL) goto _OVER;
+
   code = 0;
 
 _OVER:

source/dnode/mgmt/mgmt_mnode/src/mmInt.c

@@ -110,6 +110,7 @@ static void mmClose(SMnodeMgmt *pMgmt) {
   if (pMgmt->pMnode != NULL) {
     mmStopWorker(pMgmt);
     mndClose(pMgmt->pMnode);
+    taosThreadRwlockDestroy(&pMgmt->lock);
     pMgmt->pMnode = NULL;
   }
 
@@ -122,6 +123,11 @@ static int32_t mmOpen(SMgmtInputOpt *pInput, SMgmtOutputOpt *pOutput) {
     return -1;
   }
 
+  if (syncInit() != 0) {
+    dError("failed to init sync since %s", terrstr());
+    return -1;
+  }
+
   SMnodeMgmt *pMgmt = taosMemoryCalloc(1, sizeof(SMnodeMgmt));
   if (pMgmt == NULL) {
     terrno = TSDB_CODE_OUT_OF_MEMORY;
@@ -137,6 +143,7 @@ static int32_t mmOpen(SMgmtInputOpt *pInput, SMgmtOutputOpt *pOutput) {
   pMgmt->msgCb.queueFps[WRITE_QUEUE] = (PutToQueueFp)mmPutRpcMsgToWriteQueue;
   pMgmt->msgCb.queueFps[SYNC_QUEUE] = (PutToQueueFp)mmPutRpcMsgToSyncQueue;
   pMgmt->msgCb.mgmt = pMgmt;
+  taosThreadRwlockInit(&pMgmt->lock, NULL);
 
   bool deployed = false;
   if (mmReadFile(pMgmt, &deployed) != 0) {
@@ -206,3 +213,22 @@ SMgmtFunc mmGetMgmtFunc() {
 
   return mgmtFunc;
 }
+
+int32_t mmAcquire(SMnodeMgmt *pMgmt) {
+  int32_t code = 0;
+
+  taosThreadRwlockRdlock(&pMgmt->lock);
+  if (pMgmt->stopped) {
+    code = -1;
+  } else {
+    atomic_add_fetch_32(&pMgmt->refCount, 1);
+  }
+  taosThreadRwlockUnlock(&pMgmt->lock);
+  return code;
+}
+
+void mmRelease(SMnodeMgmt *pMgmt) {
+  taosThreadRwlockRdlock(&pMgmt->lock);
+  atomic_sub_fetch_32(&pMgmt->refCount, 1);
+  taosThreadRwlockUnlock(&pMgmt->lock);
+}
\ No newline at end of file

source/dnode/mgmt/mgmt_mnode/src/mmWorker.c

@@ -56,6 +56,12 @@ static void mmProcessQueue(SQueueInfo *pInfo, SRpcMsg *pMsg) {
   taosFreeQitem(pMsg);
 }
 
+static void mmProcessSyncQueue(SQueueInfo *pInfo, SRpcMsg *pMsg) {
+  SMnodeMgmt *pMgmt = pInfo->ahandle;
+  pMsg->info.node = pMgmt->pMnode;
+  mndProcessSyncMsg(pMsg);
+}
+
 static int32_t mmPutNodeMsgToWorker(SSingleWorker *pWorker, SRpcMsg *pMsg) {
   dTrace("msg:%p, put into worker %s, type:%s", pMsg, pWorker->name, TMSG_INFO(pMsg->msgType));
   taosWriteQitem(pWorker->queue, pMsg);
@@ -105,7 +111,17 @@ int32_t mmPutRpcMsgToReadQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg) {
 }
 
 int32_t mmPutRpcMsgToSyncQueue(SMnodeMgmt *pMgmt, SRpcMsg *pMsg) {
-  return mmPutRpcMsgToWorker(&pMgmt->syncWorker, pMsg);
+  int32_t code = -1;
+  if (mmAcquire(pMgmt) == 0) {
+    code = mmPutRpcMsgToWorker(&pMgmt->syncWorker, pMsg);
+    mmRelease(pMgmt);
+  }
+
+  if (code != 0) {
+    rpcFreeCont(pMsg->pCont);
+    pMsg->pCont = NULL;
+  }
+  return code;
 }
 
 int32_t mmStartWorker(SMnodeMgmt *pMgmt) {
@@ -149,7 +165,7 @@ int32_t mmStartWorker(SMnodeMgmt *pMgmt) {
       .min = 1,
       .max = 1,
       .name = "mnode-sync",
-      .fp = (FItem)mmProcessQueue,
+      .fp = (FItem)mmProcessSyncQueue,
       .param = pMgmt,
   };
   if (tSingleWorkerInit(&pMgmt->syncWorker, &sCfg) != 0) {
@@ -174,6 +190,11 @@ int32_t mmStartWorker(SMnodeMgmt *pMgmt) {
 }
 
 void mmStopWorker(SMnodeMgmt *pMgmt) {
+  taosThreadRwlockWrlock(&pMgmt->lock);
+  pMgmt->stopped = 1;
+  taosThreadRwlockUnlock(&pMgmt->lock);
+  while (pMgmt->refCount > 0) taosMsleep(10);
+
   tSingleWorkerCleanup(&pMgmt->monitorWorker);
   tSingleWorkerCleanup(&pMgmt->queryWorker);
   tSingleWorkerCleanup(&pMgmt->readWorker);

source/dnode/mgmt/test/mnode/CMakeLists.txt

@@ -4,7 +4,7 @@ target_link_libraries(
     dmnodeTest sut
 )
 
-add_test(
-    NAME dmnodeTest
-    COMMAND dmnodeTest
-)
+#add_test(
+#    NAME dmnodeTest
+#    COMMAND dmnodeTest
+#)

source/dnode/mnode/impl/inc/mndInt.h

@@ -19,6 +19,7 @@
 #include "mndDef.h"
 
 #include "sdb.h"
+#include "syncTools.h"
 #include "tcache.h"
 #include "tdatablock.h"
 #include "tglobal.h"
@@ -31,12 +32,14 @@
 extern "C" {
 #endif
 
+// clang-format off
 #define mFatal(...) { if (mDebugFlag & DEBUG_FATAL) { taosPrintLog("MND FATAL ", DEBUG_FATAL, 255, __VA_ARGS__); }}
 #define mError(...) { if (mDebugFlag & DEBUG_ERROR) { taosPrintLog("MND ERROR ", DEBUG_ERROR, 255, __VA_ARGS__); }}
 #define mWarn(...)  { if (mDebugFlag & DEBUG_WARN)  { taosPrintLog("MND WARN ", DEBUG_WARN, 255, __VA_ARGS__); }}
 #define mInfo(...)  { if (mDebugFlag & DEBUG_INFO)  { taosPrintLog("MND ", DEBUG_INFO, 255, __VA_ARGS__); }}
 #define mDebug(...) { if (mDebugFlag & DEBUG_DEBUG) { taosPrintLog("MND ", DEBUG_DEBUG, mDebugFlag, __VA_ARGS__); }}
 #define mTrace(...) { if (mDebugFlag & DEBUG_TRACE) { taosPrintLog("MND ", DEBUG_TRACE, mDebugFlag, __VA_ARGS__); }}
+// clang-format on
 
