SERVER-48816 Make a TenantDatabaseCloner skeleton class

9 files changed, 638 insertions, 14 deletions

diff --git a/src/mongo/db/repl/SConscript b/src/mongo/db/repl/SConscript
index 7c68f650174..c9aeb038459 100644
--- a/src/mongo/db/repl/SConscript
+++ b/src/mongo/db/repl/SConscript
@@ -689,8 +689,8 @@ env.Library(
         '$BUILD_DIR/mongo/rpc/metadata',
         '$BUILD_DIR/mongo/transport/transport_layer_common',
         '$BUILD_DIR/mongo/util/fail_point',
-        'initial_syncer',
         'data_replicator_external_state_initial_sync',
+        'initial_syncer',
         'repl_coordinator_interface',
         'repl_settings',
         'replica_set_messages',
@@ -698,6 +698,7 @@ env.Library(
         'replication_process',
         'reporter',
         'scatter_gather',
+        'tenant_migration_cloners',
         'topology_coordinator',
     ],
     LIBDEPS_PRIVATE=[
@@ -967,16 +968,16 @@ env.Library(
     target='initial_sync_cloners',
     source=[
         'all_database_cloner.cpp',
-        'base_cloner.cpp',
         'collection_cloner.cpp',
         'database_cloner.cpp',
-        env.Idlc('database_cloner.idl')[0]
     ],
     LIBDEPS = [
-        'task_runner',
+        'base_cloner',
+        'database_cloner_common',
         'initial_sync_shared_data',
         'member_data',
         'replication_consistency_markers_impl',
+        'task_runner',
         '$BUILD_DIR/mongo/base',
         '$BUILD_DIR/mongo/client/clientdriver_network',
         '$BUILD_DIR/mongo/util/concurrency/thread_pool',
@@ -985,8 +986,6 @@ env.Library(
     LIBDEPS_PRIVATE=[
         'repl_server_parameters',
         'replication_auth',
-        '$BUILD_DIR/mongo/idl/idl_parser',
-        '$BUILD_DIR/mongo/db/catalog/collection_options',
         '$BUILD_DIR/mongo/db/commands/list_collections_filter',
         '$BUILD_DIR/mongo/db/index_builds_coordinator_interface',
         '$BUILD_DIR/mongo/db/index_build_entry_helpers',
@@ -995,6 +994,23 @@ env.Library(
 )
 
 env.Library(
+    target='tenant_migration_cloners',
+    source=[
+        'tenant_collection_cloner.cpp',
+        'tenant_database_cloner.cpp',
+    ],
+    LIBDEPS = [
+        'base_cloner',
+        'database_cloner_common',
+        'initial_sync_shared_data',
+        '$BUILD_DIR/mongo/base',
+    ],
+    LIBDEPS_PRIVATE=[
+        '$BUILD_DIR/mongo/db/commands/list_collections_filter',
+    ]
+)
+
+env.Library(
     target='task_runner',
     source=[
         'task_runner.cpp',
@@ -1085,6 +1101,38 @@ env.Library(
 )
 
 env.Library(
+    target='base_cloner',
+    source=[
+        'base_cloner.cpp',
+    ],
+    LIBDEPS=[
+        'initial_sync_shared_data',
+        '$BUILD_DIR/mongo/base',
+        '$BUILD_DIR/mongo/client/clientdriver_network',
+        '$BUILD_DIR/mongo/util/fail_point',
+    ],
+    LIBDEPS_PRIVATE=[
+        'repl_server_parameters',
+        'replication_consistency_markers_impl',
+    ]
+)
+
+env.Library(
+    target='database_cloner_common',
+    source=[
+        'database_cloner_common.cpp',
+        env.Idlc('database_cloner.idl')[0],
+    ],
+    LIBDEPS=[
+        '$BUILD_DIR/mongo/base',
+    ],
+    LIBDEPS_PRIVATE=[
+        '$BUILD_DIR/mongo/db/catalog/collection_options',
+        '$BUILD_DIR/mongo/idl/idl_parser',
+    ]
+)
+
+env.Library(
     target='initial_syncer',
     source=[
         'initial_syncer.cpp',
diff --git a/src/mongo/db/repl/database_cloner.cpp b/src/mongo/db/repl/database_cloner.cpp
index ce5bb9053cc..f9bcd83fc05 100644
--- a/src/mongo/db/repl/database_cloner.cpp
+++ b/src/mongo/db/repl/database_cloner.cpp
@@ -34,6 +34,7 @@
 #include "mongo/base/string_data.h"
 #include "mongo/db/commands/list_collections_filter.h"
 #include "mongo/db/repl/database_cloner.h"
