#include "database_service.hpp" #include #include #include namespace smartbotic::database { DatabaseService::DatabaseService(Config config) : config_(std::move(config)) { } DatabaseService::~DatabaseService() { stop(); } bool DatabaseService::initialize() { spdlog::info("Initializing database service (node: {})", config_.nodeId); try { // Create data directory if needed std::error_code ec; std::filesystem::create_directories(config_.dataDirectory, ec); if (ec) { spdlog::error("Failed to create data directory: {}", ec.message()); return false; } setupComponents(); // Initialize encryption if (config_.encryptionEnabled) { if (!encryption_->initialize()) { spdlog::error("Failed to initialize encryption"); return false; } } // Recover from persistence persistence_->recover(*store_); // Set replication sequence after recovery (WAL sequence is now known) replication_->setSequence(persistence_->currentWalSequence()); // Load view definitions from the _views system collection (which is now // populated by the persistence recovery above). view_manager_->loadFromStore(); // Run migrations if enabled if (config_.migrations.enabled && !config_.migrations.directory.empty()) { if (!runMigrations()) { if (config_.migrations.failOnError) { spdlog::error("Failed to run migrations"); return false; } spdlog::warn("Some migrations failed, continuing anyway"); } } spdlog::info("Database service initialized successfully"); return true; } catch (const std::exception& e) { spdlog::error("Failed to initialize database service: {}", e.what()); return false; } } bool DatabaseService::runMigrations() { if (!config_.migrations.enabled) { return true; } MigrationRunner::Config migrationConfig; migrationConfig.directory = config_.migrations.directory; migrationConfig.autoApply = config_.migrations.autoApply; migrationConfig.failOnError = config_.migrations.failOnError; migrationRunner_ = std::make_unique(*store_, *view_manager_, migrationConfig); return migrationRunner_->runMigrations(); } void DatabaseService::start() { if (running_.exchange(true)) { return; } spdlog::info("Starting database service on {}:{}", config_.bindAddress, config_.rpcPort); // Start components store_->start(); persistence_->start(); events_->start(); files_->start(); if (config_.replicationEnabled) { replication_->start(); // Set initial local collections for discovery replication_->setLocalCollections(store_->listCollections()); } // Start gRPC server startGrpcServer(); spdlog::info("Database service started"); } void DatabaseService::stop() { if (!running_.exchange(false)) { return; } spdlog::info("Stopping database service..."); // Stop gRPC server first stopGrpcServer(); // Stop components in reverse order if (replication_) { replication_->stop(); } if (files_) { files_->stop(); } if (events_) { events_->stop(); } if (persistence_) { persistence_->stop(); } if (store_) { store_->stop(); } spdlog::info("Database service stopped"); } void DatabaseService::wait() { if (serverThread_.joinable()) { serverThread_.join(); } } void DatabaseService::signalStop() { stopRequested_ = true; stop(); } nlohmann::json DatabaseService::getStats() const { nlohmann::json stats; if (store_) { auto storeStats = store_->getStats(); stats["documents"] = storeStats.totalDocuments; stats["collections"] = storeStats.totalCollections; stats["memory_bytes"] = storeStats.estimatedMemoryBytes; stats["inserts"] = storeStats.insertCount; stats["updates"] = storeStats.updateCount; stats["deletes"] = storeStats.deleteCount; stats["queries"] = storeStats.queryCount; } if (persistence_) { auto persistStats = persistence_->getStats(); stats["wal_sequence"] = persistStats.walSequence; stats["wal_size_bytes"] = persistStats.walSizeBytes; stats["snapshot_count"] = persistStats.snapshotCount; stats["last_snapshot_sequence"] = persistStats.lastSnapshotSequence; } if (events_) { stats["subscriptions"] = events_->subscriptionCount(); } return stats; } DatabaseService::Config DatabaseService::loadConfig(const std::filesystem::path& configPath) { std::ifstream file(configPath); if (!file) { throw std::runtime_error("Failed to open config file: " + configPath.string()); } nlohmann::json json; file >> json; Config config; // Check for both "storage" and "database" keys for backward compatibility nlohmann::json* dbConfig = nullptr; if (json.contains("database")) { dbConfig = &json["database"]; } else if (json.contains("storage")) { dbConfig = &json["storage"]; } if (dbConfig) { auto& db = *dbConfig; config.nodeId = db.value("node_id", config.nodeId); config.bindAddress = db.value("bind_address", config.bindAddress); config.rpcPort = db.value("rpc_port", config.rpcPort); // Expand environment variables in data directory std::string dataDir = db.value("data_directory", ""); if (dataDir.find("${HOME}") != std::string::npos) { const char* home = std::getenv("HOME"); if (home) { size_t pos = dataDir.find("${HOME}"); dataDir.replace(pos, 7, home); } } config.dataDirectory = dataDir; // Migrations settings if (db.contains("migrations")) { auto& migrations = db["migrations"]; config.migrations.enabled = migrations.value("enabled", config.migrations.enabled); config.migrations.autoApply = migrations.value("auto_apply", config.migrations.autoApply); config.migrations.failOnError = migrations.value("fail_on_error", config.migrations.failOnError); std::string migDir = migrations.value("directory", ""); if (migDir.find("${HOME}") != std::string::npos) { const char* home = std::getenv("HOME"); if (home) { size_t pos = migDir.find("${HOME}"); migDir.replace(pos, 7, home); } } config.migrations.directory = migDir; } // Memory eviction settings if (db.contains("memory")) { auto& memory = db["memory"]; config.maxMemoryMb = memory.value("max_memory_mb", config.maxMemoryMb); config.evictionThresholdPercent = memory.value("eviction_threshold_percent", config.evictionThresholdPercent); config.evictionTargetPercent = memory.value("eviction_target_percent", config.evictionTargetPercent); config.evictionCheckIntervalMs = memory.value("eviction_check_interval_ms", config.evictionCheckIntervalMs); } // Persistence settings if (db.contains("persistence")) { auto& persistence = db["persistence"]; config.walSyncIntervalMs = persistence.value("wal_sync_interval_ms", config.walSyncIntervalMs); config.snapshotIntervalSec = persistence.value("snapshot_interval_sec", config.snapshotIntervalSec); config.compressionEnabled = persistence.value("compression", "lz4") != "none"; } // File settings if (db.contains("files")) { auto& files = db["files"]; config.maxFileSizeMb = files.value("max_file_size_mb", config.maxFileSizeMb); config.fileCleanupIntervalSec = files.value("cleanup_orphans_interval_sec", config.fileCleanupIntervalSec); if (files.contains("allowed_types")) { for (const auto& type : files["allowed_types"]) { config.allowedFileTypes.push_back(type.get()); } } } // Encryption settings if (db.contains("encryption")) { auto& encryption = db["encryption"]; config.encryptionEnabled = encryption.value("enabled", config.encryptionEnabled); config.autoGenerateKey = encryption.value("auto_generate_key", config.autoGenerateKey); std::string keyFile = encryption.value("key_file", ""); if (keyFile.find("${HOME}") != std::string::npos) { const char* home = std::getenv("HOME"); if (home) { size_t pos = keyFile.find("${HOME}"); keyFile.replace(pos, 7, home); } } config.keyFilePath = keyFile; } // Replication settings if (db.contains("replication")) { auto& replication = db["replication"]; config.replicationEnabled = replication.value("enabled", config.replicationEnabled); config.conflictResolution = replication.value("conflict_resolution", config.conflictResolution); if (replication.contains("peers")) { for (const auto& peer : replication["peers"]) { config.peerAddresses.push_back(peer.get()); } } } } return config; } void DatabaseService::setupComponents() { // Create memory store with eviction config MemoryStore::Config storeConfig; storeConfig.nodeId = config_.nodeId; storeConfig.maxMemoryBytes = config_.maxMemoryMb * 1024ULL * 1024ULL; storeConfig.evictionThresholdPercent = config_.evictionThresholdPercent; storeConfig.evictionTargetPercent = config_.evictionTargetPercent; storeConfig.evictionCheckIntervalMs = config_.evictionCheckIntervalMs; store_ = std::make_unique(storeConfig); // Create view manager (cache loaded in initialize() after persistence recovery) view_manager_ = std::make_unique(*store_); // Create persistence manager PersistenceManager::Config persistConfig; persistConfig.dataDir = config_.dataDirectory; persistConfig.walSyncIntervalMs = config_.walSyncIntervalMs; persistConfig.snapshotIntervalSec = config_.snapshotIntervalSec; persistConfig.compressionEnabled = config_.compressionEnabled; persistence_ = std::make_unique(persistConfig); // Connect store callbacks to persistence - uses setPersistCallback for WAL logging store_->setPersistCallback([this](const std::string& collection, const std::string& id, const std::optional& doc, EventType eventType) { // Log to WAL based on event type switch (eventType) { case EventType::INSERT: if (doc) persistence_->logInsert(collection, *doc); break; case EventType::UPDATE: if (doc) persistence_->logUpdate(collection, *doc); break; case EventType::DELETE: persistence_->logDelete(collection, id); break; default: break; } // Queue for replication broadcast if (replication_ && config_.replicationEnabled) { databasepb::ReplicationEntry entry; entry.set_collection(collection); entry.set_document_id(id); entry.set_global_timestamp( std::chrono::duration_cast( std::chrono::system_clock::now().time_since_epoch() ).count()); switch (eventType) { case EventType::INSERT: entry.set_op(databasepb::OP_INSERT); if (doc) entry.set_data(doc->data.dump()); break; case EventType::UPDATE: entry.set_op(databasepb::OP_UPDATE); if (doc) entry.set_data(doc->data.dump()); break; case EventType::DELETE: entry.set_op(databasepb::OP_DELETE); break; default: break; } replication_->queueForReplication(entry); } // Publish event if (events_) { DatabaseEvent event; event.type = eventType; event.collection = collection; event.documentId = id; event.timestamp = std::chrono::duration_cast( std::chrono::system_clock::now().time_since_epoch() ).count(); event.nodeId = config_.nodeId; if (doc) { event.data = doc->data; } events_->publish(event); } }); // Set up document load callback for LRU eviction recovery store_->setDocumentLoadCallback([this](const std::string& collection, const std::string& id, uint64_t walSequence) -> std::optional { // Load document from WAL via persistence manager return persistence_->loadDocument(collection, id, 0); // Search from beginning }); // Create encryption manager EncryptionManager::Config encryptConfig; encryptConfig.enabled = config_.encryptionEnabled; encryptConfig.keyFilePath = config_.keyFilePath; encryptConfig.autoGenerateKey = config_.autoGenerateKey; encryption_ = std::make_unique(encryptConfig); // Create event manager EventManager::Config eventConfig; eventConfig.nodeId = config_.nodeId; events_ = std::make_unique(eventConfig); // Create file manager FileManager::Config fileConfig; fileConfig.filesDir = config_.dataDirectory / "files"; fileConfig.maxFileSizeMb = config_.maxFileSizeMb; fileConfig.allowedMimeTypes = config_.allowedFileTypes; fileConfig.cleanupIntervalSec = config_.fileCleanupIntervalSec; files_ = std::make_unique(fileConfig); // Create replication manager ReplicationManager::Config replConfig; replConfig.nodeId = config_.nodeId; replConfig.peerAddresses = config_.peerAddresses; replConfig.conflictResolution = config_.conflictResolution; replication_ = std::make_unique(replConfig); // Wire replication entry apply callback replication_->setEntryApplyCallback([this](const databasepb::ReplicationEntry& entry) { applyReplicatedEntry(entry); }); // Create gRPC implementations storageImpl_ = std::make_unique( *store_, *persistence_, *events_, *files_, *encryption_, *view_manager_ ); replicationImpl_ = std::make_unique( *store_, *replication_, *persistence_ ); } void DatabaseService::applyReplicatedEntry(const databasepb::ReplicationEntry& entry) { try { switch (entry.op()) { case databasepb::OP_INSERT: case databasepb::OP_UPDATE: case databasepb::OP_UPSERT: { if (entry.data().empty()) { spdlog::warn("Replication entry has no data for op {}", static_cast(entry.op())); return; } auto json = nlohmann::json::parse(entry.data()); Document doc = Document::fromJson(json); doc.id = entry.document_id(); doc.nodeId = entry.node_id(); // Use loadDocument to bypass normal callbacks (avoid re-replication) store_->loadDocument(entry.collection(), doc); spdlog::trace("Applied replicated {} to {}/{} from {}", entry.op() == databasepb::OP_INSERT ? "insert" : entry.op() == databasepb::OP_UPDATE ? "update" : "upsert", entry.collection(), entry.document_id(), entry.node_id()); break; } case databasepb::OP_DELETE: { store_->remove(entry.collection(), entry.document_id()); spdlog::trace("Applied replicated delete to {}/{} from {}", entry.collection(), entry.document_id(), entry.node_id()); break; } case databasepb::OP_CREATE_COLLECTION: { CollectionOptions options; store_->createCollection(entry.collection(), options); spdlog::trace("Applied replicated create collection {} from {}", entry.collection(), entry.node_id()); break; } case databasepb::OP_DROP_COLLECTION: { store_->dropCollection(entry.collection()); spdlog::trace("Applied replicated drop collection {} from {}", entry.collection(), entry.node_id()); break; } default: spdlog::warn("Unknown replication operation type: {}", static_cast(entry.op())); break; } } catch (const std::exception& e) { spdlog::error("Failed to apply replicated entry: {}", e.what()); } } void DatabaseService::startGrpcServer() { std::string serverAddress = config_.bindAddress + ":" + std::to_string(config_.rpcPort); grpc::ServerBuilder builder; builder.AddListeningPort(serverAddress, grpc::InsecureServerCredentials()); builder.RegisterService(storageImpl_.get()); builder.RegisterService(replicationImpl_.get()); // Configure server builder.SetMaxReceiveMessageSize(100 * 1024 * 1024); // 100MB for file uploads builder.SetMaxSendMessageSize(100 * 1024 * 1024); grpcServer_ = builder.BuildAndStart(); if (!grpcServer_) { throw std::runtime_error("Failed to start gRPC server on " + serverAddress); } spdlog::info("gRPC server listening on {}", serverAddress); } void DatabaseService::stopGrpcServer() { if (grpcServer_) { grpcServer_->Shutdown(); grpcServer_.reset(); } } } // namespace smartbotic::database