 #define SYSTABLE_SCH_TABLE_NAME_LEN ((TSDB_TABLE_NAME_LEN - 1) + VARSTR_HEADER_SIZE)
 #define SYSTABLE_SCH_DB_NAME_LEN    ((TSDB_DB_NAME_LEN - 1) + VARSTR_HEADER_SIZE)
@@ -72,10 +75,11 @@ typedef struct {
 } STelemMgmt;
 
 typedef struct {
+  SWal      *pWal;
   int32_t    errCode;
+  bool       restored;
   sem_t      syncSem;
-  SWal      *pWal;
-  SSyncNode *pSyncNode;
+  int64_t    sync;
   ESyncState state;
 } SSyncMgmt;
 

source/dnode/mnode/impl/inc/mndSync.h

@@ -26,6 +26,8 @@ int32_t mndInitSync(SMnode *pMnode);
 void    mndCleanupSync(SMnode *pMnode);
 bool    mndIsMaster(SMnode *pMnode);
 int32_t mndSyncPropose(SMnode *pMnode, SSdbRaw *pRaw);
+void    mndSyncStart(SMnode *pMnode);
+void    mndSyncStop(SMnode *pMnode);
 
 #ifdef __cplusplus
 }

source/dnode/mnode/impl/src/mndSync.c

@@ -17,178 +17,146 @@
 #include "mndSync.h"
 #include "mndTrans.h"
 
-static int32_t mndInitWal(SMnode *pMnode) {
-  SSyncMgmt *pMgmt = &pMnode->syncMgmt;
+int32_t mndSyncEqMsg(const SMsgCb *msgcb, SRpcMsg *pMsg) { return tmsgPutToQueue(msgcb, SYNC_QUEUE, pMsg); }
 
-  char path[PATH_MAX] = {0};
-  snprintf(path, sizeof(path), "%s%swal", pMnode->path, TD_DIRSEP);
-  SWalCfg cfg = {
-      .vgId = 1,
-      .fsyncPeriod = 0,
-      .rollPeriod = -1,
-      .segSize = -1,
-      .retentionPeriod = -1,
-      .retentionSize = -1,
-      .level = TAOS_WAL_FSYNC,
-  };
-  pMgmt->pWal = walOpen(path, &cfg);
-  if (pMgmt->pWal == NULL) return -1;
-
-  return 0;
-}
+int32_t mndSyncSendMsg(const SEpSet *pEpSet, SRpcMsg *pMsg) { return tmsgSendReq(pEpSet, pMsg); }
 
-static void mndCloseWal(SMnode *pMnode) {
+void mndSyncCommitMsg(struct SSyncFSM *pFsm, const SRpcMsg *pMsg, SFsmCbMeta cbMeta) {
+  SMnode    *pMnode = pFsm->data;
+  SSdb      *pSdb = pMnode->pSdb;
   SSyncMgmt *pMgmt = &pMnode->syncMgmt;
-  if (pMgmt->pWal != NULL) {
-    walClose(pMgmt->pWal);
-    pMgmt->pWal = NULL;
+  SSdbRaw   *pRaw = pMsg->pCont;
+
+  mTrace("raw:%p, apply to sdb, ver:%" PRId64 " role:%s", pRaw, cbMeta.index, syncStr(cbMeta.state));
+  sdbWriteWithoutFree(pSdb, pRaw);
+  sdbSetApplyIndex(pSdb, cbMeta.index);
+  sdbSetApplyTerm(pSdb, cbMeta.term);
+  if (cbMeta.state == TAOS_SYNC_STATE_LEADER) {
+    tsem_post(&pMgmt->syncSem);
   }
 }
 
-static int32_t mndRestoreWal(SMnode *pMnode) {
-  SWal   *pWal = pMnode->syncMgmt.pWal;
-  SSdb   *pSdb = pMnode->pSdb;
-  int64_t lastSdbVer = sdbUpdateVer(pSdb, 0);
-  int32_t code = -1;
-
-  SWalReadHandle *pHandle = walOpenReadHandle(pWal);
-  if (pHandle == NULL) return -1;
-
-  int64_t first = walGetFirstVer(pWal);
-  int64_t last = walGetLastVer(pWal);
-  mDebug("start to restore wal, sdbver:%" PRId64 ", first:%" PRId64 " last:%" PRId64, lastSdbVer, first, last);
-
-  first = TMAX(lastSdbVer + 1, first);
-  for (int64_t ver = first; ver >= 0 && ver <= last; ++ver) {
-    if (walReadWithHandle(pHandle, ver) < 0) {
-      mError("ver:%" PRId64 ", failed to read from wal since %s", ver, terrstr());
-      goto _OVER;
-    }
-
-    SWalHead *pHead = pHandle->pHead;
-    int64_t   sdbVer = sdbUpdateVer(pSdb, 0);
-    if (sdbVer + 1 != ver) {
-      terrno = TSDB_CODE_SDB_INVALID_WAl_VER;
-      mError("ver:%" PRId64 ", failed to write to sdb, since inconsistent with sdbver:%" PRId64, ver, sdbVer);
-      goto _OVER;
-    }
-
-    mTrace("ver:%" PRId64 ", will be restored, content:%p", ver, pHead->head.body);
-    if (sdbWriteWithoutFree(pSdb, (void *)pHead->head.body) < 0) {
-      mError("ver:%" PRId64 ", failed to write to sdb since %s", ver, terrstr());
-      goto _OVER;
-    }
-
-    sdbUpdateVer(pSdb, 1);
-    mDebug("ver:%" PRId64 ", is restored", ver);
-  }
-
-  int64_t sdbVer = sdbUpdateVer(pSdb, 0);
-  mDebug("restore wal finished, sdbver:%" PRId64, sdbVer);
+int32_t mndSyncGetSnapshot(struct SSyncFSM *pFsm, SSnapshot *pSnapshot) {
+  SMnode *pMnode = pFsm->data;
+  pSnapshot->lastApplyIndex = sdbGetApplyIndex(pMnode->pSdb);
+  pSnapshot->lastApplyTerm = sdbGetApplyTerm(pMnode->pSdb);
+  return 0;
+}
 
+void mndRestoreFinish(struct SSyncFSM *pFsm) {
+  SMnode *pMnode = pFsm->data;
   mndTransPullup(pMnode);
-  sdbVer = sdbUpdateVer(pSdb, 0);
-  mDebug("pullup trans finished, sdbver:%" PRId64, sdbVer);
-
-  if (sdbVer != lastSdbVer) {
-    mInfo("sdb restored from %" PRId64 " to %" PRId64 ", write file", lastSdbVer, sdbVer);
-    if (sdbWriteFile(pSdb) != 0) {
-      goto _OVER;
-    }
-
-    if (walCommit(pWal, sdbVer) != 0) {
-      goto _OVER;
-    }
-
-    if (walBeginSnapshot(pWal, sdbVer) < 0) {
-      goto _OVER;
-    }
-
-    if (walEndSnapshot(pWal) < 0) {
-      goto _OVER;
-    }
-  }
-
-  code = 0;
+  pMnode->syncMgmt.restored = true;
+}
 
-_OVER:
-  walCloseReadHandle(pHandle);
-  return code;
+SSyncFSM *mndSyncMakeFsm(SMnode *pMnode) {
+  SSyncFSM *pFsm = taosMemoryCalloc(1, sizeof(SSyncFSM));
+  pFsm->data = pMnode;
+  pFsm->FpCommitCb = mndSyncCommitMsg;
+  pFsm->FpPreCommitCb = NULL;
+  pFsm->FpRollBackCb = NULL;
+  pFsm->FpGetSnapshot = mndSyncGetSnapshot;
+  pFsm->FpRestoreFinish = mndRestoreFinish;
+  pFsm->FpRestoreSnapshot = NULL;
+  return pFsm;
 }
 
 int32_t mndInitSync(SMnode *pMnode) {
   SSyncMgmt *pMgmt = &pMnode->syncMgmt;
-  tsem_init(&pMgmt->syncSem, 0, 0);
 