+#include "mongo/db/repl/database_cloner_common.h"
 #include "mongo/db/repl/database_cloner_gen.h"
 #include "mongo/logv2/log.h"
 #include "mongo/util/assert_util.h"
@@ -58,11 +59,6 @@ BaseCloner::ClonerStages DatabaseCloner::getStages() {
     return {&_listCollectionsStage};
 }
 
-/* static */
-CollectionOptions DatabaseCloner::parseCollectionOptions(const BSONObj& obj) {
-    return uassertStatusOK(CollectionOptions::parse(obj, CollectionOptions::parseForStorage));
-}
-
 void DatabaseCloner::preStage() {
     stdx::lock_guard<Latch> lk(_mutex);
     _stats.start = getSharedData()->getClock()->now();
diff --git a/src/mongo/db/repl/database_cloner.idl b/src/mongo/db/repl/database_cloner.idl
index 738098a087d..c5d5d3342d9 100644
--- a/src/mongo/db/repl/database_cloner.idl
+++ b/src/mongo/db/repl/database_cloner.idl
@@ -28,8 +28,7 @@
 global:
     cpp_namespace: mongo::repl
     cpp_includes:
-      - "mongo/db/catalog/collection_options.h"
-      - "mongo/db/repl/database_cloner.h"
+      - "mongo/db/repl/database_cloner_common.h"
 
 imports:
     - "mongo/idl/basic_types.idl"
@@ -39,7 +38,7 @@ types:
       description: "CollectionOptions as passed to createCollection()"
       cpp_type: "mongo::CollectionOptions"
       bson_serialization_type: object
-      deserializer: mongo::repl::DatabaseCloner::parseCollectionOptions
+      deserializer: mongo::repl::parseCollectionOptionsForDatabaseCloner
       serializer: mongo::CollectionOptions::toBSON
       
 structs:
diff --git a/src/mongo/db/repl/database_cloner_common.cpp b/src/mongo/db/repl/database_cloner_common.cpp
new file mode 100644
index 00000000000..68ebd7c333a
--- /dev/null
+++ b/src/mongo/db/repl/database_cloner_common.cpp
@@ -0,0 +1,42 @@
+/**
+ *    Copyright (C) 2020-present MongoDB, Inc.
+ *
+ *    This program is free software: you can redistribute it and/or modify
+ *    it under the terms of the Server Side Public License, version 1,
+ *    as published by MongoDB, Inc.
+ *
+ *    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.  See the
+ *    Server Side Public License for more details.
+ *
+ *    You should have received a copy of the Server Side Public License
+ *    along with this program. If not, see
+ *    <http://www.mongodb.com/licensing/server-side-public-license>.
+ *
+ *    As a special exception, the copyright holders give permission to link the
+ *    code of portions of this program with the OpenSSL library under certain
+ *    conditions as described in each individual source file and distribute
+ *    linked combinations including the program with the OpenSSL library. You
+ *    must comply with the Server Side Public License in all respects for
+ *    all of the code used other than as permitted herein. If you modify file(s)
+ *    with this exception, you may extend this exception to your version of the
+ *    file(s), but you are not obligated to do so. If you do not wish to do so,
+ *    delete this exception statement from your version. If you delete this
+ *    exception statement from all source files in the program, then also delete
+ *    it in the license file.
+ */
+
+#include "mongo/platform/basic.h"
+
+#include "mongo/db/repl/database_cloner_common.h"
+
+namespace mongo {
+namespace repl {
+
+CollectionOptions parseCollectionOptionsForDatabaseCloner(const BSONObj& obj) {
+    return uassertStatusOK(CollectionOptions::parse(obj, CollectionOptions::parseForStorage));
+}
+
+}  // namespace repl
+}  // namespace mongo
\ No newline at end of file
diff --git a/src/mongo/db/repl/database_cloner_common.h b/src/mongo/db/repl/database_cloner_common.h
new file mode 100644
index 00000000000..24f43cfb066
--- /dev/null
+++ b/src/mongo/db/repl/database_cloner_common.h
@@ -0,0 +1,39 @@
+/**
+ *    Copyright (C) 2020-present MongoDB, Inc.
