/* * Copyright (c) 2019 TAOS Data, Inc. * * This program is free software: you can use, redistribute, and/or modify * it under the terms of the GNU Affero General Public License, version 3 * or later ("AGPL"), as published by the Free Software Foundation. * * This program is distributed in the hope that it will be useful, but WITHOUT * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or * FITNESS FOR A PARTICULAR PURPOSE. * * You should have received a copy of the GNU Affero General Public License * along with this program. If not, see . */ #include "sync.h" #include "syncAppendEntries.h" #include "syncAppendEntriesReply.h" #include "syncCommit.h" #include "syncElection.h" #include "syncEnv.h" #include "syncIndexMgr.h" #include "syncInt.h" #include "syncMessage.h" #include "syncRaftCfg.h" #include "syncRaftLog.h" #include "syncRaftStore.h" #include "syncReplication.h" #include "syncRequestVote.h" #include "syncRequestVoteReply.h" #include "syncRespMgr.h" #include "syncSnapshot.h" #include "syncTimeout.h" #include "syncUtil.h" #include "syncVoteMgr.h" #include "tref.h" bool gRaftDetailLog = false; static int32_t tsNodeRefId = -1; // ------ local funciton --------- // enqueue message ---- static void syncNodeEqPingTimer(void* param, void* tmrId); static void syncNodeEqElectTimer(void* param, void* tmrId); static void syncNodeEqHeartbeatTimer(void* param, void* tmrId); static int32_t syncNodeEqNoop(SSyncNode* ths); static int32_t syncNodeAppendNoop(SSyncNode* ths); // process message ---- int32_t syncNodeOnPingCb(SSyncNode* ths, SyncPing* pMsg); int32_t syncNodeOnPingReplyCb(SSyncNode* ths, SyncPingReply* pMsg); // life cycle static void syncFreeNode(void* param); // --------------------------------- int32_t syncInit() { int32_t ret = 0; if (!syncEnvIsStart()) { tsNodeRefId = taosOpenRef(200, syncFreeNode); if (tsNodeRefId < 0) { sError("failed to init node ref"); syncCleanUp(); ret = -1; } else { sDebug("sync rsetId:%" PRId64 " is open", tsNodeRefId); ret = syncEnvStart(); } } return ret; } void syncCleanUp() { int32_t ret = syncEnvStop(); ASSERT(ret == 0); if (tsNodeRefId != -1) { sDebug("sync rsetId:%" PRId64 " is closed", tsNodeRefId); taosCloseRef(tsNodeRefId); tsNodeRefId = -1; } } int64_t syncOpen(const SSyncInfo* pSyncInfo) { SSyncNode* pSyncNode = syncNodeOpen(pSyncInfo); ASSERT(pSyncNode != NULL); if (gRaftDetailLog) { syncNodeLog2("syncNodeOpen open success", pSyncNode); } pSyncNode->rid = taosAddRef(tsNodeRefId, pSyncNode); if (pSyncNode->rid < 0) { syncFreeNode(pSyncNode); return -1; } sDebug("vgId:%d, rid:%" PRId64 " is added to rsetId:%" PRId64, pSyncInfo->vgId, pSyncNode->rid, tsNodeRefId); return pSyncNode->rid; } void syncStart(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return; } if (pSyncNode->pRaftCfg->isStandBy) { syncNodeStartStandBy(pSyncNode); } else { syncNodeStart(pSyncNode); } taosReleaseRef(tsNodeRefId, pSyncNode->rid); } void syncStartNormal(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return; } syncNodeStart(pSyncNode); taosReleaseRef(tsNodeRefId, pSyncNode->rid); } void syncStartStandBy(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return; } syncNodeStartStandBy(pSyncNode); taosReleaseRef(tsNodeRefId, pSyncNode->rid); } void syncStop(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) return; int32_t vgId = pSyncNode->vgId; syncNodeClose(pSyncNode); taosReleaseRef(tsNodeRefId, pSyncNode->rid); taosRemoveRef(tsNodeRefId, rid); sDebug("vgId:%d, rid:%" PRId64 " is removed from rsetId:%" PRId64, vgId, rid, tsNodeRefId); } int32_t syncSetStandby(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { terrno = TSDB_CODE_SYN_INTERNAL_ERROR; sError("failed to set standby since accquire ref error, rid:%" PRId64, rid); return -1; } if (pSyncNode->state != TAOS_SYNC_STATE_FOLLOWER) { if (pSyncNode->state == TAOS_SYNC_STATE_LEADER) { terrno = TSDB_CODE_SYN_IS_LEADER; } else { terrno = TSDB_CODE_SYN_STANDBY_NOT_READY; } sError("failed to set standby since it is not follower, state:%s rid:%" PRId64, syncStr(pSyncNode->state), rid); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return -1; } // state change pSyncNode->state = TAOS_SYNC_STATE_FOLLOWER; syncNodeStopHeartbeatTimer(pSyncNode); // reset elect timer, long enough int32_t electMS = TIMER_MAX_MS; int32_t ret = syncNodeRestartElectTimer(pSyncNode, electMS); ASSERT(ret == 0); pSyncNode->pRaftCfg->isStandBy = 1; raftCfgPersist(pSyncNode->pRaftCfg); taosReleaseRef(tsNodeRefId, pSyncNode->rid); sInfo("vgId:%d, set to standby", pSyncNode->vgId); return 0; } bool syncNodeCheckNewConfig(SSyncNode* pSyncNode, const SSyncCfg* pNewCfg) { bool IamInNew = syncNodeInConfig(pSyncNode, pNewCfg); if (!IamInNew) { return false; } if (pNewCfg->replicaNum > pSyncNode->replicaNum + 1) { return false; } if (pNewCfg->replicaNum < pSyncNode->replicaNum - 1) { return false; } return true; } int32_t syncReconfigBuild(int64_t rid, const SSyncCfg* pNewCfg, SRpcMsg* pRpcMsg) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { terrno = TSDB_CODE_SYN_INTERNAL_ERROR; return -1; } ASSERT(rid == pSyncNode->rid); int32_t ret = 0; if (!syncNodeCheckNewConfig(pSyncNode, pNewCfg)) { taosReleaseRef(tsNodeRefId, pSyncNode->rid); terrno = TSDB_CODE_SYN_NEW_CONFIG_ERROR; sError("syncNodeCheckNewConfig error"); return -1; } char* newconfig = syncCfg2Str((SSyncCfg*)pNewCfg); pRpcMsg->msgType = TDMT_SYNC_CONFIG_CHANGE; pRpcMsg->info.noResp = 1; pRpcMsg->contLen = strlen(newconfig) + 1; pRpcMsg->pCont = rpcMallocCont(pRpcMsg->contLen); snprintf(pRpcMsg->pCont, pRpcMsg->contLen, "%s", newconfig); taosMemoryFree(newconfig); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return ret; } int32_t syncReconfig(int64_t rid, const SSyncCfg* pNewCfg) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { terrno = TSDB_CODE_SYN_INTERNAL_ERROR; return -1; } ASSERT(rid == pSyncNode->rid); if (!syncNodeCheckNewConfig(pSyncNode, pNewCfg)) { taosReleaseRef(tsNodeRefId, pSyncNode->rid); terrno = TSDB_CODE_SYN_NEW_CONFIG_ERROR; sError("syncNodeCheckNewConfig error"); return -1; } char* newconfig = syncCfg2Str((SSyncCfg*)pNewCfg); if (gRaftDetailLog) { sInfo("==syncReconfig== newconfig:%s", newconfig); } int32_t ret = 0; SRpcMsg rpcMsg = {0}; rpcMsg.msgType = TDMT_SYNC_CONFIG_CHANGE; rpcMsg.info.noResp = 1; rpcMsg.contLen = strlen(newconfig) + 1; rpcMsg.pCont = rpcMallocCont(rpcMsg.contLen); snprintf(rpcMsg.pCont, rpcMsg.contLen, "%s", newconfig); taosMemoryFree(newconfig); ret = syncNodePropose(pSyncNode, &rpcMsg, false); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return ret; } int32_t syncLeaderTransfer(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { terrno = TSDB_CODE_SYN_INTERNAL_ERROR; return -1; } ASSERT(rid == pSyncNode->rid); int32_t ret = syncNodeLeaderTransfer(pSyncNode); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return ret; } int32_t syncLeaderTransferTo(int64_t rid, SNodeInfo newLeader) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { terrno = TSDB_CODE_SYN_INTERNAL_ERROR; return -1; } ASSERT(rid == pSyncNode->rid); int32_t ret = syncNodeLeaderTransferTo(pSyncNode, newLeader); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return ret; } int32_t syncNodeLeaderTransfer(SSyncNode* pSyncNode) { if (pSyncNode->peersNum == 0) { sDebug("only one replica, cannot leader transfer"); terrno = TSDB_CODE_SYN_ONE_REPLICA; return -1; } SNodeInfo newLeader = (pSyncNode->peersNodeInfo)[0]; int32_t ret = syncNodeLeaderTransferTo(pSyncNode, newLeader); return ret; } int32_t syncNodeLeaderTransferTo(SSyncNode* pSyncNode, SNodeInfo newLeader) { int32_t ret = 0; if (pSyncNode->replicaNum == 1) { sDebug("only one replica, cannot leader transfer"); terrno = TSDB_CODE_SYN_ONE_REPLICA; return -1; } do { char logBuf[128]; snprintf(logBuf, sizeof(logBuf), "begin leader transfer to %s:%u", newLeader.nodeFqdn, newLeader.nodePort); syncNodeEventLog(pSyncNode, logBuf); } while (0); SyncLeaderTransfer* pMsg = syncLeaderTransferBuild(pSyncNode->vgId); pMsg->newLeaderId.addr = syncUtilAddr2U64(newLeader.nodeFqdn, newLeader.nodePort); pMsg->newLeaderId.vgId = pSyncNode->vgId; pMsg->newNodeInfo = newLeader; ASSERT(pMsg != NULL); SRpcMsg rpcMsg = {0}; syncLeaderTransfer2RpcMsg(pMsg, &rpcMsg); syncLeaderTransferDestroy(pMsg); ret = syncNodePropose(pSyncNode, &rpcMsg, false); return ret; } bool syncCanLeaderTransfer(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return false; } ASSERT(rid == pSyncNode->rid); if (pSyncNode->replicaNum == 1) { taosReleaseRef(tsNodeRefId, pSyncNode->rid); return false; } if (pSyncNode->state == TAOS_SYNC_STATE_FOLLOWER) { taosReleaseRef(tsNodeRefId, pSyncNode->rid); return true; } bool matchOK = true; if (pSyncNode->state == TAOS_SYNC_STATE_CANDIDATE || pSyncNode->state == TAOS_SYNC_STATE_LEADER) { SyncIndex myCommitIndex = pSyncNode->commitIndex; for (int i = 0; i < pSyncNode->peersNum; ++i) { SyncIndex peerMatchIndex = syncIndexMgrGetIndex(pSyncNode->pMatchIndex, &(pSyncNode->peersId)[i]); if (peerMatchIndex < myCommitIndex) { matchOK = false; } } } taosReleaseRef(tsNodeRefId, pSyncNode->rid); return matchOK; } int32_t syncForwardToPeer(int64_t rid, SRpcMsg* pMsg, bool isWeak) { int32_t ret = syncPropose(rid, pMsg, isWeak); return ret; } ESyncState syncGetMyRole(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return TAOS_SYNC_STATE_ERROR; } ASSERT(rid == pSyncNode->rid); ESyncState state = pSyncNode->state; taosReleaseRef(tsNodeRefId, pSyncNode->rid); return state; } bool syncIsReady(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return false; } ASSERT(rid == pSyncNode->rid); bool b = (pSyncNode->state == TAOS_SYNC_STATE_LEADER) && pSyncNode->restoreFinish; taosReleaseRef(tsNodeRefId, pSyncNode->rid); // if false, set error code if (false == b) { if (pSyncNode->state != TAOS_SYNC_STATE_LEADER) { terrno = TSDB_CODE_SYN_NOT_LEADER; } else { terrno = TSDB_CODE_APP_NOT_READY; } } return b; } 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; } int32_t syncGetSnapshotByIndex(int64_t rid, SyncIndex index, SSnapshot* pSnapshot) { if (index < SYNC_INDEX_BEGIN) { return -1; } SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return -1; } ASSERT(rid == pSyncNode->rid); SSyncRaftEntry* pEntry = NULL; int32_t code = pSyncNode->pLogStore->syncLogGetEntry(pSyncNode->pLogStore, index, &pEntry); if (code != 0) { if (pEntry != NULL) { syncEntryDestory(pEntry); } taosReleaseRef(tsNodeRefId, pSyncNode->rid); return -1; } ASSERT(pEntry != NULL); pSnapshot->data = NULL; pSnapshot->lastApplyIndex = index; pSnapshot->lastApplyTerm = pEntry->term; pSnapshot->lastConfigIndex = syncNodeGetSnapshotConfigIndex(pSyncNode, index); syncEntryDestory(pEntry); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return 0; } int32_t syncGetSnapshotMeta(int64_t rid, struct SSnapshotMeta* sMeta) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return -1; } ASSERT(rid == pSyncNode->rid); sMeta->lastConfigIndex = pSyncNode->pRaftCfg->lastConfigIndex; sTrace("vgId:%d, get snapshot meta, lastConfigIndex:%" PRId64, pSyncNode->vgId, pSyncNode->pRaftCfg->lastConfigIndex); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return 0; } int32_t syncGetSnapshotMetaByIndex(int64_t rid, SyncIndex snapshotIndex, struct SSnapshotMeta* sMeta) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return -1; } ASSERT(rid == pSyncNode->rid); ASSERT(pSyncNode->pRaftCfg->configIndexCount >= 1); SyncIndex lastIndex = (pSyncNode->pRaftCfg->configIndexArr)[0]; for (int i = 0; i < pSyncNode->pRaftCfg->configIndexCount; ++i) { if ((pSyncNode->pRaftCfg->configIndexArr)[i] > lastIndex && (pSyncNode->pRaftCfg->configIndexArr)[i] <= snapshotIndex) { lastIndex = (pSyncNode->pRaftCfg->configIndexArr)[i]; } } sMeta->lastConfigIndex = lastIndex; sTrace("vgId:%d, get snapshot meta by index:%" PRId64 " lcindex:%" PRId64, pSyncNode->vgId, snapshotIndex, sMeta->lastConfigIndex); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return 0; } SyncIndex syncNodeGetSnapshotConfigIndex(SSyncNode* pSyncNode, SyncIndex snapshotLastApplyIndex) { ASSERT(pSyncNode->pRaftCfg->configIndexCount >= 1); SyncIndex lastIndex = (pSyncNode->pRaftCfg->configIndexArr)[0]; for (int i = 0; i < pSyncNode->pRaftCfg->configIndexCount; ++i) { if ((pSyncNode->pRaftCfg->configIndexArr)[i] > lastIndex && (pSyncNode->pRaftCfg->configIndexArr)[i] <= snapshotLastApplyIndex) { lastIndex = (pSyncNode->pRaftCfg->configIndexArr)[i]; } } sTrace("vgId:%d, sync get snapshot last config index, index:%" PRId64 " lcindex:%" PRId64, pSyncNode->vgId, snapshotLastApplyIndex, lastIndex); return lastIndex; } const char* syncGetMyRoleStr(int64_t rid) { const char* s = syncUtilState2String(syncGetMyRole(rid)); return s; } SyncTerm syncGetMyTerm(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return TAOS_SYNC_STATE_ERROR; } ASSERT(rid == pSyncNode->rid); SyncTerm term = pSyncNode->pRaftStore->currentTerm; taosReleaseRef(tsNodeRefId, pSyncNode->rid); return term; } SyncGroupId syncGetVgId(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return TAOS_SYNC_STATE_ERROR; } ASSERT(rid == pSyncNode->rid); SyncGroupId vgId = pSyncNode->vgId; taosReleaseRef(tsNodeRefId, pSyncNode->rid); return vgId; } void syncGetEpSet(int64_t rid, SEpSet* pEpSet) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { memset(pEpSet, 0, sizeof(*pEpSet)); return; } ASSERT(rid == pSyncNode->rid); pEpSet->numOfEps = 0; for (int i = 0; i < pSyncNode->pRaftCfg->cfg.replicaNum; ++i) { snprintf(pEpSet->eps[i].fqdn, sizeof(pEpSet->eps[i].fqdn), "%s", (pSyncNode->pRaftCfg->cfg.nodeInfo)[i].nodeFqdn); pEpSet->eps[i].port = (pSyncNode->pRaftCfg->cfg.nodeInfo)[i].nodePort; (pEpSet->numOfEps)++; sInfo("vgId:%d sync get epset: index:%d %s:%d", pSyncNode->vgId, i, pEpSet->eps[i].fqdn, pEpSet->eps[i].port); } pEpSet->inUse = pSyncNode->pRaftCfg->cfg.myIndex; sInfo("vgId:%d sync get epset in-use:%d", pSyncNode->vgId, pEpSet->inUse); taosReleaseRef(tsNodeRefId, pSyncNode->rid); } void syncGetRetryEpSet(int64_t rid, SEpSet* pEpSet) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { memset(pEpSet, 0, sizeof(*pEpSet)); return; } ASSERT(rid == pSyncNode->rid); pEpSet->numOfEps = 0; for (int i = 0; i < pSyncNode->pRaftCfg->cfg.replicaNum; ++i) { snprintf(pEpSet->eps[i].fqdn, sizeof(pEpSet->eps[i].fqdn), "%s", (pSyncNode->pRaftCfg->cfg.nodeInfo)[i].nodeFqdn); pEpSet->eps[i].port = (pSyncNode->pRaftCfg->cfg.nodeInfo)[i].nodePort; (pEpSet->numOfEps)++; sInfo("vgId:%d sync get retry epset: index:%d %s:%d", pSyncNode->vgId, i, pEpSet->eps[i].fqdn, pEpSet->eps[i].port); } pEpSet->inUse = (pSyncNode->pRaftCfg->cfg.myIndex + 1) % pEpSet->numOfEps; sInfo("vgId:%d sync get retry epset in-use:%d", pSyncNode->vgId, pEpSet->inUse); taosReleaseRef(tsNodeRefId, pSyncNode->rid); } int32_t syncGetRespRpc(int64_t rid, uint64_t index, SRpcMsg* msg) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return TAOS_SYNC_STATE_ERROR; } ASSERT(rid == pSyncNode->rid); SRespStub stub; int32_t ret = syncRespMgrGet(pSyncNode->pSyncRespMgr, index, &stub); if (ret == 1) { memcpy(msg, &(stub.rpcMsg), sizeof(SRpcMsg)); } taosReleaseRef(tsNodeRefId, pSyncNode->rid); return ret; } int32_t syncGetAndDelRespRpc(int64_t rid, uint64_t index, SRpcHandleInfo* pInfo) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return TAOS_SYNC_STATE_ERROR; } ASSERT(rid == pSyncNode->rid); SRespStub stub; int32_t ret = syncRespMgrGetAndDel(pSyncNode->pSyncRespMgr, index, &stub); if (ret == 1) { *pInfo = stub.rpcMsg.info; } sTrace("vgId:%d, get seq:%" PRIu64 " rpc handle:%p", pSyncNode->vgId, index, pInfo->handle); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return ret; } void syncSetMsgCb(int64_t rid, const SMsgCb* msgcb) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { sTrace("syncSetQ get pSyncNode is NULL, rid:%" PRId64, rid); return; } ASSERT(rid == pSyncNode->rid); pSyncNode->msgcb = msgcb; taosReleaseRef(tsNodeRefId, pSyncNode->rid); } char* sync2SimpleStr(int64_t rid) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { sTrace("syncSetRpc get pSyncNode is NULL, rid:%" PRId64, rid); return NULL; } ASSERT(rid == pSyncNode->rid); char* s = syncNode2SimpleStr(pSyncNode); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return s; } void setPingTimerMS(int64_t rid, int32_t pingTimerMS) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return; } ASSERT(rid == pSyncNode->rid); pSyncNode->pingBaseLine = pingTimerMS; pSyncNode->pingTimerMS = pingTimerMS; taosReleaseRef(tsNodeRefId, pSyncNode->rid); } void setElectTimerMS(int64_t rid, int32_t electTimerMS) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return; } ASSERT(rid == pSyncNode->rid); pSyncNode->electBaseLine = electTimerMS; taosReleaseRef(tsNodeRefId, pSyncNode->rid); } void setHeartbeatTimerMS(int64_t rid, int32_t hbTimerMS) { SSyncNode* pSyncNode = (SSyncNode*)taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { return; } ASSERT(rid == pSyncNode->rid); pSyncNode->hbBaseLine = hbTimerMS; pSyncNode->heartbeatTimerMS = hbTimerMS; taosReleaseRef(tsNodeRefId, pSyncNode->rid); } int32_t syncPropose(int64_t rid, SRpcMsg* pMsg, bool isWeak) { SSyncNode* pSyncNode = taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { taosReleaseRef(tsNodeRefId, rid); terrno = TSDB_CODE_SYN_INTERNAL_ERROR; return -1; } ASSERT(rid == pSyncNode->rid); int32_t ret = syncNodePropose(pSyncNode, pMsg, isWeak); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return ret; } int32_t syncProposeBatch(int64_t rid, SRpcMsg* pMsgArr, bool* pIsWeakArr, int32_t arrSize) { if (arrSize < 0) { terrno = TSDB_CODE_SYN_INTERNAL_ERROR; return -1; } SSyncNode* pSyncNode = taosAcquireRef(tsNodeRefId, rid); if (pSyncNode == NULL) { terrno = TSDB_CODE_SYN_INTERNAL_ERROR; return -1; } ASSERT(rid == pSyncNode->rid); int32_t ret = syncNodeProposeBatch(pSyncNode, pMsgArr, pIsWeakArr, arrSize); taosReleaseRef(tsNodeRefId, pSyncNode->rid); return ret; } static bool syncNodeBatchOK(SRpcMsg* pMsgArr, int32_t arrSize) { for (int32_t i = 0; i < arrSize; ++i) { if (pMsgArr[i].msgType == TDMT_SYNC_CONFIG_CHANGE) { return false; } if (pMsgArr[i].msgType == TDMT_SYNC_CONFIG_CHANGE_FINISH) { return false; } } return true; } int32_t syncNodeProposeBatch(SSyncNode* pSyncNode, SRpcMsg* pMsgArr, bool* pIsWeakArr, int32_t arrSize) { if (!syncNodeBatchOK(pMsgArr, arrSize)) { syncNodeErrorLog(pSyncNode, "sync propose batch error"); terrno = TSDB_CODE_SYN_BATCH_ERROR; return -1; } if (arrSize > SYNC_MAX_BATCH_SIZE) { syncNodeErrorLog(pSyncNode, "sync propose batch error"); terrno = TSDB_CODE_SYN_BATCH_ERROR; return -1; } if (pSyncNode->state != TAOS_SYNC_STATE_LEADER) { syncNodeErrorLog(pSyncNode, "sync propose not leader"); terrno = TSDB_CODE_SYN_NOT_LEADER; return -1; } if (pSyncNode->changing) { syncNodeErrorLog(pSyncNode, "sync propose not ready"); terrno = TSDB_CODE_SYN_PROPOSE_NOT_READY; return -1; } SRaftMeta raftArr[SYNC_MAX_BATCH_SIZE]; for (int i = 0; i < arrSize; ++i) { SRespStub stub; stub.createTime = taosGetTimestampMs(); stub.rpcMsg = pMsgArr[i]; uint64_t seqNum = syncRespMgrAdd(pSyncNode->pSyncRespMgr, &stub); raftArr[i].isWeak = pIsWeakArr[i]; raftArr[i].seqNum = seqNum; } SyncClientRequestBatch* pSyncMsg = syncClientRequestBatchBuild(pMsgArr, raftArr, arrSize, pSyncNode->vgId); ASSERT(pSyncMsg != NULL); SRpcMsg rpcMsg; syncClientRequestBatch2RpcMsg(pSyncMsg, &rpcMsg); taosMemoryFree(pSyncMsg); // only free msg body, do not free rpc msg content if (pSyncNode->replicaNum == 1 && pSyncNode->vgId != 1) { int32_t code = syncNodeOnClientRequestBatchCb(pSyncNode, pSyncMsg); if (code == 0) { // update rpc msg applyIndex SRpcMsg* msgArr = syncClientRequestBatchRpcMsgArr(pSyncMsg); ASSERT(arrSize == pSyncMsg->dataCount); for (int i = 0; i < arrSize; ++i) { pMsgArr[i].info.conn.applyIndex = msgArr[i].info.conn.applyIndex; syncRespMgrDel(pSyncNode->pSyncRespMgr, raftArr[i].seqNum); } rpcFreeCont(rpcMsg.pCont); terrno = 0; return 1; } else { terrno = TSDB_CODE_SYN_INTERNAL_ERROR; return -1; } } else { if (pSyncNode->FpEqMsg != NULL && (*pSyncNode->FpEqMsg)(pSyncNode->msgcb, &rpcMsg) == 0) { // enqueue msg ok return 0; } else { sError("vgId:%d, enqueue msg error, FpEqMsg is NULL", pSyncNode->vgId); terrno = TSDB_CODE_SYN_INTERNAL_ERROR; return -1; } } return 0; } int32_t syncNodePropose(SSyncNode* pSyncNode, SRpcMsg* pMsg, bool isWeak) { int32_t ret = 0; char eventLog[128]; snprintf(eventLog, sizeof(eventLog), "propose type:%s,%d", TMSG_INFO(pMsg->msgType), pMsg->msgType); syncNodeEventLog(pSyncNode, eventLog); if (pSyncNode->state == TAOS_SYNC_STATE_LEADER) { if (pSyncNode->changing && pMsg->msgType != TDMT_SYNC_CONFIG_CHANGE_FINISH) { ret = -1; terrno = TSDB_CODE_SYN_PROPOSE_NOT_READY; sError("vgId:%d, sync propose not ready, type:%s,%d", pSyncNode->vgId, TMSG_INFO(pMsg->msgType), pMsg->msgType); goto _END; } // config change if (pMsg->msgType == TDMT_SYNC_CONFIG_CHANGE) { if (!syncNodeCanChange(pSyncNode)) { ret = -1; terrno = TSDB_CODE_SYN_RECONFIG_NOT_READY; sError("vgId:%d, sync reconfig not ready, type:%s,%d", pSyncNode->vgId, TMSG_INFO(pMsg->msgType), pMsg->msgType); goto _END; } ASSERT(!pSyncNode->changing); pSyncNode->changing = true; } SRespStub stub; stub.createTime = taosGetTimestampMs(); stub.rpcMsg = *pMsg; uint64_t seqNum = syncRespMgrAdd(pSyncNode->pSyncRespMgr, &stub); SyncClientRequest* pSyncMsg = syncClientRequestBuild2(pMsg, seqNum, isWeak, pSyncNode->vgId); SRpcMsg rpcMsg; syncClientRequest2RpcMsg(pSyncMsg, &rpcMsg); // optimized one replica if (syncNodeIsOptimizedOneReplica(pSyncNode, pMsg)) { SyncIndex retIndex; int32_t code = syncNodeOnClientRequestCb(pSyncNode, pSyncMsg, &retIndex); if (code == 0) { pMsg->info.conn.applyIndex = retIndex; pMsg->info.conn.applyTerm = pSyncNode->pRaftStore->currentTerm; rpcFreeCont(rpcMsg.pCont); syncRespMgrDel(pSyncNode->pSyncRespMgr, seqNum); ret = 1; sDebug("vgId:%d optimized index:%" PRId64 " success, msgtype:%s,%d", pSyncNode->vgId, retIndex, TMSG_INFO(pMsg->msgType), pMsg->msgType); } else { ret = -1; terrno = TSDB_CODE_SYN_INTERNAL_ERROR; sError("vgId:%d optimized index:%" PRId64 " error, msgtype:%s,%d", pSyncNode->vgId, retIndex, TMSG_INFO(pMsg->msgType), pMsg->msgType); } } else { if (pSyncNode->FpEqMsg != NULL && (*pSyncNode->FpEqMsg)(pSyncNode->msgcb, &rpcMsg) == 0) { ret = 0; } else { ret = -1; terrno = TSDB_CODE_SYN_INTERNAL_ERROR; sError("vgId:%d, enqueue msg error, FpEqMsg is NULL", pSyncNode->vgId); } } syncClientRequestDestroy(pSyncMsg); goto _END; } else { ret = -1; terrno = TSDB_CODE_SYN_NOT_LEADER; sError("vgId:%d, sync propose not leader, %s", pSyncNode->vgId, syncUtilState2String(pSyncNode->state)); goto _END; } _END: return ret; } // open/close -------------- SSyncNode* syncNodeOpen(const SSyncInfo* pOldSyncInfo) { SSyncInfo* pSyncInfo = (SSyncInfo*)pOldSyncInfo; SSyncNode* pSyncNode = (SSyncNode*)taosMemoryMalloc(sizeof(SSyncNode)); ASSERT(pSyncNode != NULL); memset(pSyncNode, 0, sizeof(SSyncNode)); int32_t ret = 0; if (!taosDirExist((char*)(pSyncInfo->path))) { if (taosMkDir(pSyncInfo->path) != 0) { terrno = TAOS_SYSTEM_ERROR(errno); sError("failed to create dir:%s since %s", pSyncInfo->path, terrstr()); return NULL; } } snprintf(pSyncNode->configPath, sizeof(pSyncNode->configPath), "%s/raft_config.json", pSyncInfo->path); if (!taosCheckExistFile(pSyncNode->configPath)) { // create a new raft config file SRaftCfgMeta meta; meta.isStandBy = pSyncInfo->isStandBy; meta.snapshotStrategy = pSyncInfo->snapshotStrategy; meta.lastConfigIndex = SYNC_INDEX_INVALID; meta.batchSize = pSyncInfo->batchSize; ret = raftCfgCreateFile((SSyncCfg*)&(pSyncInfo->syncCfg), meta, pSyncNode->configPath); ASSERT(ret == 0); } else { // update syncCfg by raft_config.json pSyncNode->pRaftCfg = raftCfgOpen(pSyncNode->configPath); ASSERT(pSyncNode->pRaftCfg != NULL); pSyncInfo->syncCfg = pSyncNode->pRaftCfg->cfg; if (gRaftDetailLog) { char* seralized = raftCfg2Str(pSyncNode->pRaftCfg); sInfo("syncNodeOpen update config :%s", seralized); taosMemoryFree(seralized); } raftCfgClose(pSyncNode->pRaftCfg); } // init by SSyncInfo pSyncNode->vgId = pSyncInfo->vgId; memcpy(pSyncNode->path, pSyncInfo->path, sizeof(pSyncNode->path)); snprintf(pSyncNode->raftStorePath, sizeof(pSyncNode->raftStorePath), "%s/raft_store.json", pSyncInfo->path); snprintf(pSyncNode->configPath, sizeof(pSyncNode->configPath), "%s/raft_config.json", pSyncInfo->path); pSyncNode->pWal = pSyncInfo->pWal; pSyncNode->msgcb = pSyncInfo->msgcb; pSyncNode->FpSendMsg = pSyncInfo->FpSendMsg; pSyncNode->FpEqMsg = pSyncInfo->FpEqMsg; // init raft config pSyncNode->pRaftCfg = raftCfgOpen(pSyncNode->configPath); ASSERT(pSyncNode->pRaftCfg != NULL); // init internal pSyncNode->myNodeInfo = pSyncNode->pRaftCfg->cfg.nodeInfo[pSyncNode->pRaftCfg->cfg.myIndex]; syncUtilnodeInfo2raftId(&pSyncNode->myNodeInfo, pSyncNode->vgId, &pSyncNode->myRaftId); // init peersNum, peers, peersId pSyncNode->peersNum = pSyncNode->pRaftCfg->cfg.replicaNum - 1; int j = 0; for (int i = 0; i < pSyncNode->pRaftCfg->cfg.replicaNum; ++i) { if (i != pSyncNode->pRaftCfg->cfg.myIndex) { pSyncNode->peersNodeInfo[j] = pSyncNode->pRaftCfg->cfg.nodeInfo[i]; j++; } } for (int i = 0; i < pSyncNode->peersNum; ++i) { syncUtilnodeInfo2raftId(&pSyncNode->peersNodeInfo[i], pSyncNode->vgId, &pSyncNode->peersId[i]); } // init replicaNum, replicasId pSyncNode->replicaNum = pSyncNode->pRaftCfg->cfg.replicaNum; for (int i = 0; i < pSyncNode->pRaftCfg->cfg.replicaNum; ++i) { syncUtilnodeInfo2raftId(&pSyncNode->pRaftCfg->cfg.nodeInfo[i], pSyncNode->vgId, &pSyncNode->replicasId[i]); } // init raft algorithm pSyncNode->pFsm = pSyncInfo->pFsm; pSyncNode->quorum = syncUtilQuorum(pSyncNode->pRaftCfg->cfg.replicaNum); pSyncNode->leaderCache = EMPTY_RAFT_ID; // init life cycle outside // TLA+ Spec // InitHistoryVars == /\ elections = {} // /\ allLogs = {} // /\ voterLog = [i \in Server |-> [j \in {} |-> <<>>]] // InitServerVars == /\ currentTerm = [i \in Server |-> 1] // /\ state = [i \in Server |-> Follower] // /\ votedFor = [i \in Server |-> Nil] // InitCandidateVars == /\ votesResponded = [i \in Server |-> {}] // /\ votesGranted = [i \in Server |-> {}] // \* The values nextIndex[i][i] and matchIndex[i][i] are never read, since the // \* leader does not send itself messages. It's still easier to include these // \* in the functions. // InitLeaderVars == /\ nextIndex = [i \in Server |-> [j \in Server |-> 1]] // /\ matchIndex = [i \in Server |-> [j \in Server |-> 0]] // InitLogVars == /\ log = [i \in Server |-> << >>] // /\ commitIndex = [i \in Server |-> 0] // Init == /\ messages = [m \in {} |-> 0] // /\ InitHistoryVars // /\ InitServerVars // /\ InitCandidateVars // /\ InitLeaderVars // /\ InitLogVars // // init TLA+ server vars pSyncNode->state = TAOS_SYNC_STATE_FOLLOWER; pSyncNode->pRaftStore = raftStoreOpen(pSyncNode->raftStorePath); ASSERT(pSyncNode->pRaftStore != NULL); // init TLA+ candidate vars pSyncNode->pVotesGranted = voteGrantedCreate(pSyncNode); ASSERT(pSyncNode->pVotesGranted != NULL); pSyncNode->pVotesRespond = votesRespondCreate(pSyncNode); ASSERT(pSyncNode->pVotesRespond != NULL); // init TLA+ leader vars pSyncNode->pNextIndex = syncIndexMgrCreate(pSyncNode); ASSERT(pSyncNode->pNextIndex != NULL); pSyncNode->pMatchIndex = syncIndexMgrCreate(pSyncNode); ASSERT(pSyncNode->pMatchIndex != NULL); // init TLA+ log vars pSyncNode->pLogStore = logStoreCreate(pSyncNode); ASSERT(pSyncNode->pLogStore != NULL); pSyncNode->commitIndex = SYNC_INDEX_INVALID; // timer ms init pSyncNode->pingBaseLine = PING_TIMER_MS; pSyncNode->electBaseLine = ELECT_TIMER_MS_MIN; pSyncNode->hbBaseLine = HEARTBEAT_TIMER_MS; // init ping timer pSyncNode->pPingTimer = NULL; pSyncNode->pingTimerMS = pSyncNode->pingBaseLine; atomic_store_64(&pSyncNode->pingTimerLogicClock, 0); atomic_store_64(&pSyncNode->pingTimerLogicClockUser, 0); pSyncNode->FpPingTimerCB = syncNodeEqPingTimer; pSyncNode->pingTimerCounter = 0; // init elect timer pSyncNode->pElectTimer = NULL; pSyncNode->electTimerMS = syncUtilElectRandomMS(pSyncNode->electBaseLine, 2 * pSyncNode->electBaseLine); atomic_store_64(&pSyncNode->electTimerLogicClock, 0); atomic_store_64(&pSyncNode->electTimerLogicClockUser, 0); pSyncNode->FpElectTimerCB = syncNodeEqElectTimer; pSyncNode->electTimerCounter = 0; // init heartbeat timer pSyncNode->pHeartbeatTimer = NULL; pSyncNode->heartbeatTimerMS = pSyncNode->hbBaseLine; atomic_store_64(&pSyncNode->heartbeatTimerLogicClock, 0); atomic_store_64(&pSyncNode->heartbeatTimerLogicClockUser, 0); pSyncNode->FpHeartbeatTimerCB = syncNodeEqHeartbeatTimer; pSyncNode->heartbeatTimerCounter = 0; // init callback pSyncNode->FpOnPing = syncNodeOnPingCb; pSyncNode->FpOnPingReply = syncNodeOnPingReplyCb; pSyncNode->FpOnClientRequest = syncNodeOnClientRequestCb; pSyncNode->FpOnTimeout = syncNodeOnTimeoutCb; pSyncNode->FpOnSnapshotSend = syncNodeOnSnapshotSendCb; pSyncNode->FpOnSnapshotRsp = syncNodeOnSnapshotRspCb; if (pSyncNode->pRaftCfg->snapshotStrategy) { sInfo("vgId:%d, sync node use snapshot", pSyncNode->vgId); pSyncNode->FpOnRequestVote = syncNodeOnRequestVoteSnapshotCb; pSyncNode->FpOnRequestVoteReply = syncNodeOnRequestVoteReplySnapshotCb; pSyncNode->FpOnAppendEntries = syncNodeOnAppendEntriesSnapshotCb; pSyncNode->FpOnAppendEntriesReply = syncNodeOnAppendEntriesReplySnapshotCb; } else { sInfo("vgId:%d, sync node do not use snapshot", pSyncNode->vgId); pSyncNode->FpOnRequestVote = syncNodeOnRequestVoteCb; pSyncNode->FpOnRequestVoteReply = syncNodeOnRequestVoteReplyCb; pSyncNode->FpOnAppendEntries = syncNodeOnAppendEntriesCb; pSyncNode->FpOnAppendEntriesReply = syncNodeOnAppendEntriesReplyCb; } // tools pSyncNode->pSyncRespMgr = syncRespMgrCreate(pSyncNode, 0); ASSERT(pSyncNode->pSyncRespMgr != NULL); // restore state pSyncNode->restoreFinish = false; // pSyncNode->pSnapshot = NULL; // if (pSyncNode->pFsm->FpGetSnapshotInfo != NULL) { // pSyncNode->pSnapshot = taosMemoryMalloc(sizeof(SSnapshot)); // pSyncNode->pFsm->FpGetSnapshotInfo(pSyncNode->pFsm, pSyncNode->pSnapshot); // } // tsem_init(&(pSyncNode->restoreSem), 0, 0); // snapshot senders for (int i = 0; i < TSDB_MAX_REPLICA; ++i) { SSyncSnapshotSender* pSender = snapshotSenderCreate(pSyncNode, i); // ASSERT(pSender != NULL); (pSyncNode->senders)[i] = pSender; } // snapshot receivers pSyncNode->pNewNodeReceiver = snapshotReceiverCreate(pSyncNode, EMPTY_RAFT_ID); // is config changing pSyncNode->changing = false; // start in syncNodeStart // start raft // syncNodeBecomeFollower(pSyncNode); syncNodeEventLog(pSyncNode, "sync open"); return pSyncNode; } void syncNodeStart(SSyncNode* pSyncNode) { // start raft if (pSyncNode->replicaNum == 1) { raftStoreNextTerm(pSyncNode->pRaftStore); syncNodeBecomeLeader(pSyncNode, "one replica start"); // Raft 3.6.2 Committing entries from previous terms syncNodeAppendNoop(pSyncNode); syncMaybeAdvanceCommitIndex(pSyncNode); } else { syncNodeBecomeFollower(pSyncNode, "first start"); } int32_t ret = 0; ret = syncNodeStartPingTimer(pSyncNode); ASSERT(ret == 0); } void syncNodeStartStandBy(SSyncNode* pSyncNode) { // state change pSyncNode->state = TAOS_SYNC_STATE_FOLLOWER; syncNodeStopHeartbeatTimer(pSyncNode); // reset elect timer, long enough int32_t electMS = TIMER_MAX_MS; int32_t ret = syncNodeRestartElectTimer(pSyncNode, electMS); ASSERT(ret == 0); } void syncNodeClose(SSyncNode* pSyncNode) { syncNodeEventLog(pSyncNode, "sync close"); int32_t ret; ASSERT(pSyncNode != NULL); ret = raftStoreClose(pSyncNode->pRaftStore); ASSERT(ret == 0); syncRespMgrDestroy(pSyncNode->pSyncRespMgr); voteGrantedDestroy(pSyncNode->pVotesGranted); votesRespondDestory(pSyncNode->pVotesRespond); syncIndexMgrDestroy(pSyncNode->pNextIndex); syncIndexMgrDestroy(pSyncNode->pMatchIndex); logStoreDestory(pSyncNode->pLogStore); raftCfgClose(pSyncNode->pRaftCfg); syncNodeStopPingTimer(pSyncNode); syncNodeStopElectTimer(pSyncNode); syncNodeStopHeartbeatTimer(pSyncNode); if (pSyncNode->pFsm != NULL) { taosMemoryFree(pSyncNode->pFsm); } for (int i = 0; i < TSDB_MAX_REPLICA; ++i) { if ((pSyncNode->senders)[i] != NULL) { snapshotSenderDestroy((pSyncNode->senders)[i]); (pSyncNode->senders)[i] = NULL; } } if (pSyncNode->pNewNodeReceiver != NULL) { snapshotReceiverDestroy(pSyncNode->pNewNodeReceiver); pSyncNode->pNewNodeReceiver = NULL; } // free memory in syncFreeNode // taosMemoryFree(pSyncNode); } // option // bool syncNodeSnapshotEnable(SSyncNode* pSyncNode) { return pSyncNode->pRaftCfg->snapshotEnable; } ESyncStrategy syncNodeStrategy(SSyncNode* pSyncNode) { return pSyncNode->pRaftCfg->snapshotStrategy; } // ping -------------- int32_t syncNodePing(SSyncNode* pSyncNode, const