-  if (mndInitWal(pMnode) < 0) {
+  char path[PATH_MAX + 20] = {0};
+  snprintf(path, sizeof(path), "%s%swal", pMnode->path, TD_DIRSEP);
+  SWalCfg cfg = {
+      .vgId = 1,
+      .fsyncPeriod = 0,
+      .rollPeriod = -1,
+      .segSize = -1,
+      .retentionPeriod = -1,
+      .retentionSize = -1,
+      .level = TAOS_WAL_FSYNC,
+  };
+
+  pMgmt->pWal = walOpen(path, &cfg);
+  if (pMgmt->pWal == NULL) {
     mError("failed to open wal since %s", terrstr());
     return -1;
   }
 
-  if (mndRestoreWal(pMnode) < 0) {
-    mError("failed to restore wal since %s", terrstr());
-    return -1;
+  SSyncInfo syncInfo = {.vgId = 1, .FpSendMsg = mndSyncSendMsg, .FpEqMsg = mndSyncEqMsg};
+  snprintf(syncInfo.path, sizeof(syncInfo.path), "%s%ssync", pMnode->path, TD_DIRSEP);
+  syncInfo.pWal = pMgmt->pWal;
+  syncInfo.pFsm = mndSyncMakeFsm(pMnode);
+
+  SSyncCfg *pCfg = &syncInfo.syncCfg;
+  pCfg->replicaNum = pMnode->replica;
+  pCfg->myIndex = pMnode->selfIndex;
+  for (int32_t i = 0; i < pMnode->replica; ++i) {
+    SNodeInfo *pNode = &pCfg->nodeInfo[i];
+    tstrncpy(pNode->nodeFqdn, pMnode->replicas[i].fqdn, sizeof(pNode->nodeFqdn));
+    pNode->nodePort = pMnode->replicas[i].port;
   }
 
-  if (pMnode->selfId == 1) {
-    pMgmt->state = TAOS_SYNC_STATE_LEADER;
+  tsem_init(&pMgmt->syncSem, 0, 0);
+  pMgmt->sync = syncOpen(&syncInfo);
+  if (pMgmt->sync <= 0) {
+    mError("failed to open sync since %s", terrstr());
+    return -1;
   }
-  pMgmt->pSyncNode = NULL;
+
+  mDebug("mnode sync is opened, id:%" PRId64, pMgmt->sync);
   return 0;
 }
 
 void mndCleanupSync(SMnode *pMnode) {
   SSyncMgmt *pMgmt = &pMnode->syncMgmt;
-  tsem_destroy(&pMgmt->syncSem);
-  mndCloseWal(pMnode);
-}
-
-static int32_t mndSyncApplyCb(struct SSyncFSM *fsm, SyncIndex index, const SSyncBuffer *buf, void *pData) {
-  SMnode    *pMnode = pData;
-  SSyncMgmt *pMgmt = &pMnode->syncMgmt;
+  syncStop(pMgmt->sync);
+  mDebug("sync:%" PRId64 " is stopped", pMgmt->sync);
 
-  pMgmt->errCode = 0;
-  tsem_post(&pMgmt->syncSem);
+  tsem_destroy(&pMgmt->syncSem);
+  if (pMgmt->pWal != NULL) {
+    walClose(pMgmt->pWal);
+  }
 
-  return 0;
+  memset(pMgmt, 0, sizeof(SSyncMgmt));
 }
 
 int32_t mndSyncPropose(SMnode *pMnode, SSdbRaw *pRaw) {
-  SWal *pWal = pMnode->syncMgmt.pWal;
-  SSdb *pSdb = pMnode->pSdb;
-
-  int64_t ver = sdbUpdateVer(pSdb, 1);
-  if (walWrite(pWal, ver, 1, pRaw, sdbGetRawTotalSize(pRaw)) < 0) {
-    sdbUpdateVer(pSdb, -1);
-    mError("ver:%" PRId64 ", failed to write raw:%p to wal since %s", ver, pRaw, terrstr());
-    return -1;
-  }
-
-  mTrace("ver:%" PRId64 ", write to wal, raw:%p", ver, pRaw);
-  walCommit(pWal, ver);
-  walFsync(pWal, true);
-
-#if 1
-  return 0;
-#else
-  if (pMnode->replica == 1) return 0;
-
   SSyncMgmt *pMgmt = &pMnode->syncMgmt;
   pMgmt->errCode = 0;
 
-  SSyncBuffer buf = {.data = pRaw, .len = sdbGetRawTotalSize(pRaw)};
-
-  bool    isWeak = false;
-  int32_t code = syncPropose(pMgmt->pSyncNode, &buf, pMnode, isWeak);
+  SRpcMsg rsp = {.code = TDMT_MND_APPLY_MSG, .contLen = sdbGetRawTotalSize(pRaw)};
+  rsp.pCont = rpcMallocCont(rsp.contLen);
+  if (rsp.pCont == NULL) return -1;
+  memcpy(rsp.pCont, pRaw, rsp.contLen);
+
+  const bool isWeak = false;
+  int32_t    code = syncPropose(pMgmt->sync, &rsp, isWeak);
+  if (code == 0) {
+    tsem_wait(&pMgmt->syncSem);
+  } else if (code == TAOS_SYNC_PROPOSE_NOT_LEADER) {
+    terrno = TSDB_CODE_APP_NOT_READY;
+  } else if (code == TAOS_SYNC_PROPOSE_OTHER_ERROR) {
+    terrno = TSDB_CODE_SYN_INTERNAL_ERROR;
+  } else {
+    terrno = TSDB_CODE_APP_ERROR;
+  }
 
   if (code != 0) return code;
-
-  tsem_wait(&pMgmt->syncSem);
   return pMgmt->errCode;
-#endif
 }
 
+void mndSyncStart(SMnode *pMnode) {
+  SSyncMgmt *pMgmt = &pMnode->syncMgmt;
+  syncSetMsgCb(pMgmt->sync, &pMnode->msgCb);
+  syncStart(pMgmt->sync);
+  mDebug("sync:%" PRId64 " is started", pMgmt->sync);
+}
+
+void mndSyncStop(SMnode *pMnode) {}
+
 bool mndIsMaster(SMnode *pMnode) {
   SSyncMgmt *pMgmt = &pMnode->syncMgmt;
-  return pMgmt->state == TAOS_SYNC_STATE_LEADER;
+  pMgmt->state = syncGetMyRole(pMgmt->sync);
+
+  return (pMgmt->state == TAOS_SYNC_STATE_LEADER) && (pMnode->syncMgmt.restored);
 }

source/dnode/mnode/impl/src/mndTrans.c

@@ -682,13 +682,6 @@ static int32_t mndTransSync(SMnode *pMnode, STrans *pTrans) {
   }
 
   mDebug("trans:%d, sync finished", pTrans->id);
-
-  code = sdbWrite(pMnode->pSdb, pRaw);
-  if (code != 0) {
-    mError("trans:%d, failed to write sdb since %s", pTrans->id, terrstr());
-    return -1;
-  }
-
   return 0;
 }
 
@@ -1360,19 +1353,35 @@ _OVER:
   return code;
 }
 
+static int32_t mndCompareTransId(int32_t *pTransId1, int32_t *pTransId2) { return *pTransId1 >= *pTransId2 ? 1 : 0; }
+
 void mndTransPullup(SMnode *pMnode) {
-  STrans *pTrans = NULL;
-  void   *pIter = NULL;
+  SSdb   *pSdb = pMnode->pSdb;
+  SArray *pArray = taosArrayInit(sdbGetSize(pSdb, SDB_TRANS), sizeof(int32_t));
+  if (pArray == NULL) return;
 