+ *
+ *    This program is free software: you can redistribute it and/or modify
+ *    it under the terms of the Server Side Public License, version 1,
+ *    as published by MongoDB, Inc.
+ *
+ *    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.  See the
+ *    Server Side Public License for more details.
+ *
+ *    You should have received a copy of the Server Side Public License
+ *    along with this program. If not, see
+ *    <http://www.mongodb.com/licensing/server-side-public-license>.
+ *
+ *    As a special exception, the copyright holders give permission to link the
+ *    code of portions of this program with the OpenSSL library under certain
+ *    conditions as described in each individual source file and distribute
+ *    linked combinations including the program with the OpenSSL library. You
+ *    must comply with the Server Side Public License in all respects for
+ *    all of the code used other than as permitted herein. If you modify file(s)
+ *    with this exception, you may extend this exception to your version of the
+ *    file(s), but you are not obligated to do so. If you do not wish to do so,
+ *    delete this exception statement from your version. If you delete this
+ *    exception statement from all source files in the program, then also delete
+ *    it in the license file.
+ */
+
+#pragma once
+
+#include "mongo/db/catalog/collection_options.h"
+
+namespace mongo {
+namespace repl {
+
+CollectionOptions parseCollectionOptionsForDatabaseCloner(const BSONObj& obj);
+}
+}  // namespace mongo
\ No newline at end of file
diff --git a/src/mongo/db/repl/tenant_collection_cloner.cpp b/src/mongo/db/repl/tenant_collection_cloner.cpp
new file mode 100644
index 00000000000..be1d6c0a3de
--- /dev/null
+++ b/src/mongo/db/repl/tenant_collection_cloner.cpp
@@ -0,0 +1,93 @@
+/**
+ *    Copyright (C) 2020-present MongoDB, Inc.
+ *
+ *    This program is free software: you can redistribute it and/or modify
+ *    it under the terms of the Server Side Public License, version 1,
+ *    as published by MongoDB, Inc.
+ *
+ *    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.  See the
+ *    Server Side Public License for more details.
+ *
+ *    You should have received a copy of the Server Side Public License
+ *    along with this program. If not, see
+ *    <http://www.mongodb.com/licensing/server-side-public-license>.
+ *
+ *    As a special exception, the copyright holders give permission to link the
+ *    code of portions of this program with the OpenSSL library under certain
+ *    conditions as described in each individual source file and distribute
+ *    linked combinations including the program with the OpenSSL library. You
+ *    must comply with the Server Side Public License in all respects for
+ *    all of the code used other than as permitted herein. If you modify file(s)
+ *    with this exception, you may extend this exception to your version of the
+ *    file(s), but you are not obligated to do so. If you do not wish to do so,
+ *    delete this exception statement from your version. If you delete this
+ *    exception statement from all source files in the program, then also delete
+ *    it in the license file.