SRaftId* destRaftId, SyncPing* pMsg) { syncPingLog2((char*)"==syncNodePing==", pMsg); int32_t ret = 0; SRpcMsg rpcMsg; syncPing2RpcMsg(pMsg, &rpcMsg); syncRpcMsgLog2((char*)"==syncNodePing==", &rpcMsg); ret = syncNodeSendMsgById(destRaftId, pSyncNode, &rpcMsg); return ret; } int32_t syncNodePingSelf(SSyncNode* pSyncNode) { int32_t ret = 0; SyncPing* pMsg = syncPingBuild3(&pSyncNode->myRaftId, &pSyncNode->myRaftId, pSyncNode->vgId); ret = syncNodePing(pSyncNode, &pMsg->destId, pMsg); ASSERT(ret == 0); syncPingDestroy(pMsg); return ret; } int32_t syncNodePingPeers(SSyncNode* pSyncNode) { int32_t ret = 0; for (int i = 0; i < pSyncNode->peersNum; ++i) { SRaftId* destId = &(pSyncNode->peersId[i]); SyncPing* pMsg = syncPingBuild3(&pSyncNode->myRaftId, destId, pSyncNode->vgId); ret = syncNodePing(pSyncNode, destId, pMsg); ASSERT(ret == 0); syncPingDestroy(pMsg); } return ret; } int32_t syncNodePingAll(SSyncNode* pSyncNode) { int32_t ret = 0; for (int i = 0; i < pSyncNode->pRaftCfg->cfg.replicaNum; ++i) { SRaftId* destId = &(pSyncNode->replicasId[i]); SyncPing* pMsg = syncPingBuild3(&pSyncNode->myRaftId, destId, pSyncNode->vgId); ret = syncNodePing(pSyncNode, destId, pMsg); ASSERT(ret == 0); syncPingDestroy(pMsg); } return ret; } // timer control -------------- int32_t syncNodeStartPingTimer(SSyncNode* pSyncNode) { int32_t ret = 0; if (syncEnvIsStart()) { taosTmrReset(pSyncNode->FpPingTimerCB, pSyncNode->pingTimerMS, pSyncNode, gSyncEnv->pTimerManager, &pSyncNode->pPingTimer); atomic_store_64(&pSyncNode->pingTimerLogicClock, pSyncNode->pingTimerLogicClockUser); } else { sError("vgId:%d, start ping timer error, sync env is stop", pSyncNode->vgId); } return ret; } int32_t syncNodeStopPingTimer(SSyncNode* pSyncNode) { int32_t ret = 0; atomic_add_fetch_64(&pSyncNode->pingTimerLogicClockUser, 1); taosTmrStop(pSyncNode->pPingTimer); pSyncNode->pPingTimer = NULL; return ret; } int32_t syncNodeStartElectTimer(SSyncNode* pSyncNode, int32_t ms) { int32_t ret = 0; if (syncEnvIsStart()) { pSyncNode->electTimerMS = ms; taosTmrReset(pSyncNode->FpElectTimerCB, pSyncNode->electTimerMS, pSyncNode, gSyncEnv->pTimerManager, &pSyncNode->pElectTimer); atomic_store_64(&pSyncNode->electTimerLogicClock, pSyncNode->electTimerLogicClockUser); } else { sError("vgId:%d, start elect timer error, sync env is stop", pSyncNode->vgId); } return ret; } int32_t syncNodeStopElectTimer(SSyncNode* pSyncNode) { int32_t ret = 0; atomic_add_fetch_64(&pSyncNode->electTimerLogicClockUser, 1); taosTmrStop(pSyncNode->pElectTimer); pSyncNode->pElectTimer = NULL; return ret; } int32_t syncNodeRestartElectTimer(SSyncNode* pSyncNode, int32_t ms) { int32_t ret = 0; syncNodeStopElectTimer(pSyncNode); syncNodeStartElectTimer(pSyncNode, ms); return ret; } int32_t syncNodeResetElectTimer(SSyncNode* pSyncNode) { int32_t ret = 0; int32_t electMS; if (pSyncNode->pRaftCfg->isStandBy) { electMS = TIMER_MAX_MS; } else { electMS = syncUtilElectRandomMS(pSyncNode->electBaseLine, 2 * pSyncNode->electBaseLine); } ret = syncNodeRestartElectTimer(pSyncNode, electMS); return ret; } int32_t syncNodeStartHeartbeatTimer(SSyncNode* pSyncNode) { int32_t ret = 0; if (syncEnvIsStart()) { taosTmrReset(pSyncNode->FpHeartbeatTimerCB, pSyncNode->heartbeatTimerMS, pSyncNode, gSyncEnv->pTimerManager, &pSyncNode->pHeartbeatTimer); atomic_store_64(&pSyncNode->heartbeatTimerLogicClock, pSyncNode->heartbeatTimerLogicClockUser); } else { sError("vgId:%d, start heartbeat timer error, sync env is stop", pSyncNode->vgId); } return ret; } int32_t syncNodeStopHeartbeatTimer(SSyncNode* pSyncNode) { int32_t ret = 0; atomic_add_fetch_64(&pSyncNode->heartbeatTimerLogicClockUser, 1); taosTmrStop(pSyncNode->pHeartbeatTimer); pSyncNode->pHeartbeatTimer = NULL; return ret; } // utils -------------- int32_t syncNodeSendMsgById(const SRaftId* destRaftId, SSyncNode* pSyncNode, SRpcMsg* pMsg) { SEpSet epSet; syncUtilraftId2EpSet(destRaftId, &epSet); if (pSyncNode->FpSendMsg != NULL) { if (gRaftDetailLog) { char* JsonStr = syncRpcMsg2Str(pMsg); syncUtilJson2Line(JsonStr); sTrace("sync send msg, vgId:%d, type:%d, msg:%s", pSyncNode->vgId, pMsg->msgType, JsonStr); taosMemoryFree(JsonStr); } // htonl syncUtilMsgHtoN(pMsg->pCont); pMsg->info.noResp = 1; pSyncNode->FpSendMsg(&epSet, pMsg); } else { sError("vgId:%d, sync send msg by id error, fp-send-msg is null", pSyncNode->vgId); return -1; } return 0; } int32_t syncNodeSendMsgByInfo(const SNodeInfo* nodeInfo, SSyncNode* pSyncNode, SRpcMsg* pMsg) { SEpSet epSet; syncUtilnodeInfo2EpSet(nodeInfo, &epSet); if (pSyncNode->FpSendMsg != NULL) { // htonl syncUtilMsgHtoN(pMsg->pCont); pMsg->info.noResp = 1; pSyncNode->FpSendMsg(&epSet, pMsg); } else { sError("vgId:%d, sync send msg by info error, fp-send-msg is null", pSyncNode->vgId); } return 0; } cJSON* syncNode2Json(const SSyncNode* pSyncNode) { char u64buf[128] = {0}; cJSON* pRoot = cJSON_CreateObject(); if (pSyncNode != NULL) { // init by SSyncInfo cJSON_AddNumberToObject(pRoot, "vgId", pSyncNode->vgId); cJSON_AddItemToObject(pRoot, "SRaftCfg", raftCfg2Json(pSyncNode->pRaftCfg)); cJSON_AddStringToObject(pRoot, "path", pSyncNode->path); cJSON_AddStringToObject(pRoot, "raftStorePath", pSyncNode->raftStorePath); cJSON_AddStringToObject(pRoot, "configPath", pSyncNode->configPath); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->pWal); cJSON_AddStringToObject(pRoot, "pWal", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->msgcb); cJSON_AddStringToObject(pRoot, "rpcClient", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpSendMsg); cJSON_AddStringToObject(pRoot, "FpSendMsg", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->msgcb); cJSON_AddStringToObject(pRoot, "queue", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpEqMsg); cJSON_AddStringToObject(pRoot, "FpEqMsg", u64buf); // init internal cJSON* pMe = syncUtilNodeInfo2Json(&pSyncNode->myNodeInfo); cJSON_AddItemToObject(pRoot, "myNodeInfo", pMe); cJSON* pRaftId = syncUtilRaftId2Json(&pSyncNode->myRaftId); cJSON_AddItemToObject(pRoot, "myRaftId", pRaftId); cJSON_AddNumberToObject(pRoot, "peersNum", pSyncNode->peersNum); cJSON* pPeers = cJSON_CreateArray(); cJSON_AddItemToObject(pRoot, "peersNodeInfo", pPeers); for (int i = 0; i < pSyncNode->peersNum; ++i) { cJSON_AddItemToArray(pPeers, syncUtilNodeInfo2Json(&pSyncNode->peersNodeInfo[i])); } cJSON* pPeersId = cJSON_CreateArray(); cJSON_AddItemToObject(pRoot, "peersId", pPeersId); for (int i = 0; i < pSyncNode->peersNum; ++i) { cJSON_AddItemToArray(pPeersId, syncUtilRaftId2Json(&pSyncNode->peersId[i])); } cJSON_AddNumberToObject(pRoot, "replicaNum", pSyncNode->replicaNum); cJSON* pReplicasId = cJSON_CreateArray(); cJSON_AddItemToObject(pRoot, "replicasId", pReplicasId); for (int i = 0; i < pSyncNode->replicaNum; ++i) { cJSON_AddItemToArray(pReplicasId, syncUtilRaftId2Json(&pSyncNode->replicasId[i])); } // raft algorithm snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->pFsm); cJSON_AddStringToObject(pRoot, "pFsm", u64buf); cJSON_AddNumberToObject(pRoot, "quorum", pSyncNode->quorum); cJSON* pLaderCache = syncUtilRaftId2Json(&pSyncNode->leaderCache); cJSON_AddItemToObject(pRoot, "leaderCache", pLaderCache); // life cycle snprintf(u64buf, sizeof(u64buf), "%" PRId64, pSyncNode->rid); cJSON_AddStringToObject(pRoot, "rid", u64buf); // tla+ server vars cJSON_AddNumberToObject(pRoot, "state", pSyncNode->state); cJSON_AddStringToObject(pRoot, "state_str", syncUtilState2String(pSyncNode->state)); cJSON_AddItemToObject(pRoot, "pRaftStore", raftStore2Json(pSyncNode->pRaftStore)); // tla+ candidate vars cJSON_AddItemToObject(pRoot, "pVotesGranted", voteGranted2Json(pSyncNode->pVotesGranted)); cJSON_AddItemToObject(pRoot, "pVotesRespond", votesRespond2Json(pSyncNode->pVotesRespond)); // tla+ leader vars cJSON_AddItemToObject(pRoot, "pNextIndex", syncIndexMgr2Json(pSyncNode->pNextIndex)); cJSON_AddItemToObject(pRoot, "pMatchIndex", syncIndexMgr2Json(pSyncNode->pMatchIndex)); // tla+ log vars cJSON_AddItemToObject(pRoot, "pLogStore", logStore2Json(pSyncNode->pLogStore)); snprintf(u64buf, sizeof(u64buf), "%" PRId64, pSyncNode->commitIndex); cJSON_AddStringToObject(pRoot, "commitIndex", u64buf); // timer ms init cJSON_AddNumberToObject(pRoot, "pingBaseLine", pSyncNode->pingBaseLine); cJSON_AddNumberToObject(pRoot, "electBaseLine", pSyncNode->electBaseLine); cJSON_AddNumberToObject(pRoot, "hbBaseLine", pSyncNode->hbBaseLine); // ping timer snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->pPingTimer); cJSON_AddStringToObject(pRoot, "pPingTimer", u64buf); cJSON_AddNumberToObject(pRoot, "pingTimerMS", pSyncNode->pingTimerMS); snprintf(u64buf, sizeof(u64buf), "%" PRIu64, pSyncNode->pingTimerLogicClock); cJSON_AddStringToObject(pRoot, "pingTimerLogicClock", u64buf); snprintf(u64buf, sizeof(u64buf), "%" PRIu64, pSyncNode->pingTimerLogicClockUser); cJSON_AddStringToObject(pRoot, "pingTimerLogicClockUser", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpPingTimerCB); cJSON_AddStringToObject(pRoot, "FpPingTimerCB", u64buf); snprintf(u64buf, sizeof(u64buf), "%" PRIu64, pSyncNode->pingTimerCounter); cJSON_AddStringToObject(pRoot, "pingTimerCounter", u64buf); // elect timer snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->pElectTimer); cJSON_AddStringToObject(pRoot, "pElectTimer", u64buf); cJSON_AddNumberToObject(pRoot, "electTimerMS", pSyncNode->electTimerMS); snprintf(u64buf, sizeof(u64buf), "%" PRIu64, pSyncNode->electTimerLogicClock); cJSON_AddStringToObject(pRoot, "electTimerLogicClock", u64buf); snprintf(u64buf, sizeof(u64buf), "%" PRIu64, pSyncNode->electTimerLogicClockUser); cJSON_AddStringToObject(pRoot, "electTimerLogicClockUser", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpElectTimerCB); cJSON_AddStringToObject(pRoot, "FpElectTimerCB", u64buf); snprintf(u64buf, sizeof(u64buf), "%" PRIu64, pSyncNode->electTimerCounter); cJSON_AddStringToObject(pRoot, "electTimerCounter", u64buf); // heartbeat timer snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->pHeartbeatTimer); cJSON_AddStringToObject(pRoot, "pHeartbeatTimer", u64buf); cJSON_AddNumberToObject(pRoot, "heartbeatTimerMS", pSyncNode->heartbeatTimerMS); snprintf(u64buf, sizeof(u64buf), "%" PRIu64, pSyncNode->heartbeatTimerLogicClock); cJSON_AddStringToObject(pRoot, "heartbeatTimerLogicClock", u64buf); snprintf(u64buf, sizeof(u64buf), "%" PRIu64, pSyncNode->heartbeatTimerLogicClockUser); cJSON_AddStringToObject(pRoot, "heartbeatTimerLogicClockUser", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpHeartbeatTimerCB); cJSON_AddStringToObject(pRoot, "FpHeartbeatTimerCB", u64buf); snprintf(u64buf, sizeof(u64buf), "%" PRIu64, pSyncNode->heartbeatTimerCounter); cJSON_AddStringToObject(pRoot, "heartbeatTimerCounter", u64buf); // callback snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpOnPing); cJSON_AddStringToObject(pRoot, "FpOnPing", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpOnPingReply); cJSON_AddStringToObject(pRoot, "FpOnPingReply", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpOnRequestVote); cJSON_AddStringToObject(pRoot, "FpOnRequestVote", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpOnRequestVoteReply); cJSON_AddStringToObject(pRoot, "FpOnRequestVoteReply", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpOnAppendEntries); cJSON_AddStringToObject(pRoot, "FpOnAppendEntries", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpOnAppendEntriesReply); cJSON_AddStringToObject(pRoot, "FpOnAppendEntriesReply", u64buf); snprintf(u64buf, sizeof(u64buf), "%p", pSyncNode->FpOnTimeout); cJSON_AddStringToObject(pRoot, "FpOnTimeout", u64buf); // restoreFinish cJSON_AddNumberToObject(pRoot, "restoreFinish", pSyncNode->restoreFinish); // snapshot senders cJSON* pSenders = cJSON_CreateArray(); cJSON_AddItemToObject(pRoot, "senders", pSenders); for (int i = 0; i < TSDB_MAX_REPLICA; ++i) { cJSON_AddItemToArray(pSenders, snapshotSender2Json((pSyncNode->senders)[i])); } // snapshot receivers cJSON* pReceivers = cJSON_CreateArray(); cJSON_AddItemToObject(pRoot, "receiver", snapshotReceiver2Json(pSyncNode->pNewNodeReceiver)); // changing cJSON_AddNumberToObject(pRoot, "changing", pSyncNode->changing); } cJSON* pJson = cJSON_CreateObject(); cJSON_AddItemToObject(pJson, "SSyncNode", pRoot); return pJson; } char* syncNode2Str(const SSyncNode* pSyncNode) { cJSON* pJson = syncNode2Json(pSyncNode); char* serialized = cJSON_Print(pJson); cJSON_Delete(pJson); return serialized; } void syncNodeEventLog(const SSyncNode* pSyncNode, char* str) { int32_t userStrLen = strlen(str); SSnapshot snapshot = {.data = NULL, .lastApplyIndex = -1, .lastApplyTerm = 0}; if (pSyncNode->pFsm != NULL && pSyncNode->pFsm->FpGetSnapshotInfo != NULL) { pSyncNode->pFsm->FpGetSnapshotInfo(pSyncNode->pFsm, &snapshot); } SyncIndex logLastIndex = SYNC_INDEX_INVALID; SyncIndex logBeginIndex = SYNC_INDEX_INVALID; if (pSyncNode->pLogStore != NULL) { logLastIndex = pSyncNode->pLogStore->syncLogLastIndex(pSyncNode->pLogStore); logBeginIndex = pSyncNode->pLogStore->syncLogBeginIndex(pSyncNode->pLogStore); } char* pCfgStr = syncCfg2SimpleStr(&(pSyncNode->pRaftCfg->cfg)); char* printStr = ""; if (pCfgStr != NULL) { printStr = pCfgStr; } if (userStrLen < 256) { char logBuf[256 + 256]; if (pSyncNode != NULL && pSyncNode->pRaftCfg != NULL && pSyncNode->pRaftStore != NULL) { snprintf(logBuf, sizeof(logBuf), "vgId:%d, sync %s %s, term:%" PRIu64 ", commit:%" PRId64 ", first:%" PRId64 ", last:%" PRId64 ", snapshot:%" PRId64 ", standby:%d, " "strategy:%d, batch:%d, " "replica-num:%d, " "lconfig:%" PRId64 ", changing:%d, restore:%d, %s", pSyncNode->vgId, syncUtilState2String(pSyncNode->state), str, pSyncNode->pRaftStore->currentTerm, pSyncNode->commitIndex, logBeginIndex, logLastIndex, snapshot.lastApplyIndex, pSyncNode->pRaftCfg->isStandBy, pSyncNode->pRaftCfg->snapshotStrategy, pSyncNode->pRaftCfg->batchSize, pSyncNode->replicaNum, pSyncNode->pRaftCfg->lastConfigIndex, pSyncNode->changing, pSyncNode->restoreFinish, printStr); } else { snprintf(logBuf, sizeof(logBuf), "%s", str); } sDebug("%s", logBuf); } else { int len = 256 + userStrLen; char* s = (char*)taosMemoryMalloc(len); if (pSyncNode != NULL && pSyncNode->pRaftCfg != NULL && pSyncNode->pRaftStore != NULL) { snprintf(s, len, "vgId:%d, sync %s %s, term:%" PRIu64 ", commit:%" PRId64 ", first:%" PRId64 ", last:%" PRId64 ", snapshot:%" PRId64 ", standby:%d, " "strategy:%d, batch:%d, " "replica-num:%d, " "lconfig:%" PRId64 ", changing:%d, restore:%d, %s", pSyncNode->vgId, syncUtilState2String(pSyncNode->state), str, pSyncNode->pRaftStore->currentTerm, pSyncNode->commitIndex, logBeginIndex, logLastIndex, snapshot.lastApplyIndex, pSyncNode->pRaftCfg->isStandBy, pSyncNode->pRaftCfg->snapshotStrategy, pSyncNode->pRaftCfg->batchSize, pSyncNode->replicaNum, pSyncNode->pRaftCfg->lastConfigIndex, pSyncNode->changing, pSyncNode->restoreFinish, printStr); } else { snprintf(s, len, "%s", str); } sDebug("%s", s); taosMemoryFree(s); } taosMemoryFree(pCfgStr); } void syncNodeErrorLog(const SSyncNode* pSyncNode, char* str) { int32_t userStrLen = strlen(str); SSnapshot snapshot = {.data = NULL, .lastApplyIndex = -1, .lastApplyTerm = 0}; if (pSyncNode->pFsm != NULL && pSyncNode->pFsm->FpGetSnapshotInfo != NULL) { pSyncNode->pFsm->FpGetSnapshotInfo(pSyncNode->pFsm, &snapshot); } SyncIndex logLastIndex = SYNC_INDEX_INVALID; SyncIndex logBeginIndex = SYNC_INDEX_INVALID; if (pSyncNode->pLogStore != NULL) { logLastIndex = pSyncNode->pLogStore->syncLogLastIndex(pSyncNode->pLogStore); logBeginIndex = pSyncNode->pLogStore->syncLogBeginIndex(pSyncNode->pLogStore); } char* pCfgStr = syncCfg2SimpleStr(&(pSyncNode->pRaftCfg->cfg)); char* printStr = ""; if (pCfgStr != NULL) { printStr = pCfgStr; } if (userStrLen < 256) { char logBuf[256 + 256]; if (pSyncNode != NULL && pSyncNode->pRaftCfg != NULL && pSyncNode->pRaftStore != NULL) { snprintf(logBuf, sizeof(logBuf), "vgId:%d, sync %s %s, term:%" PRIu64 ", commit:%" PRId64 ", first:%" PRId64 ", last:%" PRId64 ", snapshot:%" PRId64 ", standby:%d, " "replica-num:%d, " "lconfig:%" PRId64 ", changing:%d, restore:%d, %s", pSyncNode->vgId, syncUtilState2String(pSyncNode->state), str, pSyncNode->pRaftStore->currentTerm, pSyncNode->commitIndex, logBeginIndex, logLastIndex, snapshot.lastApplyIndex, pSyncNode->pRaftCfg->isStandBy, pSyncNode->replicaNum, pSyncNode->pRaftCfg->lastConfigIndex, pSyncNode->changing, pSyncNode->restoreFinish, printStr); } else { snprintf(logBuf, sizeof(logBuf), "%s", str); } sError("%s", logBuf); } else { int len = 256 + userStrLen; char* s = (char*)taosMemoryMalloc(len); if (pSyncNode != NULL && pSyncNode->pRaftCfg != NULL && pSyncNode->pRaftStore != NULL) { snprintf(s, len, "vgId:%d, sync %s %s, term:%" PRIu64 ", commit:%" PRId64 ", first:%" PRId64 ", last:%" PRId64 ", snapshot:%" PRId64 ", standby:%d, " "replica-num:%d, " "lconfig:%" PRId64 ", changing:%d, restore:%d, %s", pSyncNode->vgId, syncUtilState2String(pSyncNode->state), str, pSyncNode->pRaftStore->currentTerm, pSyncNode->commitIndex, logBeginIndex, logLastIndex, snapshot.lastApplyIndex, pSyncNode->pRaftCfg->isStandBy, pSyncNode->replicaNum, pSyncNode->pRaftCfg->lastConfigIndex, pSyncNode->changing, pSyncNode->restoreFinish, printStr); } else { snprintf(s, len, "%s", str); } sError("%s", s); taosMemoryFree(s); } taosMemoryFree(pCfgStr); } char* syncNode2SimpleStr(const SSyncNode* pSyncNode) { int len = 256; char* s = (char*)taosMemoryMalloc(len); SSnapshot snapshot = {.data = NULL, .lastApplyIndex = -1, .lastApplyTerm = 0}; if (pSyncNode->pFsm->FpGetSnapshotInfo != NULL) { pSyncNode->pFsm->FpGetSnapshotInfo(pSyncNode->pFsm, &snapshot); } SyncIndex logLastIndex = pSyncNode->pLogStore->syncLogLastIndex(pSyncNode->pLogStore); SyncIndex logBeginIndex = pSyncNode->pLogStore->syncLogBeginIndex(pSyncNode->pLogStore); snprintf(s, len, "vgId:%d, sync %s, term:%" PRIu64 ", commit:%" PRId64 ", first:%" PRId64 ", last:%" PRId64 ", snapshot:%" PRId64 ", standby:%d, " "replica-num:%d, " "lconfig:%" PRId64 ", changing:%d, restore:%d", pSyncNode->vgId, syncUtilState2String(pSyncNode->state), pSyncNode->pRaftStore->currentTerm, pSyncNode->commitIndex, logBeginIndex, logLastIndex, snapshot.lastApplyIndex, pSyncNode->pRaftCfg->isStandBy, pSyncNode->replicaNum, pSyncNode->pRaftCfg->lastConfigIndex, pSyncNode->changing, pSyncNode->restoreFinish); return s; } bool syncNodeInConfig(SSyncNode* pSyncNode, const SSyncCfg* config) { bool b1 = false; bool b2 = false; for (int i = 0; i < config->replicaNum; ++i) { if (strcmp((config->nodeInfo)[i].nodeFqdn, pSyncNode->myNodeInfo.nodeFqdn) == 0 && (config->nodeInfo)[i].nodePort == pSyncNode->myNodeInfo.nodePort) { b1 = true; break; } } for (int i = 0; i < config->replicaNum; ++i) { SRaftId raftId; raftId.addr = syncUtilAddr2U64((config->nodeInfo)[i].nodeFqdn, (config->nodeInfo)[i].nodePort); raftId.vgId = pSyncNode->vgId; if (syncUtilSameId(&raftId, &(pSyncNode->myRaftId))) { b2 = true; break; } } ASSERT(b1 == b2); return b1; } void syncNodeDoConfigChange(SSyncNode* pSyncNode, SSyncCfg* pNewConfig, SyncIndex lastConfigChangeIndex) { SSyncCfg oldConfig = pSyncNode->pRaftCfg->cfg; pSyncNode->pRaftCfg->cfg = *pNewConfig; pSyncNode->pRaftCfg->lastConfigIndex = lastConfigChangeIndex; bool IamInOld = syncNodeInConfig(pSyncNode, &oldConfig); bool IamInNew = syncNodeInConfig(pSyncNode, pNewConfig); bool isDrop = false; bool isAdd = false; if (IamInOld && !IamInNew) { isDrop = true; } else { isDrop = false; } if (!IamInOld && IamInNew) { isAdd = true; } else { isAdd = false; } // log begin config change do { char eventLog[256]; char* pOldCfgStr = syncCfg2SimpleStr(&oldConfig); char* pNewCfgStr = syncCfg2SimpleStr(pNewConfig); snprintf(eventLog, sizeof(eventLog), "begin do config change, from %s to %s", pOldCfgStr, pNewCfgStr); syncNodeEventLog(pSyncNode, eventLog); taosMemoryFree(pOldCfgStr); taosMemoryFree(pNewCfgStr); } while (0); if (IamInNew) { pSyncNode->pRaftCfg->isStandBy = 0; // change isStandBy to normal } if (isDrop) { pSyncNode->pRaftCfg->isStandBy = 1; // set standby } // add last config index raftCfgAddConfigIndex(pSyncNode->pRaftCfg, lastConfigChangeIndex); if (IamInNew) { //----------------------------------------- int32_t ret = 0; // save snapshot senders int32_t oldReplicaNum = pSyncNode->replicaNum; SRaftId oldReplicasId[TSDB_MAX_REPLICA]; memcpy(oldReplicasId, pSyncNode->replicasId, sizeof(oldReplicasId)); SSyncSnapshotSender* oldSenders[TSDB_MAX_REPLICA]; for (int i = 0; i < TSDB_MAX_REPLICA; ++i) { oldSenders[i] = (pSyncNode->senders)[i]; char* eventLog = snapshotSender2SimpleStr(oldSenders[i], "snapshot sender save old"); syncNodeEventLog(pSyncNode, eventLog); taosMemoryFree(eventLog); } // init internal pSyncNode->myNodeInfo = pSyncNode->pRaftCfg->cfg.nodeInfo[pSyncNode->pRaftCfg->cfg.myIndex]; syncUtilnodeInfo2raftId(&pSyncNode->myNodeInfo, pSyncNode->vgId, &pSyncNode->myRaftId); // init peersNum, peers, peersId pSyncNode->peersNum = pSyncNode->pRaftCfg->cfg.replicaNum - 1; int j = 0; for (int i = 0; i < pSyncNode->pRaftCfg->cfg.replicaNum; ++i) { if (i != pSyncNode->pRaftCfg->cfg.myIndex) { pSyncNode->peersNodeInfo[j] = pSyncNode->pRaftCfg->cfg.nodeInfo[i]; j++; } } for (int i = 0; i < pSyncNode->peersNum; ++i) { syncUtilnodeInfo2raftId(&pSyncNode->peersNodeInfo[i], pSyncNode->vgId, &pSyncNode->peersId[i]); } // init replicaNum, replicasId pSyncNode->replicaNum = pSyncNode->pRaftCfg->cfg.replicaNum; for (int i = 0; i < pSyncNode->pRaftCfg->cfg.replicaNum; ++i) { syncUtilnodeInfo2raftId(&pSyncNode->pRaftCfg->cfg.nodeInfo[i], pSyncNode->vgId, &pSyncNode->replicasId[i]); } syncIndexMgrUpdate(pSyncNode->pNextIndex, pSyncNode); syncIndexMgrUpdate(pSyncNode->pMatchIndex, pSyncNode); voteGrantedUpdate(pSyncNode->pVotesGranted, pSyncNode); votesRespondUpdate(pSyncNode->pVotesRespond, pSyncNode); pSyncNode->quorum = syncUtilQuorum(pSyncNode->pRaftCfg->cfg.replicaNum); // reset snapshot senders // clear new for (int i = 0; i < TSDB_MAX_REPLICA; ++i) { (pSyncNode->senders)[i] = NULL; } // reset new for (int i = 0; i < pSyncNode->replicaNum; ++i) { // reset sender bool reset = false; for (int j = 0; j < TSDB_MAX_REPLICA; ++j) { if (syncUtilSameId(&(pSyncNode->replicasId)[i], &oldReplicasId[j])) { char host[128]; uint16_t port; syncUtilU642Addr((pSyncNode->replicasId)[i].addr, host, sizeof(host), &port); do { char eventLog[256]; snprintf(eventLog, sizeof(eventLog), "snapshot sender reset for: %" PRIu64 ", newIndex:%d, %s:%d, %p", (pSyncNode->replicasId)[i].addr, i, host, port, oldSenders[j]); syncNodeEventLog(pSyncNode, eventLog); } while (0); (pSyncNode->senders)[i] = oldSenders[j]; oldSenders[j] = NULL; reset = true; // reset replicaIndex int32_t oldreplicaIndex = (pSyncNode->senders)[i]->replicaIndex; (pSyncNode->senders)[i]->replicaIndex = i; do { char eventLog[256]; snprintf(eventLog, sizeof(eventLog), "snapshot sender udpate replicaIndex from %d to %d, %s:%d, %p, reset:%d", oldreplicaIndex, i, host, port, (pSyncNode->senders)[i], reset); syncNodeEventLog(pSyncNode, eventLog); } while (0); } } } // create new for (int i = 0; i < TSDB_MAX_REPLICA; ++i) { if ((pSyncNode->senders)[i] == NULL) { (pSyncNode->senders)[i] = snapshotSenderCreate(pSyncNode, i); char* eventLog = snapshotSender2SimpleStr((pSyncNode->senders)[i], "snapshot sender create new"); syncNodeEventLog(pSyncNode, eventLog); taosMemoryFree(eventLog); } } // free old for (int i = 0; i < TSDB_MAX_REPLICA; ++i) { if (oldSenders[i] != NULL) { snapshotSenderDestroy(oldSenders[i]); do { char eventLog[128]; snprintf(eventLog, sizeof(eventLog), "snapshot sender delete old %p replica-index:%d", oldSenders[i], i); syncNodeEventLog(pSyncNode, eventLog); } while (0); oldSenders[i] = NULL; } } // persist cfg raftCfgPersist(pSyncNode->pRaftCfg); char tmpbuf[512]; char* oldStr = syncCfg2SimpleStr(&oldConfig); char* newStr = syncCfg2SimpleStr(pNewConfig); snprintf(tmpbuf, sizeof(tmpbuf), "config change from %d to %d, index:%" PRId64 ", %s --> %s", oldConfig.replicaNum, pNewConfig->replicaNum, lastConfigChangeIndex, oldStr, newStr); taosMemoryFree(oldStr); taosMemoryFree(newStr); // change isStandBy to normal (election timeout) if (pSyncNode->state == TAOS_SYNC_STATE_LEADER) { syncNodeBecomeLeader(pSyncNode, tmpbuf); // Raft 3.6.2 Committing entries from previous terms syncNodeReplicate(pSyncNode); syncNodeAppendNoop(pSyncNode); syncMaybeAdvanceCommitIndex(pSyncNode); } else { syncNodeBecomeFollower(pSyncNode, tmpbuf); } } else { // persist cfg raftCfgPersist(pSyncNode->pRaftCfg); char tmpbuf[512]; char* oldStr = syncCfg2SimpleStr(&oldConfig); char* newStr = syncCfg2SimpleStr(pNewConfig); snprintf(tmpbuf, sizeof(tmpbuf), "do not config change from %d to %d, index:%" PRId64 ", %s --> %s", oldConfig.replicaNum, pNewConfig->replicaNum, lastConfigChangeIndex, oldStr, newStr); taosMemoryFree(oldStr); taosMemoryFree(newStr); syncNodeEventLog(pSyncNode, tmpbuf); } _END: // log end config change do { char eventLog[256]; char* pOldCfgStr = syncCfg2SimpleStr(&oldConfig); char* pNewCfgStr = syncCfg2SimpleStr(pNewConfig); snprintf(eventLog, sizeof(eventLog), "end do config change, from %s to %s", pOldCfgStr, pNewCfgStr); syncNodeEventLog(pSyncNode, eventLog); taosMemoryFree(pOldCfgStr); taosMemoryFree(pNewCfgStr); } while (0); return; } SSyncNode* syncNodeAcquire(int64_t rid) { SSyncNode* pNode = taosAcquireRef(tsNodeRefId, rid); if (pNode == NULL) { sTrace("failed to acquire node from refId:%" PRId64, rid); } return pNode; } void syncNodeRelease(SSyncNode* pNode) { taosReleaseRef(tsNodeRefId, pNode->rid); } // raft state change -------------- void syncNodeUpdateTerm(SSyncNode* pSyncNode, SyncTerm term) { if (term > pSyncNode->pRaftStore->currentTerm) { raftStoreSetTerm(pSyncNode->pRaftStore, term); char tmpBuf[64]; snprintf(tmpBuf, sizeof(tmpBuf), "update term to %" PRIu64, term); syncNodeBecomeFollower(pSyncNode, tmpBuf); raftStoreClearVote(pSyncNode->pRaftStore); } } void syncNodeBecomeFollower(SSyncNode* pSyncNode, const char* debugStr) { // maybe clear leader cache if (pSyncNode->state == TAOS_SYNC_STATE_LEADER) { pSyncNode->leaderCache = EMPTY_RAFT_ID; } // state change pSyncNode->state = TAOS_SYNC_STATE_FOLLOWER; syncNodeStopHeartbeatTimer(pSyncNode); // reset elect timer syncNodeResetElectTimer(pSyncNode); // trace log do { int32_t debugStrLen = strlen(debugStr); if (debugStrLen < 256) { char eventLog[256 + 64]; snprintf(eventLog, sizeof(eventLog), "become follower %s", debugStr); syncNodeEventLog(pSyncNode, eventLog); } else { char* eventLog = taosMemoryMalloc(debugStrLen + 64); snprintf(eventLog, debugStrLen, "become follower %s", debugStr); syncNodeEventLog(pSyncNode, eventLog); taosMemoryFree(eventLog); } } while (0); } // TLA+ Spec // \* Candidate i transitions to leader. // BecomeLeader(i) == // /\ state[i] = Candidate // /\ votesGranted[i] \in Quorum // /\ state' = [state EXCEPT ![i] = Leader] // /\ nextIndex' = [nextIndex EXCEPT ![i] = // [j \in Server |-> Len(log[i]) + 1]] // /\ matchIndex' = [matchIndex EXCEPT ![i] = // [j \in Server |-> 0]] // /\ elections' = elections \cup // {[eterm |-> currentTerm[i], // eleader |-> i, // elog |-> log[i], // evotes |-> votesGranted[i], // evoterLog |-> voterLog[i]]} // /\ UNCHANGED <> // void syncNodeBecomeLeader(SSyncNode* pSyncNode, const char* debugStr) { // reset restoreFinish pSyncNode->restoreFinish = false; // state change pSyncNode->state = TAOS_SYNC_STATE_LEADER; // set leader cache pSyncNode->leaderCache = pSyncNode->myRaftId; for (int i = 0; i < pSyncNode->pNextIndex->replicaNum; ++i) { // maybe overwrite myself, no harm // just do it! // pSyncNode->pNextIndex->index[i] = pSyncNode->pLogStore->getLastIndex(pSyncNode->pLogStore) + 1; // maybe wal is deleted SyncIndex lastIndex; SyncTerm lastTerm; int32_t code = syncNodeGetLastIndexTerm(pSyncNode, &lastIndex, &lastTerm); ASSERT(code == 0); pSyncNode->pNextIndex->index[i] = lastIndex + 1; } for (int i = 0; i < pSyncNode->pMatchIndex->replicaNum; ++i) { // maybe overwrite myself, no harm // just do it! pSyncNode->pMatchIndex->index[i] = SYNC_INDEX_INVALID; } // update sender private term SSyncSnapshotSender* pMySender = syncNodeGetSnapshotSender(pSyncNode, &(pSyncNode->myRaftId)); if (pMySender != NULL) { for (int i = 0; i < pSyncNode->pMatchIndex->replicaNum; ++i) { if ((pSyncNode->senders)[i]->privateTerm > pMySender->privateTerm) { pMySender->privateTerm = (pSyncNode->senders)[i]->privateTerm; } } (pMySender->privateTerm) += 100; } // stop elect timer syncNodeStopElectTimer(pSyncNode); // start heartbeat timer syncNodeStartHeartbeatTimer(pSyncNode); // trace log do { int32_t debugStrLen = strlen(debugStr); if (debugStrLen < 256) { char eventLog[256 + 64]; snprintf(eventLog, sizeof(eventLog), "become leader %s", debugStr); syncNodeEventLog(pSyncNode, eventLog); } else { char* eventLog = taosMemoryMalloc(debugStrLen + 64); snprintf(eventLog, debugStrLen, "become leader %s", debugStr); syncNodeEventLog(pSyncNode, eventLog); taosMemoryFree(eventLog); } } while (0); } void syncNodeCandidate2Leader(SSyncNode* pSyncNode) { ASSERT(pSyncNode->state == TAOS_SYNC_STATE_CANDIDATE); ASSERT(voteGrantedMajority(pSyncNode->pVotesGranted)); syncNodeBecomeLeader(pSyncNode, "candidate to leader"); syncNodeLog2("==state change syncNodeCandidate2Leader==", pSyncNode); // Raft 3.6.2 Committing entries from previous terms syncNodeReplicate(pSyncNode); syncNodeAppendNoop(pSyncNode); syncMaybeAdvanceCommitIndex(pSyncNode); } void syncNodeFollower2Candidate(SSyncNode* pSyncNode) { ASSERT(pSyncNode->state == TAOS_SYNC_STATE_FOLLOWER); pSyncNode->state = TAOS_SYNC_STATE_CANDIDATE; syncNodeLog2("==state change syncNodeFollower2Candidate==", pSyncNode); } void syncNodeLeader2Follower(SSyncNode* pSyncNode) { ASSERT(pSyncNode->state == TAOS_SYNC_STATE_LEADER); syncNodeBecomeFollower(pSyncNode, "leader to follower"); syncNodeLog2("==state change syncNodeLeader2Follower==", pSyncNode); } void syncNodeCandidate2Follower(SSyncNode* pSyncNode) { ASSERT(pSyncNode->state == TAOS_SYNC_STATE_CANDIDATE); syncNodeBecomeFollower(pSyncNode, "candidate to follower"); syncNodeLog2("==state change syncNodeCandidate2Follower==", pSyncNode); } // raft vote -------------- // just called by syncNodeVoteForSelf // need assert void syncNodeVoteForTerm(SSyncNode* pSyncNode, SyncTerm term, SRaftId* pRaftId) { ASSERT(term == pSyncNode->pRaftStore->currentTerm); ASSERT(!raftStoreHasVoted(pSyncNode->pRaftStore)); raftStoreVote(pSyncNode->pRaftStore, pRaftId); } // simulate get vote from outside void syncNodeVoteForSelf(SSyncNode* pSyncNode) { syncNodeVoteForTerm(pSyncNode, pSyncNode->pRaftStore->currentTerm, &(pSyncNode->myRaftId)); SyncRequestVoteReply* pMsg = syncRequestVoteReplyBuild(pSyncNode->vgId); pMsg->srcId = pSyncNode->myRaftId; pMsg->destId = pSyncNode->myRaftId; pMsg->term = pSyncNode->pRaftStore->currentTerm; pMsg->voteGranted = true; voteGrantedVote(pSyncNode->pVotesGranted, pMsg); votesRespondAdd(pSyncNode->pVotesRespond, pMsg); syncRequestVoteReplyDestroy(pMsg); } // snapshot -------------- // return if has a snapshot bool syncNodeHasSnapshot(SSyncNode* pSyncNode) { bool ret = false; SSnapshot snapshot = {.data = NULL, .lastApplyIndex = -1, .lastApplyTerm = 0, .lastConfigIndex = -1}; if (pSyncNode->pFsm->FpGetSnapshotInfo != NULL) { pSyncNode->pFsm->FpGetSnapshotInfo(pSyncNode->pFsm, &snapshot); if (snapshot.lastApplyIndex >= SYNC_INDEX_BEGIN) { ret = true; } } return ret; } // return max(logLastIndex, snapshotLastIndex) // if no snapshot and log, return -1 SyncIndex syncNodeGetLastIndex(SSyncNode* pSyncNode) { SSnapshot snapshot = {.data = NULL, .lastApplyIndex = -1, .lastApplyTerm = 0, .lastConfigIndex = -1}; if (pSyncNode->pFsm->FpGetSnapshotInfo != NULL) { pSyncNode->pFsm->FpGetSnapshotInfo(pSyncNode->pFsm, &snapshot); } SyncIndex logLastIndex = pSyncNode->pLogStore->syncLogLastIndex(pSyncNode->pLogStore); SyncIndex lastIndex = logLastIndex > snapshot.lastApplyIndex ? logLastIndex : snapshot.lastApplyIndex; return lastIndex; } // return the last term of snapshot and log // if error, return SYNC_TERM_INVALID (by syncLogLastTerm) SyncTerm syncNodeGetLastTerm(SSyncNode* pSyncNode) { SyncTerm lastTerm = 0; if (syncNodeHasSnapshot(pSyncNode)) { // has snapshot SSnapshot snapshot = {.data = NULL, .lastApplyIndex = -1, .lastApplyTerm = 0, .lastConfigIndex = -1}; if (pSyncNode->pFsm->FpGetSnapshotInfo != NULL) { pSyncNode->pFsm->FpGetSnapshotInfo(pSyncNode->pFsm, &snapshot); } SyncIndex logLastIndex = pSyncNode->pLogStore->syncLogLastIndex(pSyncNode->pLogStore); if (logLastIndex > snapshot.lastApplyIndex) { lastTerm = pSyncNode->pLogStore->syncLogLastTerm(pSyncNode->pLogStore); } else { lastTerm = snapshot.lastApplyTerm; } } else { // no snapshot lastTerm = pSyncNode->pLogStore->syncLogLastTerm(pSyncNode->pLogStore); } return lastTerm; } // get last index and term along with snapshot int32_t syncNodeGetLastIndexTerm(SSyncNode* pSyncNode, SyncIndex* pLastIndex, SyncTerm* pLastTerm) { *pLastIndex = syncNodeGetLastIndex(pSyncNode); *pLastTerm = syncNodeGetLastTerm(pSyncNode); return 0; } // return append-entries first try index SyncIndex syncNodeSyncStartIndex(SSyncNode* pSyncNode) { SyncIndex syncStartIndex = syncNodeGetLastIndex(pSyncNode) + 1; return syncStartIndex; } // if index > 0, return index - 1 // else, return -1 SyncIndex syncNodeGetPreIndex(SSyncNode* pSyncNode, SyncIndex index) { SyncIndex preIndex = index - 1; if (preIndex < SYNC_INDEX_INVALID) { preIndex = SYNC_INDEX_INVALID; } return preIndex; } // if index < 0, return SYNC_TERM_INVALID // if index == 0, return 0 // if index > 0, return preTerm // if error, return SYNC_TERM_INVALID SyncTerm syncNodeGetPreTerm(SSyncNode* pSyncNode, SyncIndex index) { if (index < SYNC_INDEX_BEGIN) { return SYNC_TERM_INVALID; } if (index == SYNC_INDEX_BEGIN) { return 0; } SyncTerm preTerm = 0; SyncIndex preIndex = index - 1; SSyncRaftEntry* pPreEntry = NULL; int32_t code = pSyncNode->pLogStore->syncLogGetEntry(pSyncNode->pLogStore, preIndex, &pPreEntry); if (code == 0) { ASSERT(pPreEntry != NULL); preTerm = pPreEntry->term; taosMemoryFree(pPreEntry); return preTerm; } else { SSnapshot snapshot = {.data = NULL, .lastApplyIndex = -1, .lastApplyTerm = 0, .lastConfigIndex = -1}; if (pSyncNode->pFsm->FpGetSnapshotInfo != NULL) { pSyncNode->pFsm->FpGetSnapshotInfo(pSyncNode->pFsm, &snapshot); if (snapshot.lastApplyIndex == preIndex) { return snapshot.lastApplyTerm; } } } do { char logBuf[128]; snprintf(logBuf, sizeof(logBuf), "sync node get pre term error, index:%" PRId64, index); syncNodeErrorLog(pSyncNode, logBuf); } while (0); return SYNC_TERM_INVALID; } // get pre index and term of "index" int32_t syncNodeGetPreIndexTerm(SSyncNode* pSyncNode, SyncIndex index, SyncIndex* pPreIndex, SyncTerm* pPreTerm) { *pPreIndex = syncNodeGetPreIndex(pSyncNode, index); *pPreTerm = syncNodeGetPreTerm(pSyncNode, index); return 0; } // for debug -------------- void syncNodePrint(SSyncNode* pObj) { char* serialized = syncNode2Str(pObj); printf("syncNodePrint | len:%" PRIu64 " | %s \n", strlen(serialized), serialized); fflush(NULL); taosMemoryFree(serialized); } void syncNodePrint2(char* s, SSyncNode* pObj) { char* serialized = syncNode2Str(pObj); printf("syncNodePrint2 | len:%" PRIu64 " | %s | %s \n", strlen(serialized), s, serialized); fflush(NULL); taosMemoryFree(serialized); } void syncNodeLog(SSyncNode* pObj) { char* serialized = syncNode2Str(pObj); sTraceLong("syncNodeLog | len:%" PRIu64 " | %s", strlen(serialized), serialized); taosMemoryFree(serialized); } void syncNodeLog2(char* s, SSyncNode* pObj) { if (gRaftDetailLog) { char* serialized = syncNode2Str(pObj); sTraceLong("syncNodeLog2 | len:%" PRIu64 " | %s | %s", strlen(serialized), s, serialized); taosMemoryFree(serialized); } } void syncNodeLog3(char* s, SSyncNode* pObj) { char* serialized = syncNode2Str(pObj); sTraceLong("syncNodeLog3 | len:%" PRIu64 " | %s | %s", strlen(serialized), s, serialized); taosMemoryFree(serialized); } // ------ local funciton --------- // enqueue message ---- static void syncNodeEqPingTimer(void* param, void* tmrId) { SSyncNode* pSyncNode = (SSyncNode*)param; if (atomic_load_64(&pSyncNode->pingTimerLogicClockUser) <= atomic_load_64(&pSyncNode->pingTimerLogicClock)) { SyncTimeout* pSyncMsg = syncTimeoutBuild2(SYNC_TIMEOUT_PING, atomic_load_64(&pSyncNode->pingTimerLogicClock), pSyncNode->pingTimerMS, pSyncNode->vgId, pSyncNode); SRpcMsg rpcMsg; syncTimeout2RpcMsg(pSyncMsg, &rpcMsg); syncRpcMsgLog2((char*)"==syncNodeEqPingTimer==", &rpcMsg); if (pSyncNode->FpEqMsg != NULL) { int32_t code = pSyncNode->FpEqMsg(pSyncNode->msgcb, &rpcMsg); if (code != 0) { sError("vgId:%d, sync enqueue ping msg error, code:%d", pSyncNode->vgId, code); rpcFreeCont(rpcMsg.pCont); syncTimeoutDestroy(pSyncMsg); return; } } else { sTrace("syncNodeEqPingTimer pSyncNode->FpEqMsg is NULL"); } syncTimeoutDestroy(pSyncMsg); if (syncEnvIsStart()) { taosTmrReset(syncNodeEqPingTimer, pSyncNode->pingTimerMS, pSyncNode, gSyncEnv->pTimerManager, &pSyncNode->pPingTimer); } else { sError("sync env is stop, syncNodeEqPingTimer"); } } else { sTrace("==syncNodeEqPingTimer== pingTimerLogicClock:%" PRIu64 ", pingTimerLogicClockUser:%" PRIu64, pSyncNode->pingTimerLogicClock, pSyncNode->pingTimerLogicClockUser); } } static void syncNodeEqElectTimer(void* param, void* tmrId) { SSyncNode* pSyncNode = (SSyncNode*)param; if (atomic_load_64(&pSyncNode->electTimerLogicClockUser) <= atomic_load_64(&pSyncNode->electTimerLogicClock)) { SyncTimeout* pSyncMsg = syncTimeoutBuild2(SYNC_TIMEOUT_ELECTION, atomic_load_64(&pSyncNode->electTimerLogicClock), pSyncNode->electTimerMS, pSyncNode->vgId, pSyncNode); SRpcMsg rpcMsg; syncTimeout2RpcMsg(pSyncMsg, &rpcMsg); syncRpcMsgLog2((char*)"==syncNodeEqElectTimer==", &rpcMsg); if (pSyncNode->FpEqMsg != NULL) { int32_t code = pSyncNode->FpEqMsg(pSyncNode->msgcb, &rpcMsg); if (code != 0) { sError("vgId:%d, sync enqueue elect msg error, code:%d", pSyncNode->vgId, code); rpcFreeCont(rpcMsg.pCont); syncTimeoutDestroy(pSyncMsg); return; } } else { sTrace("syncNodeEqElectTimer FpEqMsg is NULL"); } syncTimeoutDestroy(pSyncMsg); // reset timer ms if (syncEnvIsStart()) { pSyncNode->electTimerMS = syncUtilElectRandomMS(pSyncNode->electBaseLine, 2 * pSyncNode->electBaseLine); taosTmrReset(syncNodeEqElectTimer, pSyncNode->electTimerMS, pSyncNode, gSyncEnv->pTimerManager, &pSyncNode->pElectTimer); } else { sError("sync env is stop, syncNodeEqElectTimer"); } } else { sTrace("==syncNodeEqElectTimer== electTimerLogicClock:%" PRIu64 ", electTimerLogicClockUser:%" PRIu64, pSyncNode->electTimerLogicClock, pSyncNode->electTimerLogicClockUser); } } static void syncNodeEqHeartbeatTimer(void* param, void* tmrId) { SSyncNode* pSyncNode = (SSyncNode*)param; if (pSyncNode->replicaNum > 1) { if (atomic_load_64(&pSyncNode->heartbeatTimerLogicClockUser) <= atomic_load_64(&pSyncNode->heartbeatTimerLogicClock)) { SyncTimeout* pSyncMsg = syncTimeoutBuild2(SYNC_TIMEOUT_HEARTBEAT, atomic_load_64(&pSyncNode->heartbeatTimerLogicClock), pSyncNode->heartbeatTimerMS, pSyncNode->vgId, pSyncNode); SRpcMsg rpcMsg; syncTimeout2RpcMsg(pSyncMsg, &rpcMsg); syncRpcMsgLog2((char*)"==syncNodeEqHeartbeatTimer==", &rpcMsg); if (pSyncNode->FpEqMsg != NULL) { int32_t code = pSyncNode->FpEqMsg(pSyncNode->msgcb, &rpcMsg); if (code != 0) { sError("vgId:%d, sync enqueue timer msg error, code:%d", pSyncNode->vgId, code); rpcFreeCont(rpcMsg.pCont); syncTimeoutDestroy(pSyncMsg); return; } } else { sError("syncNodeEqHeartbeatTimer FpEqMsg is NULL"); } syncTimeoutDestroy(pSyncMsg); if (syncEnvIsStart()) { taosTmrReset(syncNodeEqHeartbeatTimer, pSyncNode->heartbeatTimerMS, pSyncNode, gSyncEnv->pTimerManager, &pSyncNode->pHeartbeatTimer); } else { sError("sync env is stop, syncNodeEqHeartbeatTimer"); } } else { sTrace("==syncNodeEqHeartbeatTimer== heartbeatTimerLogicClock:%" PRIu64 ", heartbeatTimerLogicClockUser:%" PRIu64 "", pSyncNode->heartbeatTimerLogicClock, pSyncNode->heartbeatTimerLogicClockUser); } } } static int32_t syncNodeEqNoop(SSyncNode* ths) { int32_t ret = 0; ASSERT(ths->state == TAOS_SYNC_STATE_LEADER); SyncIndex index = ths->pLogStore->syncLogWriteIndex(ths->pLogStore); SyncTerm term = ths->pRaftStore->currentTerm; SSyncRaftEntry* pEntry = syncEntryBuildNoop(term, index, ths->vgId); ASSERT(pEntry != NULL); uint32_t entryLen; char* serialized = syncEntrySerialize(pEntry, &entryLen); SyncClientRequest* pSyncMsg = syncClientRequestBuild(entryLen); ASSERT(pSyncMsg->dataLen == entryLen); memcpy(pSyncMsg->data, serialized, entryLen); SRpcMsg rpcMsg = {0}; syncClientRequest2RpcMsg(pSyncMsg, &rpcMsg); if (ths->FpEqMsg != NULL) { ths->FpEqMsg(ths->msgcb, &rpcMsg); } else { sTrace("syncNodeEqNoop pSyncNode->FpEqMsg is NULL"); } taosMemoryFree(serialized); syncClientRequestDestroy(pSyncMsg); return ret; } static int32_t syncNodeAppendNoop(SSyncNode* ths) { int32_t ret = 0; SyncIndex index = ths->pLogStore->syncLogWriteIndex(ths->pLogStore); SyncTerm term = ths->pRaftStore->currentTerm; SSyncRaftEntry* pEntry = syncEntryBuildNoop(term, index, ths->vgId); ASSERT(pEntry != NULL); if (ths->state == TAOS_SYNC_STATE_LEADER) { int32_t code = ths->pLogStore->syncLogAppendEntry(ths->pLogStore, pEntry); ASSERT(code == 0); syncNodeReplicate(ths); } syncEntryDestory(pEntry); return ret; } // on message ---- int32_t syncNodeOnPingCb(SSyncNode* ths, SyncPing* pMsg) { // log state char logBuf[1024] = {0}; snprintf(logBuf, sizeof(logBuf), "==syncNodeOnPingCb== vgId:%d, state: %d, %s, term:%" PRIu64 " electTimerLogicClock:%" PRIu64 ", " "electTimerLogicClockUser:%" PRIu64 ", electTimerMS:%d", ths->vgId, ths->state, syncUtilState2String(ths->state), ths->pRaftStore->currentTerm, ths->electTimerLogicClock, ths->electTimerLogicClockUser, ths->electTimerMS); int32_t ret = 0; syncPingLog2(logBuf, pMsg); SyncPingReply* pMsgReply = syncPingReplyBuild3(&ths->myRaftId, &pMsg->srcId, ths->vgId); SRpcMsg rpcMsg; syncPingReply2RpcMsg(pMsgReply, &rpcMsg); /* // htonl SMsgHead* pHead = rpcMsg.pCont; pHead->contLen = htonl(pHead->contLen); pHead->vgId = htonl(pHead->vgId); */ syncNodeSendMsgById(&pMsgReply->destId, ths, &rpcMsg); return ret; } int32_t syncNodeOnPingReplyCb(SSyncNode* ths, SyncPingReply* pMsg) { int32_t ret = 0; syncPingReplyLog2("==syncNodeOnPingReplyCb==", pMsg); return ret; } // TLA+ Spec // ClientRequest(i, v) == // /\ state[i] = Leader // /\ LET entry == [term |-> currentTerm[i], // value |-> v] // newLog == Append(log[i], entry) // IN log' = [log EXCEPT ![i] = newLog] // /\ UNCHANGED <> // int32_t syncNodeOnClientRequestCb(SSyncNode* ths, SyncClientRequest* pMsg, SyncIndex* pRetIndex) { int32_t ret = 0; int32_t code = 0; syncClientRequestLog2("==syncNodeOnClientRequestCb==", pMsg); SyncIndex index = ths->pLogStore->syncLogWriteIndex(ths->pLogStore); SyncTerm term = ths->pRaftStore->currentTerm; SSyncRaftEntry* pEntry = syncEntryBuild2((SyncClientRequest*)pMsg, term, index); ASSERT(pEntry != NULL); if (ths->state == TAOS_SYNC_STATE_LEADER) { // append entry code = ths->pLogStore->syncLogAppendEntry(ths->pLogStore, pEntry); if (code != 0) { // del resp mgr, call FpCommitCb ASSERT(0); return -1; } // if mulit replica, start replicate right now if (ths->replicaNum > 1) { syncNodeReplicate(ths); } // pre commit SRpcMsg rpcMsg; syncEntry2OriginalRpc(pEntry, &rpcMsg); if (ths->pFsm != NULL) { if (ths->pFsm->FpPreCommitCb != NULL && syncUtilUserPreCommit(pEntry->originalRpcType)) { SFsmCbMeta cbMeta = {0}; cbMeta.index = pEntry->index; cbMeta.lastConfigIndex = syncNodeGetSnapshotConfigIndex(ths, cbMeta.index); cbMeta.isWeak = pEntry->isWeak; cbMeta.code = 0; cbMeta.state = ths->state; cbMeta.seqNum = pEntry->seqNum; ths->pFsm->FpPreCommitCb(ths->pFsm, &rpcMsg, cbMeta); } } rpcFreeCont(rpcMsg.pCont); // if only myself, maybe commit right now if (ths->replicaNum == 1) { syncMaybeAdvanceCommitIndex(ths); } } else { // pre commit SRpcMsg rpcMsg; syncEntry2OriginalRpc(pEntry, &rpcMsg); if (ths->pFsm != NULL) { if (ths->pFsm->FpPreCommitCb != NULL && syncUtilUserPreCommit(pEntry->originalRpcType)) { SFsmCbMeta cbMeta = {0}; cbMeta.index = pEntry->index; cbMeta.lastConfigIndex = syncNodeGetSnapshotConfigIndex(ths, cbMeta.index); cbMeta.isWeak = pEntry->isWeak; cbMeta.code = 1; cbMeta.state = ths->state; cbMeta.seqNum = pEntry->seqNum; ths->pFsm->FpPreCommitCb(ths->pFsm, &rpcMsg, cbMeta); } } rpcFreeCont(rpcMsg.pCont); } if (pRetIndex != NULL) { if (ret == 0 && pEntry != NULL) { *pRetIndex = pEntry->index; } else { *pRetIndex = SYNC_INDEX_INVALID; } } syncEntryDestory(pEntry); return ret; } int32_t syncNodeOnClientRequestBatchCb(SSyncNode* ths, SyncClientRequestBatch* pMsg) { int32_t code = 0; if (ths->state != TAOS_SYNC_STATE_LEADER) { // call FpCommitCb, delete resp mgr return -1; } SyncIndex index = ths->pLogStore->syncLogWriteIndex(ths->pLogStore); SyncTerm term = ths->pRaftStore->currentTerm; int32_t raftMetaArrayLen = sizeof(SRaftMeta) * pMsg->dataCount; int32_t rpcArrayLen = sizeof(SRpcMsg) * pMsg->dataCount; SRaftMeta* raftMetaArr = (SRaftMeta*)(pMsg->data); SRpcMsg* msgArr = (SRpcMsg*)((char*)(pMsg->data) + raftMetaArrayLen); for (int32_t i = 0; i < pMsg->dataCount; ++i) { SSyncRaftEntry* pEntry = syncEntryBuild(msgArr[i].contLen); ASSERT(pEntry != NULL); pEntry->originalRpcType = msgArr[i].msgType; pEntry->seqNum = raftMetaArr[i].seqNum; pEntry->isWeak = raftMetaArr[i].isWeak; pEntry->term = term; pEntry->index = index; memcpy(pEntry->data, msgArr[i].pCont, msgArr[i].contLen); ASSERT(msgArr[i].contLen == pEntry->dataLen); code = ths->pLogStore->syncLogAppendEntry(ths->pLogStore, pEntry); if (code != 0) { // del resp mgr, call FpCommitCb ASSERT(0); return -1; } // update rpc msg conn apply.index msgArr[i].info.conn.applyIndex = pEntry->index; } // fsync once SSyncLogStoreData* pData = ths->pLogStore->data; SWal* pWal = pData->pWal; walFsync(pWal, true); if (ths->replicaNum > 1) { // if multi replica, start replicate right now syncNodeReplicate(ths); } else if (ths->replicaNum == 1) { // one replica syncMaybeAdvanceCommitIndex(ths); } return 0; } static void syncFreeNode(void* param) { SSyncNode* pNode = param; // inner object already free // syncNodePrint2((char*)"==syncFreeNode==", pNode); taosMemoryFree(pNode); } const char* syncStr(ESyncState state) { switch (state) { case TAOS_SYNC_STATE_FOLLOWER: return "follower"; case TAOS_SYNC_STATE_CANDIDATE: return "candidate"; case TAOS_SYNC_STATE_LEADER: return "leader"; default: return "error"; } } static int32_t syncDoLeaderTransfer(SSyncNode* ths, SRpcMsg* pRpcMsg, SSyncRaftEntry* pEntry) { SyncLeaderTransfer* pSyncLeaderTransfer = syncLeaderTransferFromRpcMsg2(pRpcMsg); syncNodeEventLog(ths, "do leader transfer"); bool sameId = syncUtilSameId(&(pSyncLeaderTransfer->newLeaderId), &(ths->myRaftId)); bool sameNodeInfo = strcmp(pSyncLeaderTransfer->newNodeInfo.nodeFqdn, ths->myNodeInfo.nodeFqdn) == 0 && pSyncLeaderTransfer->newNodeInfo.nodePort == ths->myNodeInfo.nodePort; bool same = sameId || sameNodeInfo; if (same) { // reset elect timer now! int32_t electMS = 1; int32_t ret = syncNodeRestartElectTimer(ths, electMS); ASSERT(ret == 0); char eventLog[256]; snprintf(eventLog, sizeof(eventLog), "maybe leader transfer to %s:%d %" PRIu64, pSyncLeaderTransfer->newNodeInfo.nodeFqdn, pSyncLeaderTransfer->newNodeInfo.nodePort, pSyncLeaderTransfer->newLeaderId.addr); syncNodeEventLog(ths, eventLog); } if (ths->pFsm->FpLeaderTransferCb != NULL) { SFsmCbMeta cbMeta = {0}; cbMeta.code = 0; cbMeta.currentTerm = ths->pRaftStore->currentTerm; cbMeta.flag = 0; cbMeta.index = pEntry->index; cbMeta.lastConfigIndex = syncNodeGetSnapshotConfigIndex(ths, cbMeta.index); cbMeta.isWeak = pEntry->isWeak; cbMeta.seqNum = pEntry->seqNum; cbMeta.state = ths->state; cbMeta.term = pEntry->term; ths->pFsm->FpLeaderTransferCb(ths->pFsm, pRpcMsg, cbMeta); } syncLeaderTransferDestroy(pSyncLeaderTransfer); return 0; } int32_t syncNodeUpdateNewConfigIndex(SSyncNode* ths, SSyncCfg* pNewCfg) { for (int i = 0; i < pNewCfg->replicaNum; ++i) { SRaftId raftId; raftId.addr = syncUtilAddr2U64((pNewCfg->nodeInfo)[i].nodeFqdn, (pNewCfg->nodeInfo)[i].nodePort); raftId.vgId = ths->vgId; if (syncUtilSameId(&(ths->myRaftId), &raftId)) { pNewCfg->myIndex = i; return 0; } } return -1; } static int32_t syncNodeConfigChangeFinish(SSyncNode* ths, SRpcMsg* pRpcMsg, SSyncRaftEntry* pEntry) { SyncReconfigFinish* pFinish = syncReconfigFinishFromRpcMsg2(pRpcMsg); ASSERT(pFinish); if (ths->pFsm->FpReConfigCb != NULL) { SReConfigCbMeta cbMeta = {0}; cbMeta.code = 0; cbMeta.index = pEntry->index; cbMeta.term = pEntry->term; cbMeta.seqNum = pEntry->seqNum; cbMeta.lastConfigIndex = syncNodeGetSnapshotConfigIndex(ths, pEntry->index); cbMeta.state = ths->state; cbMeta.currentTerm = ths->pRaftStore->currentTerm; cbMeta.isWeak = pEntry->isWeak; cbMeta.flag = 0; cbMeta.oldCfg = pFinish->oldCfg; cbMeta.newCfg = pFinish->newCfg; cbMeta.newCfgIndex = pFinish->newCfgIndex; cbMeta.newCfgTerm = pFinish->newCfgTerm; cbMeta.newCfgSeqNum = pFinish->newCfgSeqNum; ths->pFsm->FpReConfigCb(ths->pFsm, pRpcMsg, cbMeta); } // clear changing ths->changing = false; char tmpbuf[512]; char* oldStr = syncCfg2SimpleStr(&(pFinish->oldCfg)); char* newStr = syncCfg2SimpleStr(&(pFinish->newCfg)); snprintf(tmpbuf, sizeof(tmpbuf), "config change finish from %d to %d, index:%" PRId64 ", %s --> %s", pFinish->oldCfg.replicaNum, pFinish->newCfg.replicaNum, pFinish->newCfgIndex, oldStr, newStr); taosMemoryFree(oldStr); taosMemoryFree(newStr); syncNodeEventLog(ths, tmpbuf); syncReconfigFinishDestroy(pFinish); return 0; } static int32_t syncNodeConfigChange(SSyncNode* ths, SRpcMsg* pRpcMsg, SSyncRaftEntry* pEntry, SyncReconfigFinish* pFinish) { // set changing ths->changing = true; // old config SSyncCfg oldSyncCfg = ths->pRaftCfg->cfg; // new config SSyncCfg newSyncCfg; int32_t ret = syncCfgFromStr(pRpcMsg->pCont, &newSyncCfg); ASSERT(ret == 0); // update new config myIndex syncNodeUpdateNewConfigIndex(ths, &newSyncCfg); // do config change syncNodeDoConfigChange(ths, &newSyncCfg, pEntry->index); // set pFinish pFinish->oldCfg = oldSyncCfg; pFinish->newCfg = newSyncCfg; pFinish->newCfgIndex = pEntry->index; pFinish->newCfgTerm = pEntry->term; pFinish->newCfgSeqNum = pEntry->seqNum; return 0; } static int32_t syncNodeProposeConfigChangeFinish(SSyncNode* ths, SyncReconfigFinish* pFinish) { SRpcMsg rpcMsg; syncReconfigFinish2RpcMsg(pFinish, &rpcMsg); int32_t code = syncNodePropose(ths, &rpcMsg, false); if (code != 0) { sError("syncNodeProposeConfigChangeFinish error"); ths->changing = false; } return 0; } bool syncNodeIsOptimizedOneReplica(SSyncNode* ths, SRpcMsg* pMsg) { return (ths->replicaNum == 1 && syncUtilUserCommit(pMsg->msgType) && ths->vgId != 1); } int32_t syncNodeCommit(SSyncNode* ths, SyncIndex beginIndex, SyncIndex endIndex, uint64_t flag) { int32_t code = 0; ESyncState state = flag; char eventLog[128]; snprintf(eventLog, sizeof(eventLog), "commit by wal from index:%" PRId64 " to index:%" PRId64, beginIndex, endIndex); syncNodeEventLog(ths, eventLog); // execute fsm if (ths->pFsm != NULL) { for (SyncIndex i = beginIndex; i <= endIndex; ++i) { if (i != SYNC_INDEX_INVALID) { SSyncRaftEntry* pEntry; code = ths->pLogStore->syncLogGetEntry(ths->pLogStore, i, &pEntry); ASSERT(code == 0); ASSERT(pEntry != NULL); SRpcMsg rpcMsg; syncEntry2OriginalRpc(pEntry, &rpcMsg); // user commit if ((ths->pFsm->FpCommitCb != NULL) && syncUtilUserCommit(pEntry->originalRpcType)) { bool internalExecute = true; if ((ths->replicaNum == 1) && ths->restoreFinish && (ths->vgId != 1)) { internalExecute = false; } do { char logBuf[128]; snprintf(logBuf, sizeof(logBuf), "index:%" PRId64 ", internalExecute:%d", i, internalExecute); syncNodeEventLog(ths, logBuf); } while (0); // execute fsm in apply thread, or execute outside syncPropose if (internalExecute) { SFsmCbMeta cbMeta = {0}; cbMeta.index = pEntry->index; cbMeta.lastConfigIndex = syncNodeGetSnapshotConfigIndex(ths, cbMeta.index); cbMeta.isWeak = pEntry->isWeak; cbMeta.code = 0; cbMeta.state = ths->state; cbMeta.seqNum = pEntry->seqNum; cbMeta.term = pEntry->term; cbMeta.currentTerm = ths->pRaftStore->currentTerm; cbMeta.flag = flag; ths->pFsm->FpCommitCb(ths->pFsm, &rpcMsg, cbMeta); } } // config change if (pEntry->originalRpcType == TDMT_SYNC_CONFIG_CHANGE) { SyncReconfigFinish* pFinish = syncReconfigFinishBuild(ths->vgId); ASSERT(pFinish != NULL); code = syncNodeConfigChange(ths, &rpcMsg, pEntry, pFinish); ASSERT(code == 0); if (ths->state == TAOS_SYNC_STATE_LEADER) { syncNodeProposeConfigChangeFinish(ths, pFinish); } syncReconfigFinishDestroy(pFinish); } // config change finish if (pEntry->originalRpcType == TDMT_SYNC_CONFIG_CHANGE_FINISH) { code = syncNodeConfigChangeFinish(ths, &rpcMsg, pEntry); ASSERT(code == 0); } // leader transfer if (pEntry->originalRpcType == TDMT_SYNC_LEADER_TRANSFER) { code = syncDoLeaderTransfer(ths, &rpcMsg, pEntry); ASSERT(code == 0); } // restore finish // if only snapshot, a noop entry will be append, so syncLogLastIndex is always ok if (pEntry->index == ths->pLogStore->syncLogLastIndex(ths->pLogStore)) { if (ths->restoreFinish == false) { if (ths->pFsm->FpRestoreFinishCb != NULL) { ths->pFsm->FpRestoreFinishCb(ths->pFsm); } ths->restoreFinish = true; char eventLog[128]; snprintf(eventLog, sizeof(eventLog), "restore finish, index:%" PRId64, pEntry->index); syncNodeEventLog(ths, eventLog); } } rpcFreeCont(rpcMsg.pCont); syncEntryDestory(pEntry); } } } return 0; } bool syncNodeInRaftGroup(SSyncNode* ths, SRaftId* pRaftId) { for (int i = 0; i < ths->replicaNum; ++i) { if (syncUtilSameId(&((ths->replicasId)[i]), pRaftId)) { return true; } } return false; } SSyncSnapshotSender* syncNodeGetSnapshotSender(SSyncNode* ths, SRaftId* pDestId) { SSyncSnapshotSender* pSender = NULL; for (int i = 0; i < ths->replicaNum; ++i) { if (syncUtilSameId(pDestId, &((ths->replicasId)[i]))) { pSender = (ths->senders)[i]; } } return pSender; } bool syncNodeCanChange(SSyncNode* pSyncNode) { if (pSyncNode->changing) { sError("sync cannot change"); return false; } if ((pSyncNode->commitIndex >= SYNC_INDEX_BEGIN)) { SyncIndex lastIndex = syncNodeGetLastIndex(pSyncNode); if (pSyncNode->commitIndex != lastIndex) { sError("sync cannot change2"); return false; } } for (int i = 0; i < pSyncNode->peersNum; ++i) { SSyncSnapshotSender* pSender = syncNodeGetSnapshotSender(pSyncNode, &(pSyncNode->peersId)[i]); if (pSender->start) { sError("sync cannot change3"); return false; } } return true; }