+  void *pIter = NULL;
   while (1) {
+    STrans *pTrans = NULL;
     pIter = sdbFetch(pMnode->pSdb, SDB_TRANS, pIter, (void **)&pTrans);
     if (pIter == NULL) break;
+    taosArrayPush(pArray, &pTrans->id);
+    sdbRelease(pSdb, pTrans);
+  }
 
-    mndTransExecute(pMnode, pTrans);
-    sdbRelease(pMnode->pSdb, pTrans);
+  taosArraySort(pArray, (__compar_fn_t)mndCompareTransId);
+
+  for (int32_t i = 0; i < taosArrayGetSize(pArray); ++i) {
+    int32_t *pTransId = taosArrayGet(pArray, i);
+    STrans  *pTrans = mndAcquireTrans(pMnode, *pTransId);
+    if (pTrans != NULL) {
+      mndTransExecute(pMnode, pTrans);
+    }
+    mndReleaseTrans(pMnode, pTrans);
   }
 
   sdbWriteFile(pMnode->pSdb);
+  taosArrayDestroy(pArray);
 }
 
 static int32_t mndRetrieveTrans(SRpcMsg *pReq, SShowObj *pShow, SSDataBlock *pBlock, int32_t rows) {

source/dnode/mnode/impl/src/mnode.c

@@ -335,9 +335,107 @@ int32_t mndAlter(SMnode *pMnode, const SMnodeOpt *pOption) {
   return 0;
 }
 
-int32_t mndStart(SMnode *pMnode) { return mndInitTimer(pMnode); }
+int32_t mndStart(SMnode *pMnode) {
+  mndSyncStart(pMnode);
+  return mndInitTimer(pMnode);
+}
+
+void mndStop(SMnode *pMnode) {
+  mndSyncStop(pMnode);
+  return mndCleanupTimer(pMnode);
+}
+
+int32_t mndProcessSyncMsg(SRpcMsg *pMsg) {
+  SMnode *pMnode = pMsg->info.node;
+  void   *ahandle = pMsg->info.ahandle;
+  int32_t ret = TAOS_SYNC_PROPOSE_OTHER_ERROR;
+
+  if (syncEnvIsStart()) {
+    SSyncNode *pSyncNode = syncNodeAcquire(pMnode->syncMgmt.sync);
+    assert(pSyncNode != NULL);
+
+    ESyncState state = syncGetMyRole(pMnode->syncMgmt.sync);
+    SyncTerm   currentTerm = syncGetMyTerm(pMnode->syncMgmt.sync);
+
+    SMsgHead *pHead = pMsg->pCont;
+
+    char  logBuf[512];
+    char *syncNodeStr = sync2SimpleStr(pMnode->syncMgmt.sync);
+    snprintf(logBuf, sizeof(logBuf), "==vnodeProcessSyncReq== msgType:%d, syncNode: %s", pMsg->msgType, syncNodeStr);
+    syncRpcMsgLog2(logBuf, pMsg);
+    taosMemoryFree(syncNodeStr);
+
+    SRpcMsg *pRpcMsg = pMsg;
+
+    if (pRpcMsg->msgType == TDMT_VND_SYNC_TIMEOUT) {
+      SyncTimeout *pSyncMsg = syncTimeoutFromRpcMsg2(pRpcMsg);
+      assert(pSyncMsg != NULL);
+
+      ret = syncNodeOnTimeoutCb(pSyncNode, pSyncMsg);
+      syncTimeoutDestroy(pSyncMsg);
+
+    } else if (pRpcMsg->msgType == TDMT_VND_SYNC_PING) {
+      SyncPing *pSyncMsg = syncPingFromRpcMsg2(pRpcMsg);
+      assert(pSyncMsg != NULL);
+
+      ret = syncNodeOnPingCb(pSyncNode, pSyncMsg);
+      syncPingDestroy(pSyncMsg);
+
+    } else if (pRpcMsg->msgType == TDMT_VND_SYNC_PING_REPLY) {
+      SyncPingReply *pSyncMsg = syncPingReplyFromRpcMsg2(pRpcMsg);
+      assert(pSyncMsg != NULL);
+
+      ret = syncNodeOnPingReplyCb(pSyncNode, pSyncMsg);
+      syncPingReplyDestroy(pSyncMsg);
 
-void mndStop(SMnode *pMnode) { return mndCleanupTimer(pMnode); }
+    } else if (pRpcMsg->msgType == TDMT_VND_SYNC_CLIENT_REQUEST) {
+      SyncClientRequest *pSyncMsg = syncClientRequestFromRpcMsg2(pRpcMsg);
+      assert(pSyncMsg != NULL);
+
+      ret = syncNodeOnClientRequestCb(pSyncNode, pSyncMsg);
+      syncClientRequestDestroy(pSyncMsg);
+
+    } else if (pRpcMsg->msgType == TDMT_VND_SYNC_REQUEST_VOTE) {
+      SyncRequestVote *pSyncMsg = syncRequestVoteFromRpcMsg2(pRpcMsg);
+      assert(pSyncMsg != NULL);
+
+      ret = syncNodeOnRequestVoteCb(pSyncNode, pSyncMsg);
+      syncRequestVoteDestroy(pSyncMsg);
+
+    } else if (pRpcMsg->msgType == TDMT_VND_SYNC_REQUEST_VOTE_REPLY) {
+      SyncRequestVoteReply *pSyncMsg = syncRequestVoteReplyFromRpcMsg2(pRpcMsg);
+      assert(pSyncMsg != NULL);
+
+      ret = syncNodeOnRequestVoteReplyCb(pSyncNode, pSyncMsg);
+      syncRequestVoteReplyDestroy(pSyncMsg);
+
+    } else if (pRpcMsg->msgType == TDMT_VND_SYNC_APPEND_ENTRIES) {
+      SyncAppendEntries *pSyncMsg = syncAppendEntriesFromRpcMsg2(pRpcMsg);
+      assert(pSyncMsg != NULL);
+
+      ret = syncNodeOnAppendEntriesCb(pSyncNode, pSyncMsg);
+      syncAppendEntriesDestroy(pSyncMsg);
+
+    } else if (pRpcMsg->msgType == TDMT_VND_SYNC_APPEND_ENTRIES_REPLY) {
+      SyncAppendEntriesReply *pSyncMsg = syncAppendEntriesReplyFromRpcMsg2(pRpcMsg);
+      assert(pSyncMsg != NULL);
+
+      ret = syncNodeOnAppendEntriesReplyCb(pSyncNode, pSyncMsg);
+      syncAppendEntriesReplyDestroy(pSyncMsg);
+
+    } else {
+      mError("==mndProcessSyncMsg== error msg type:%d", pRpcMsg->msgType);
+      ret = TAOS_SYNC_PROPOSE_OTHER_ERROR;
+    }
+
+    syncNodeRelease(pSyncNode);
+  } else {
+    mError("==mndProcessSyncMsg== error syncEnv stop");
+    ret = TAOS_SYNC_PROPOSE_OTHER_ERROR;
+  }
+
+  return ret;
+}
 
 int32_t mndProcessMsg(SRpcMsg *pMsg) {
   SMnode *pMnode = pMsg->info.node;

source/dnode/mnode/impl/test/sdb/sdbTest.cpp

@@ -493,9 +493,8 @@ TEST_F(MndTestSdb, 01_Write_Str) {
   ASSERT_EQ(sdbGetSize(pSdb, SDB_USER), 2);
   ASSERT_EQ(sdbGetMaxId(pSdb, SDB_USER), -1);
   ASSERT_EQ(sdbGetTableVer(pSdb, SDB_USER), 2 );
-  ASSERT_EQ(sdbUpdateVer(pSdb, 0), -1);
-  ASSERT_EQ(sdbUpdateVer(pSdb, 1), 0);
-  ASSERT_EQ(sdbUpdateVer(pSdb, -1), -1);
+  sdbSetApplyIndex(pSdb, -1);
+  ASSERT_EQ(sdbGetApplyIndex(pSdb), -1);
   ASSERT_EQ(mnode.insertTimes, 2);
   ASSERT_EQ(mnode.deleteTimes, 0);
 