+ */
+
+#define MONGO_LOGV2_DEFAULT_COMPONENT ::mongo::logv2::LogComponent::kReplicationInitialSync
+
+#include "mongo/platform/basic.h"
+
+#include "mongo/base/string_data.h"
+#include "mongo/db/commands/list_collections_filter.h"
+#include "mongo/db/repl/database_cloner_gen.h"
+#include "mongo/db/repl/tenant_collection_cloner.h"
+#include "mongo/logv2/log.h"
+#include "mongo/util/assert_util.h"
+
+namespace mongo {
+namespace repl {
+
+TenantCollectionCloner::TenantCollectionCloner(const NamespaceString& sourceNss,
+                                               const CollectionOptions& collectionOptions,
+                                               InitialSyncSharedData* sharedData,
+                                               const HostAndPort& source,
+                                               DBClientConnection* client,
+                                               StorageInterface* storageInterface,
+                                               ThreadPool* dbPool)
+    : BaseCloner("TenantCollectionCloner"_sd, sharedData, source, client, storageInterface, dbPool),
+      _sourceNss(sourceNss),
+      _collectionOptions(collectionOptions),
+      _sourceDbAndUuid(NamespaceString("UNINITIALIZED")),
+      _placeholderStage("placeholder", this, &TenantCollectionCloner::placeholderStage) {
+    invariant(sourceNss.isValid());
+    invariant(collectionOptions.uuid);
+    _sourceDbAndUuid = NamespaceStringOrUUID(sourceNss.db().toString(), *collectionOptions.uuid);
+    _stats.ns = _sourceNss.ns();
+}
+
+BaseCloner::ClonerStages TenantCollectionCloner::getStages() {
+    return {&_placeholderStage};
+}
+
+BaseCloner::AfterStageBehavior TenantCollectionCloner::placeholderStage() {
+    return kContinueNormally;
+}
+
+TenantCollectionCloner::Stats TenantCollectionCloner::getStats() const {
+    stdx::lock_guard<Latch> lk(_mutex);
+    return _stats;
+}
+
+void TenantCollectionCloner::Stats::append(BSONObjBuilder* builder) const {
+    builder->appendNumber(kDocumentsToCopyFieldName, documentToCopy);
+    builder->appendNumber(kDocumentsCopiedFieldName, documentsCopied);
+    builder->appendNumber("indexes", indexes);
+    builder->appendNumber("insertedBatches", insertedBatches);
+    if (start != Date_t()) {
+        builder->appendDate("start", start);
+        if (end != Date_t()) {
+            builder->appendDate("end", end);
+            auto elapsed = end - start;
+            long long elapsedMillis = duration_cast<Milliseconds>(elapsed).count();
+            builder->appendNumber("elapsedMillis", elapsedMillis);
+        }
+    }
+    builder->appendNumber("receivedBatches", receivedBatches);
+}
+
+}  // namespace repl
+}  // namespace mongo
diff --git a/src/mongo/db/repl/tenant_collection_cloner.h b/src/mongo/db/repl/tenant_collection_cloner.h
new file mode 100644
index 00000000000..ee8c9bf2ced
--- /dev/null
+++ b/src/mongo/db/repl/tenant_collection_cloner.h
@@ -0,0 +1,106 @@
+/**
+ *    Copyright (C) 2020-present MongoDB, Inc.
+ *
+ *    This program is free software: you can redistribute it and/or modify
+ *    it under the terms of the Server Side Public License, version 1,
+ *    as published by MongoDB, Inc.
+ *
+ *    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.  See the
+ *    Server Side Public License for more details.
+ *
+ *    You should have received a copy of the Server Side Public License
+ *    along with this program. If not, see
+ *    <http://www.mongodb.com/licensing/server-side-public-license>.
+ *
+ *    As a special exception, the copyright holders give permission to link the
+ *    code of portions of this program with the OpenSSL library under certain
+ *    conditions as described in each individual source file and distribute
+ *    linked combinations including the program with the OpenSSL library. You
+ *    must comply with the Server Side Public License in all respects for
+ *    all of the code used other than as permitted herein. If you modify file(s)
+ *    with this exception, you may extend this exception to your version of the
+ *    file(s), but you are not obligated to do so. If you do not wish to do so,
+ *    delete this exception statement from your version. If you delete this
+ *    exception statement from all source files in the program, then also delete
+ *    it in the license file.
+ */
+
+#pragma once
+
+#include <memory>
+#include <vector>
+
+#include "mongo/db/repl/base_cloner.h"
+#include "mongo/db/repl/task_runner.h"
+#include "mongo/util/progress_meter.h"
+
+namespace mongo {
+namespace repl {
+
+class TenantCollectionCloner : public BaseCloner {
+public:
+    struct Stats {
+        static constexpr StringData kDocumentsToCopyFieldName = "documentsToCopy"_sd;
+        static constexpr StringData kDocumentsCopiedFieldName = "documentsCopied"_sd;
+
+        std::string ns;
+        Date_t start;
+        Date_t end;
+        size_t documentToCopy{0};
+        size_t documentsCopied{0};
+        size_t indexes{0};
+        size_t insertedBatches{0};
+        size_t receivedBatches{0};
+
+        std::string toString() const;
+        BSONObj toBSON() const;
+        void append(BSONObjBuilder* builder) const;
+    };
+
+    TenantCollectionCloner(const NamespaceString& ns,
+                           const CollectionOptions& collectionOptions,
+                           InitialSyncSharedData* sharedData,
+                           const HostAndPort& source,
+                           DBClientConnection* client,
+                           StorageInterface* storageInterface,
+                           ThreadPool* dbPool);
+
+    virtual ~TenantCollectionCloner() = default;
+
+    Stats getStats() const;
+
+protected:
+    ClonerStages getStages() final;
+
+private:
+    std::string describeForFuzzer(BaseClonerStage* stage) const final {
+        return _sourceNss.db() + " db: { " + stage->getName() + ": UUID(\"" +
+            _sourceDbAndUuid.uuid()->toString() + "\") coll: " + _sourceNss.coll() + " }";
+    }
+
+    /**
+     * Temporary no-op stage.