@@ -537,9 +536,6 @@ TEST_F(MndTestSdb, 01_Write_Str) {
 
     ASSERT_EQ(sdbGetSize(pSdb, SDB_USER), 3);
     ASSERT_EQ(sdbGetTableVer(pSdb, SDB_USER), 4);
-    ASSERT_EQ(sdbUpdateVer(pSdb, 0), -1);
-    ASSERT_EQ(sdbUpdateVer(pSdb, 1), 0);
-    ASSERT_EQ(sdbUpdateVer(pSdb, -1), -1);
     ASSERT_EQ(mnode.insertTimes, 3);
     ASSERT_EQ(mnode.deleteTimes, 0);
 
@@ -704,8 +700,9 @@ TEST_F(MndTestSdb, 01_Write_Str) {
   }
 
   // write version
-  ASSERT_EQ(sdbUpdateVer(pSdb, 1), 0);
-  ASSERT_EQ(sdbUpdateVer(pSdb, 1), 1);
+  sdbSetApplyIndex(pSdb, 0);
+  sdbSetApplyIndex(pSdb, 1);
+  ASSERT_EQ(sdbGetApplyIndex(pSdb), 1);
   ASSERT_EQ(sdbWriteFile(pSdb), 0);
   ASSERT_EQ(sdbWriteFile(pSdb), 0);
 
@@ -775,7 +772,7 @@ TEST_F(MndTestSdb, 01_Read_Str) {
   ASSERT_EQ(sdbGetSize(pSdb, SDB_USER), 2);
   ASSERT_EQ(sdbGetMaxId(pSdb, SDB_USER), -1);
   ASSERT_EQ(sdbGetTableVer(pSdb, SDB_USER), 5);
-  ASSERT_EQ(sdbUpdateVer(pSdb, 0), 1);
+  ASSERT_EQ(sdbGetApplyIndex(pSdb), 1);
   ASSERT_EQ(mnode.insertTimes, 4);
   ASSERT_EQ(mnode.deleteTimes, 0);
 

source/dnode/mnode/impl/test/trans/CMakeLists.txt

@@ -31,7 +31,7 @@ target_include_directories(
     PUBLIC "${TD_SOURCE_DIR}/include/dnode/mnode"
     PRIVATE "${CMAKE_CURRENT_SOURCE_DIR}/../../inc"
 )
-# add_test(
-#     NAME transTest2
-#     COMMAND transTest2
-# )
+#add_test(
+#    NAME transTest2
+#    COMMAND transTest2
+#)

source/dnode/mnode/impl/test/trans/trans2.cpp

@@ -23,6 +23,11 @@ int32_t sendReq(const SEpSet *pEpSet, SRpcMsg *pMsg) {
   return -1;
 }
 
+int32_t putToQueue(void *pMgmt, SRpcMsg *pMsg) {
+  terrno = TSDB_CODE_INVALID_PTR;
+  return -1;
+}
+
 class MndTestTrans2 : public ::testing::Test {
  protected:
   static void InitLog() {
@@ -55,6 +60,9 @@ class MndTestTrans2 : public ::testing::Test {
     msgCb.reportStartupFp = reportStartup;
     msgCb.sendReqFp = sendReq;
     msgCb.sendRspFp = sendRsp;
+    msgCb.queueFps[SYNC_QUEUE] = putToQueue;
+    msgCb.queueFps[WRITE_QUEUE] = putToQueue;
+     msgCb.queueFps[READ_QUEUE] = putToQueue;
     msgCb.mgmt = (SMgmtWrapper *)(&msgCb);  // hack
     tmsgSetDefault(&msgCb);
 
@@ -77,6 +85,7 @@ class MndTestTrans2 : public ::testing::Test {
   static void SetUpTestSuite() {
     InitLog();
     walInit();
+    syncInit();
     InitMnode();
   }
 

source/dnode/mnode/sdb/src/sdb.c

@@ -31,11 +31,9 @@ SSdb *sdbInit(SSdbOpt *pOption) {
   char path[PATH_MAX + 100] = {0};
   snprintf(path, sizeof(path), "%s%sdata", pOption->path, TD_DIRSEP);
   pSdb->currDir = strdup(path);
-  snprintf(path, sizeof(path), "%s%ssync", pOption->path, TD_DIRSEP);
-  pSdb->syncDir = strdup(path);
   snprintf(path, sizeof(path), "%s%stmp", pOption->path, TD_DIRSEP);
   pSdb->tmpDir = strdup(path);
-  if (pSdb->currDir == NULL || pSdb->currDir == NULL || pSdb->currDir == NULL) {
+  if (pSdb->currDir == NULL || pSdb->tmpDir == NULL) {
     sdbCleanup(pSdb);
     terrno = TSDB_CODE_OUT_OF_MEMORY;
     mError("failed to init sdb since %s", terrstr());
@@ -55,6 +53,7 @@ SSdb *sdbInit(SSdbOpt *pOption) {
   }
 
   pSdb->curVer = -1;
+  pSdb->curTerm = -1;
   pSdb->lastCommitVer = -1;
   pSdb->pMnode = pOption->pMnode;
   mDebug("sdb init successfully");
@@ -149,12 +148,6 @@ static int32_t sdbCreateDir(SSdb *pSdb) {
     return -1;
   }
 
-  if (taosMkDir(pSdb->syncDir) != 0) {
-    terrno = TAOS_SYSTEM_ERROR(errno);
-    mError("failed to create dir:%s since %s", pSdb->syncDir, terrstr());
-    return -1;
-  }
-
   if (taosMkDir(pSdb->tmpDir) != 0) {
     terrno = TAOS_SYSTEM_ERROR(errno);
     mError("failed to create dir:%s since %s", pSdb->tmpDir, terrstr());
@@ -164,4 +157,10 @@ static int32_t sdbCreateDir(SSdb *pSdb) {
   return 0;
 }
 
-int64_t sdbUpdateVer(SSdb *pSdb, int32_t val) { return atomic_add_fetch_64(&pSdb->curVer, val); }
+void sdbSetApplyIndex(SSdb *pSdb, int64_t index) { pSdb->curVer = index; }
\ No newline at end of file
+
+int64_t sdbGetApplyIndex(SSdb *pSdb) { return pSdb->curVer; }
+
+void sdbSetApplyTerm(SSdb *pSdb, int64_t term) { pSdb->curTerm = term; }
+
+int64_t sdbGetApplyTerm(SSdb *pSdb) { return pSdb->curTerm; }

source/dnode/mnode/sdb/src/sdbFile.c

@@ -65,6 +65,16 @@ static int32_t sdbReadFileHead(SSdb *pSdb, TdFilePtr pFile) {
     return -1;
   }
 
+  ret = taosReadFile(pFile, &pSdb->curTerm, sizeof(int64_t));
+  if (ret < 0) {
+    terrno = TAOS_SYSTEM_ERROR(errno);
+    return -1;
+  }
+  if (ret != sizeof(int64_t)) {
+    terrno = TSDB_CODE_FILE_CORRUPTED;
+    return -1;
+  }
+
   for (int32_t i = 0; i < SDB_TABLE_SIZE; ++i) {
     int64_t maxId = 0;
     ret = taosReadFile(pFile, &maxId, sizeof(int64_t));
@@ -123,6 +133,11 @@ static int32_t sdbWriteFileHead(SSdb *pSdb, TdFilePtr pFile) {
     return -1;
   }
 