+     */
+    AfterStageBehavior placeholderStage();
+
+    // All member variables are labeled with one of the following codes indicating the
+    // synchronization rules for accessing them.
+    //
+    // (R)  Read-only in concurrent operation; no synchronization required.
+    // (S)  Self-synchronizing; access according to class's own rules.
+    // (M)  Reads and writes guarded by _mutex (defined in base class).
+    // (X)  Access only allowed from the main flow of control called from run() or constructor.
+    const NamespaceString _sourceNss;            // (R)
+    const CollectionOptions _collectionOptions;  // (R)
+    // Despite the type name, this member must always contain a UUID.
+    NamespaceStringOrUUID _sourceDbAndUuid;  // (R)
+
+    ClonerStage<TenantCollectionCloner> _placeholderStage;  // (R)
+
+    Stats _stats;  // (M)
+};
+
+}  // namespace repl
+}  // namespace mongo
\ No newline at end of file
diff --git a/src/mongo/db/repl/tenant_database_cloner.cpp b/src/mongo/db/repl/tenant_database_cloner.cpp
new file mode 100644
index 00000000000..9e9fbc5c72b
--- /dev/null
+++ b/src/mongo/db/repl/tenant_database_cloner.cpp
@@ -0,0 +1,171 @@
+/**
+ *    Copyright (C) 2020-present MongoDB, Inc.
+ *
+ *    This program is free software: you can redistribute it and/or modify
+ *    it under the terms of the Server Side Public License, version 1,
+ *    as published by MongoDB, Inc.
+ *
+ *    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.  See the
+ *    Server Side Public License for more details.
+ *
+ *    You should have received a copy of the Server Side Public License
+ *    along with this program. If not, see
+ *    <http://www.mongodb.com/licensing/server-side-public-license>.
+ *
+ *    As a special exception, the copyright holders give permission to link the
+ *    code of portions of this program with the OpenSSL library under certain
+ *    conditions as described in each individual source file and distribute
+ *    linked combinations including the program with the OpenSSL library. You
+ *    must comply with the Server Side Public License in all respects for
+ *    all of the code used other than as permitted herein. If you modify file(s)
+ *    with this exception, you may extend this exception to your version of the
+ *    file(s), but you are not obligated to do so. If you do not wish to do so,
+ *    delete this exception statement from your version. If you delete this
+ *    exception statement from all source files in the program, then also delete
+ *    it in the license file.
+ */
+
+#define MONGO_LOGV2_DEFAULT_COMPONENT ::mongo::logv2::LogComponent::kReplicationInitialSync
+
+#include "mongo/platform/basic.h"
+
+#include "mongo/base/string_data.h"
+#include "mongo/db/commands/list_collections_filter.h"
+#include "mongo/db/repl/database_cloner_gen.h"
+#include "mongo/db/repl/tenant_collection_cloner.h"
+#include "mongo/db/repl/tenant_database_cloner.h"
+#include "mongo/logv2/log.h"
+#include "mongo/util/assert_util.h"
+
+namespace mongo {
+namespace repl {
+
+TenantDatabaseCloner::TenantDatabaseCloner(const std::string& dbName,
+                                           InitialSyncSharedData* sharedData,
+                                           const HostAndPort& source,
+                                           DBClientConnection* client,
+                                           StorageInterface* storageInterface,
+                                           ThreadPool* dbPool)
+    : BaseCloner("TenantDatabaseCloner"_sd, sharedData, source, client, storageInterface, dbPool),
+      _dbName(dbName),
+      _listCollectionsStage("listCollections", this, &TenantDatabaseCloner::listCollectionsStage) {
+    invariant(!dbName.empty());
+    _stats.dbname = dbName;
+}
+
+BaseCloner::ClonerStages TenantDatabaseCloner::getStages() {
+    return {&_listCollectionsStage};
+}
+
+void TenantDatabaseCloner::preStage() {
+    stdx::lock_guard<Latch> lk(_mutex);
+    _stats.start = getSharedData()->getClock()->now();
+}
+
+BaseCloner::AfterStageBehavior TenantDatabaseCloner::listCollectionsStage() {
+    // TODO(SERVER-48816): Implement this stage.