+  if (taosWriteFile(pFile, &pSdb->curTerm, sizeof(int64_t)) != sizeof(int64_t)) {
+    terrno = TAOS_SYSTEM_ERROR(errno);
+    return -1;
+  }
+
   for (int32_t i = 0; i < SDB_TABLE_SIZE; ++i) {
     int64_t maxId = 0;
     if (i < SDB_MAX) {
@@ -182,6 +197,7 @@ int32_t sdbReadFile(SSdb *pSdb) {
   if (sdbReadFileHead(pSdb, pFile) != 0) {
     mError("failed to read file:%s head since %s", file, terrstr());
     pSdb->curVer = -1;
+    pSdb->curTerm = -1;
     taosMemoryFree(pRaw);
     taosCloseFile(&pFile);
     return -1;
@@ -256,8 +272,8 @@ static int32_t sdbWriteFileImp(SSdb *pSdb) {
   char curfile[PATH_MAX] = {0};
   snprintf(curfile, sizeof(curfile), "%s%ssdb.data", pSdb->currDir, TD_DIRSEP);
 
-  mDebug("start to write file:%s, current ver:%" PRId64 ", commit ver:%" PRId64, curfile, pSdb->curVer,
-         pSdb->lastCommitVer);
+  mDebug("start to write file:%s, current ver:%" PRId64 " term:%" PRId64 ", commit ver:%" PRId64, curfile, pSdb->curVer,
+         pSdb->curTerm, pSdb->lastCommitVer);
 
   TdFilePtr pFile = taosOpenFile(tmpfile, TD_FILE_CREATE | TD_FILE_WRITE | TD_FILE_TRUNC);
   if (pFile == NULL) {
@@ -350,7 +366,7 @@ static int32_t sdbWriteFileImp(SSdb *pSdb) {
     mError("failed to write file:%s since %s", curfile, tstrerror(code));
   } else {
     pSdb->lastCommitVer = pSdb->curVer;
-    mDebug("write file:%s successfully, ver:%" PRId64, curfile, pSdb->lastCommitVer);
+    mDebug("write file:%s successfully, ver:%" PRId64 " term:%" PRId64, curfile, pSdb->lastCommitVer, pSdb->curTerm);
   }
 
   terrno = code;

source/dnode/vnode/src/vnd/vnodeSync.c

@@ -56,7 +56,13 @@ void vnodeSyncStart(SVnode *pVnode) {
 
 void vnodeSyncClose(SVnode *pVnode) { syncStop(pVnode->sync); }
 
-int32_t vnodeSyncEqMsg(const SMsgCb *msgcb, SRpcMsg *pMsg) { return tmsgPutToQueue(msgcb, SYNC_QUEUE, pMsg); }
+int32_t vnodeSyncEqMsg(const SMsgCb *msgcb, SRpcMsg *pMsg) { 
+  int32_t code = tmsgPutToQueue(msgcb, SYNC_QUEUE, pMsg);
+  if (code != 0) {
+    rpcFreeCont(pMsg->pCont);
+  }
+  return code;
+}
 
 int32_t vnodeSyncSendMsg(const SEpSet *pEpSet, SRpcMsg *pMsg) { return tmsgSendReq(pEpSet, pMsg); }
 
@@ -141,5 +147,6 @@ SSyncFSM *vnodeSyncMakeFsm(SVnode *pVnode) {
   pFsm->FpPreCommitCb = vnodeSyncPreCommitMsg;
   pFsm->FpRollBackCb = vnodeSyncRollBackMsg;
   pFsm->FpGetSnapshot = vnodeSyncGetSnapshot;
+  pFsm->FpRestoreFinish = NULL;
   return pFsm;
 }
\ No newline at end of file

source/libs/sync/inc/syncIO.h

@@ -36,8 +36,8 @@ typedef struct SSyncIO {
   STaosQueue *pMsgQ;
   STaosQset * pQset;
   TdThread    consumerTid;
-  void       *serverRpc;
-  void       *clientRpc;
+  void *      serverRpc;
+  void *      clientRpc;
   SEpSet      myAddr;
   SMsgCb      msgcb;
 

source/libs/sync/inc/syncInt.h

@@ -147,6 +147,11 @@ typedef struct SSyncNode {
   // tools
   SSyncRespMgr* pSyncRespMgr;
 
+  // restore state
+  bool       restoreFinish;
+  //sem_t      restoreSem;
+  SSnapshot* pSnapshot;
+
 } SSyncNode;
 
 // open/close --------------

source/libs/sync/src/syncAppendEntries.c

@@ -324,7 +324,6 @@ int32_t syncNodeOnAppendEntriesCb(SSyncNode* ths, SyncAppendEntries* pMsg) {
               SRpcMsg rpcMsg;
               syncEntry2OriginalRpc(pEntry, &rpcMsg);
 
-              // if (ths->pFsm->FpCommitCb != NULL && pEntry->originalRpcType != TDMT_VND_SYNC_NOOP) {
               if (ths->pFsm->FpCommitCb != NULL && syncUtilUserCommit(pEntry->originalRpcType)) {
                 SFsmCbMeta cbMeta;
                 cbMeta.index = pEntry->index;
@@ -332,7 +331,18 @@ int32_t syncNodeOnAppendEntriesCb(SSyncNode* ths, SyncAppendEntries* pMsg) {
                 cbMeta.code = 0;
                 cbMeta.state = ths->state;
                 cbMeta.seqNum = pEntry->seqNum;
+                cbMeta.term = pEntry->term;
+                cbMeta.currentTerm = ths->pRaftStore->currentTerm;
                 ths->pFsm->FpCommitCb(ths->pFsm, &rpcMsg, cbMeta);
+
+                bool needExecute = true;
+                if (ths->pSnapshot != NULL && cbMeta.index <= ths->pSnapshot->lastApplyIndex) {
+                  needExecute = false;
+                }
+
+                if (needExecute) {
+                  ths->pFsm->FpCommitCb(ths->pFsm, &rpcMsg, cbMeta);
+                }
               }
 
               // config change
@@ -349,6 +359,22 @@ int32_t syncNodeOnAppendEntriesCb(SSyncNode* ths, SyncAppendEntries* pMsg) {
                 }
               }
 
+              // restore finish
+              if (pEntry->index == ths->pLogStore->getLastIndex(ths->pLogStore)) {
+                if (ths->restoreFinish == false) {
+                  if (ths->pFsm->FpRestoreFinish != NULL) {
+                    ths->pFsm->FpRestoreFinish(ths->pFsm);
+                  }
+                  ths->restoreFinish = true;
+                  sInfo("==syncNodeOnAppendEntriesCb== restoreFinish set true %p vgId:%d", ths, ths->vgId);
+
+                  /*
+                  tsem_post(&ths->restoreSem);
+                  sInfo("==syncNodeOnAppendEntriesCb== RestoreFinish tsem_post %p", ths);
+                  */
+                }
+              }
+
               rpcFreeCont(rpcMsg.pCont);
               syncEntryDestory(pEntry);
             }

source/libs/sync/src/syncCommit.c

@@ -102,7 +102,6 @@ void syncMaybeAdvanceCommitIndex(SSyncNode* pSyncNode) {
           SRpcMsg rpcMsg;
           syncEntry2OriginalRpc(pEntry, &rpcMsg);
 
-          // if (pSyncNode->pFsm->FpCommitCb != NULL && pEntry->originalRpcType != TDMT_VND_SYNC_NOOP) {
           if (pSyncNode->pFsm->FpCommitCb != NULL && syncUtilUserCommit(pEntry->originalRpcType)) {
             SFsmCbMeta cbMeta;
             cbMeta.index = pEntry->index;
@@ -110,7 +109,17 @@ void syncMaybeAdvanceCommitIndex(SSyncNode* pSyncNode) {
             cbMeta.code = 0;
             cbMeta.state = pSyncNode->state;
             cbMeta.seqNum = pEntry->seqNum;
-            pSyncNode->pFsm->FpCommitCb(pSyncNode->pFsm, &rpcMsg, cbMeta);
+            cbMeta.term = pEntry->term;
+            cbMeta.currentTerm = pSyncNode->pRaftStore->currentTerm;
+
+            bool needExecute = true;
+            if (pSyncNode->pSnapshot != NULL && cbMeta.index <= pSyncNode->pSnapshot->lastApplyIndex) {
+              needExecute = false;
+            }
+
+            if (needExecute) {
+              pSyncNode->pFsm->FpCommitCb(pSyncNode->pFsm, &rpcMsg, cbMeta);
+            }
           }
 
           // config change
@@ -127,6 +136,22 @@ void syncMaybeAdvanceCommitIndex(SSyncNode* pSyncNode) {
             }
           }
 
+          // restore finish
+          if (pEntry->index == pSyncNode->pLogStore->getLastIndex(pSyncNode->pLogStore)) {
+            if (pSyncNode->restoreFinish == false) {
+              if (pSyncNode->pFsm->FpRestoreFinish != NULL) {
+                pSyncNode->pFsm->FpRestoreFinish(pSyncNode->pFsm);
+              }
+              pSyncNode->restoreFinish = true;
+              sInfo("==syncMaybeAdvanceCommitIndex== restoreFinish set true %p vgId:%d", pSyncNode, pSyncNode->vgId);
+
+              /*
+              tsem_post(&pSyncNode->restoreSem);
+              sInfo("==syncMaybeAdvanceCommitIndex== RestoreFinish tsem_post %p", pSyncNode);
+              */
+            }
+          }
+
           rpcFreeCont(rpcMsg.pCont);
           syncEntryDestory(pEntry);
         }
@@ -162,4 +187,4 @@ bool syncAgree(SSyncNode* pSyncNode, SyncIndex index) {
     }
   }
   return false;
-}
+}
\ No newline at end of file

source/libs/sync/src/syncMain.c

@@ -13,7 +13,6 @@
  * along with this program. If not, see <http://www.gnu.org/licenses/>.
  */
 
-#include <stdint.h>
 #include "sync.h"
 #include "syncAppendEntries.h"
 #include "syncAppendEntriesReply.h"
@@ -55,14 +54,17 @@ static void syncFreeNode(void* param);
 // ---------------------------------
 
 int32_t syncInit() {
-  int32_t ret;
-  tsNodeRefId = taosOpenRef(200, syncFreeNode);
-  if (tsNodeRefId < 0) {
-    sError("failed to init node ref");
-    syncCleanUp();
-    ret = -1;
-  } else {
-    ret = syncEnvStart();
+  int32_t ret = 0;
+
+  if (!syncEnvIsStart()) {
+    tsNodeRefId = taosOpenRef(200, syncFreeNode);
+    if (tsNodeRefId < 0) {
+      sError("failed to init node ref");
+      syncCleanUp();
+      ret = -1;
+    } else {
+      ret = syncEnvStart();
+    }
   }
 
   return ret;
@@ -155,6 +157,18 @@ ESyncState syncGetMyRole(int64_t rid) {
   return state;
 }
 
+bool syncIsRestoreFinish(int64_t rid) {
+  SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid);
+  if (pSyncNode == NULL) {
+    return false;
+  }
+  assert(rid == pSyncNode->rid);
+  bool b = pSyncNode->restoreFinish;
+
+  taosReleaseRef(tsNodeRefId, pSyncNode->rid);
+  return b;
+}
+
 const char* syncGetMyRoleStr(int64_t rid) {
   const char* s = syncUtilState2String(syncGetMyRole(rid));
   return s;
@@ -240,7 +254,7 @@ int32_t syncGetAndDelRespRpc(int64_t rid, uint64_t index, SRpcMsg* msg) {
   return ret;
 }
 
-void syncSetMsgCb(int64_t rid, const SMsgCb *msgcb) {
+void syncSetMsgCb(int64_t rid, const SMsgCb* msgcb) {
   SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid);
   if (pSyncNode == NULL) {
     sTrace("syncSetQ get pSyncNode is NULL, rid:%ld", rid);
@@ -306,8 +320,10 @@ int32_t syncPropose(int64_t rid, const SRpcMsg* pMsg, bool isWeak) {
   int32_t    ret = TAOS_SYNC_PROPOSE_SUCCESS;
   SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid);
   if (pSyncNode == NULL) {
+    rpcFreeCont(pMsg->pCont);
     return TAOS_SYNC_PROPOSE_OTHER_ERROR;
   }
+
   assert(rid == pSyncNode->rid);
 
   if (pSyncNode->state == TAOS_SYNC_STATE_LEADER) {
@@ -490,6 +506,15 @@ SSyncNode* syncNodeOpen(const SSyncInfo* pSyncInfo) {
   pSyncNode->pSyncRespMgr = syncRespMgrCreate(NULL, 0);
   assert(pSyncNode->pSyncRespMgr != NULL);
 
+  // restore state
+  pSyncNode->restoreFinish = false;
+  pSyncNode->pSnapshot = NULL;
+  if (pSyncNode->pFsm->FpGetSnapshot != NULL) {
+    pSyncNode->pSnapshot = taosMemoryMalloc(sizeof(SSnapshot));
+    pSyncNode->pFsm->FpGetSnapshot(pSyncNode->pFsm, pSyncNode->pSnapshot);
+  }
+  //tsem_init(&(pSyncNode->restoreSem), 0, 0);
+
   // start in syncNodeStart
   // start raft
   // syncNodeBecomeFollower(pSyncNode);
@@ -509,6 +534,20 @@ void syncNodeStart(SSyncNode* pSyncNode) {
     // use this now
     syncNodeAppendNoop(pSyncNode);
     syncMaybeAdvanceCommitIndex(pSyncNode);  // maybe only one replica
+
+    /*
+    sInfo("==syncNodeStart== RestoreFinish begin 1 replica tsem_wait %p", pSyncNode);
+    tsem_wait(&pSyncNode->restoreSem);
+    sInfo("==syncNodeStart== RestoreFinish end 1 replica tsem_wait %p", pSyncNode);
+    */
+
+    /*
+    while (pSyncNode->restoreFinish != true) {
+      taosMsleep(10);
+    }
+    */
+
+    sInfo("==syncNodeStart== restoreFinish ok 1 replica %p vgId:%d", pSyncNode, pSyncNode->vgId);
     return;
   }
 
@@ -518,6 +557,19 @@ void syncNodeStart(SSyncNode* pSyncNode) {
   int32_t ret = 0;
   // ret = syncNodeStartPingTimer(pSyncNode);
   assert(ret == 0);
+
+  /*
+  sInfo("==syncNodeStart== RestoreFinish begin multi replica tsem_wait %p", pSyncNode);
+  tsem_wait(&pSyncNode->restoreSem);
+  sInfo("==syncNodeStart== RestoreFinish end multi replica tsem_wait %p", pSyncNode);
+  */
+
+  /*
+  while (pSyncNode->restoreFinish != true) {
+    taosMsleep(10);
+  }
+  */
+  sInfo("==syncNodeStart== restoreFinish ok multi replica %p vgId:%d", pSyncNode, pSyncNode->vgId);
 }
 
 void syncNodeStartStandBy(SSyncNode* pSyncNode) {
@@ -554,6 +606,12 @@ void syncNodeClose(SSyncNode* pSyncNode) {
     taosMemoryFree(pSyncNode->pFsm);
   }
 
+  if (pSyncNode->pSnapshot != NULL) {
+    taosMemoryFree(pSyncNode->pSnapshot);
+  }
+
+  //tsem_destroy(&pSyncNode->restoreSem);
+
   // free memory in syncFreeNode
   // taosMemoryFree(pSyncNode);
 }

source/libs/sync/test/syncConfigChangeTest.cpp

@@ -73,12 +73,17 @@ int32_t GetSnapshotCb(struct SSyncFSM* pFsm, SSnapshot* pSnapshot) {
   return 0;
 }
 
+void FpRestoreFinishCb(struct SSyncFSM* pFsm) {
+  sTrace("==callback== ==FpRestoreFinishCb==");
+}
+
 SSyncFSM* createFsm() {
   SSyncFSM* pFsm = (SSyncFSM*)taosMemoryMalloc(sizeof(SSyncFSM));
   pFsm->FpCommitCb = CommitCb;
   pFsm->FpPreCommitCb = PreCommitCb;
   pFsm->FpRollBackCb = RollBackCb;
   pFsm->FpGetSnapshot = GetSnapshotCb;
+  pFsm->FpRestoreFinish = FpRestoreFinishCb;
   return pFsm;
 }
 

source/libs/sync/test/syncSnapshotTest.cpp

@@ -160,6 +160,8 @@ SyncClientRequest *step1(const SRpcMsg *pMsg) {
 }
 
 int main(int argc, char **argv) {
+  sprintf(tsTempDir, "%s", ".");
+
   // taosInitLog((char *)"syncTest.log", 100000, 10);
   tsAsyncLog = 0;
   sDebugFlag = 143 + 64;

source/libs/transport/inc/transComm.h

@@ -94,7 +94,7 @@ typedef void* queue[2];
 /* Return the structure holding the given element. */
 #define QUEUE_DATA(e, type, field) ((type*)((void*)((char*)(e)-offsetof(type, field))))
 
-#define TRANS_RETRY_COUNT_LIMIT 20  // retry count limit
+#define TRANS_RETRY_COUNT_LIMIT 100  // retry count limit
 #define TRANS_RETRY_INTERVAL    15  // ms retry interval
 #define TRANS_CONN_TIMEOUT      3   // connect timeout
 

source/libs/transport/src/trans.c

@@ -17,7 +17,7 @@
 
 #include "transComm.h"
 
-void* (*taosInitHandle[])(uint32_t ip, uint32_t port, char* label, int numOfThreads, void* fp, void* shandle) = {
+void* (*taosInitHandle[])(uint32_t ip, uint32_t port, char* label, int32_t numOfThreads, void* fp, void* shandle) = {
     transInitServer, transInitClient};
 
 void (*taosCloseHandle[])(void* arg) = {transCloseServer, transCloseClient};
@@ -77,37 +77,38 @@ void rpcClose(void* arg) {
   taosMemoryFree(pRpc);
   return;
 }
-void* rpcMallocCont(int contLen) {
-  int size = contLen + TRANS_MSG_OVERHEAD;
 
-  char* start = (char*)taosMemoryCalloc(1, (size_t)size);
+void* rpcMallocCont(int32_t contLen) {
+  int32_t size = contLen + TRANS_MSG_OVERHEAD;
+  char*   start = taosMemoryCalloc(1, size);
   if (start == NULL) {
     tError("failed to malloc msg, size:%d", size);
+    terrno = TSDB_CODE_OUT_OF_MEMORY;
     return NULL;
   } else {
     tTrace("malloc mem:%p size:%d", start, size);
   }
+
   return start + sizeof(STransMsgHead);
 }
-void rpcFreeCont(void* cont) {
-  // impl
-  if (cont == NULL) {
-    return;
-  }
 
+void rpcFreeCont(void* cont) {
+  if (cont == NULL) return;
   taosMemoryFree((char*)cont - TRANS_MSG_OVERHEAD);
   tTrace("free mem: %p", (char*)cont - TRANS_MSG_OVERHEAD);
 }
-void* rpcReallocCont(void* ptr, int contLen) {
-  if (ptr == NULL) {
-    return rpcMallocCont(contLen);
-  }
-  char* st = (char*)ptr - TRANS_MSG_OVERHEAD;
-  int   sz = contLen + TRANS_MSG_OVERHEAD;
+
+void* rpcReallocCont(void* ptr, int32_t contLen) {
+  if (ptr == NULL) return rpcMallocCont(contLen);
+
+  char*   st = (char*)ptr - TRANS_MSG_OVERHEAD;
+  int32_t sz = contLen + TRANS_MSG_OVERHEAD;
   st = taosMemoryRealloc(st, sz);
   if (st == NULL) {
+    terrno = TSDB_CODE_OUT_OF_MEMORY;
     return NULL;
   }
+
   return st + TRANS_MSG_OVERHEAD;
 }
 
@@ -116,8 +117,8 @@ void rpcSendRedirectRsp(void* thandle, const SEpSet* pEpSet) {
   assert(0);
 }
 
-int  rpcReportProgress(void* pConn, char* pCont, int contLen) { return -1; }
-void rpcCancelRequest(int64_t rid) { return; }
+int32_t rpcReportProgress(void* pConn, char* pCont, int32_t contLen) { return -1; }
+void    rpcCancelRequest(int64_t rid) { return; }
 
 void rpcSendRequest(void* shandle, const SEpSet* pEpSet, SRpcMsg* pMsg, int64_t* pRid) {
   transSendRequest(shandle, pEpSet, pMsg, NULL);
@@ -129,8 +130,8 @@ void rpcSendRecv(void* shandle, SEpSet* pEpSet, SRpcMsg* pMsg, SRpcMsg* pRsp) {
   transSendRecv(shandle, pEpSet, pMsg, pRsp);
 }
 
-void rpcSendResponse(const SRpcMsg* pMsg) { transSendResponse(pMsg); }
-int  rpcGetConnInfo(void* thandle, SRpcConnInfo* pInfo) { return transGetConnInfo((void*)thandle, pInfo); }
+void    rpcSendResponse(const SRpcMsg* pMsg) { transSendResponse(pMsg); }
+int32_t rpcGetConnInfo(void* thandle, SRpcConnInfo* pInfo) { return transGetConnInfo((void*)thandle, pInfo); }
 
 void rpcRefHandle(void* handle, int8_t type) {
   assert(type == TAOS_CONN_SERVER || type == TAOS_CONN_CLIENT);

tests/test/c/sdbDump.c

@@ -262,7 +262,7 @@ void dumpCluster(SSdb *pSdb, SJson *json) {
 }
 
 void dumpTrans(SSdb *pSdb, SJson *json) {
-    void  *pIter = NULL;
+  void  *pIter = NULL;
   SJson *items = tjsonCreateObject();
   tjsonAddItemToObject(json, "transactions", items);
 
@@ -294,6 +294,7 @@ void dumpTrans(SSdb *pSdb, SJson *json) {
 void dumpHeader(SSdb *pSdb, SJson *json) {
   tjsonAddIntegerToObject(json, "sver", 1);
   tjsonAddStringToObject(json, "curVer", i642str(pSdb->curVer));
+  tjsonAddStringToObject(json, "curTerm", i642str(pSdb->curTerm));
 
   SJson *maxIdsJson = tjsonCreateObject();
   tjsonAddItemToObject(json, "maxIds", maxIdsJson);