+    return kContinueNormally;
+}
+
+bool TenantDatabaseCloner::isMyFailPoint(const BSONObj& data) const {
+    return data["database"].str() == _dbName && BaseCloner::isMyFailPoint(data);
+}
+
+void TenantDatabaseCloner::postStage() {
+    {
+        stdx::lock_guard<Latch> lk(_mutex);
+        _stats.collections = _collections.size();
+        _stats.collectionStats.reserve(_collections.size());
+        for (const auto& coll : _collections) {
+            _stats.collectionStats.emplace_back();
+            _stats.collectionStats.back().ns = coll.first.ns();
+        }
+    }
+    for (const auto& coll : _collections) {
+        auto& sourceNss = coll.first;
+        auto& collectionOptions = coll.second;
+        {
+            stdx::lock_guard<Latch> lk(_mutex);
+            _currentCollectionCloner =
+                std::make_unique<TenantCollectionCloner>(sourceNss,
+                                                         collectionOptions,
+                                                         getSharedData(),
+                                                         getSource(),
+                                                         getClient(),
+                                                         getStorageInterface(),
+                                                         getDBPool());
+        }
+        auto collStatus = _currentCollectionCloner->run();
+        if (collStatus.isOK()) {
+            LOGV2_DEBUG(
+                4881600, 1, "Tenant collection clone finished", "namespace"_attr = sourceNss);
+        } else {
+            LOGV2_ERROR(4881601,
+                        "Tenant collection clone failed",
+                        "namespace"_attr = sourceNss,
+                        "error"_attr = collStatus.toString());
+            setInitialSyncFailedStatus(
+                {collStatus.code(),
+                 collStatus
+                     .withContext(str::stream()
+                                  << "Error cloning collection '" << sourceNss.toString() << "'")
+                     .toString()});
+        }
+        {
+            stdx::lock_guard<Latch> lk(_mutex);
+            _stats.collectionStats[_stats.clonedCollections] = _currentCollectionCloner->getStats();
+            _currentCollectionCloner = nullptr;
+            // Abort the tenant database cloner if the collection clone failed.
+            if (!collStatus.isOK())
+                return;
+            _stats.clonedCollections++;
+        }
+    }
+    stdx::lock_guard<Latch> lk(_mutex);
+    _stats.end = getSharedData()->getClock()->now();
+}
+
+TenantDatabaseCloner::Stats TenantDatabaseCloner::getStats() const {
+    stdx::lock_guard<Latch> lk(_mutex);
+    TenantDatabaseCloner::Stats stats = _stats;
+    if (_currentCollectionCloner) {
+        stats.collectionStats[_stats.clonedCollections] = _currentCollectionCloner->getStats();
+    }
+    return stats;
+}
+
+std::string TenantDatabaseCloner::Stats::toString() const {
+    return toBSON().toString();
+}
+
+BSONObj TenantDatabaseCloner::Stats::toBSON() const {
+    BSONObjBuilder bob;
+    bob.append("dbname", dbname);
+    append(&bob);
+    return bob.obj();
+}
+
+void TenantDatabaseCloner::Stats::append(BSONObjBuilder* builder) const {
+    builder->appendNumber("collections", collections);
+    builder->appendNumber("clonedCollections", clonedCollections);
+    if (start != Date_t()) {
+        builder->appendDate("start", start);
+        if (end != Date_t()) {
+            builder->appendDate("end", end);
+            auto elapsed = end - start;
+            long long elapsedMillis = duration_cast<Milliseconds>(elapsed).count();
+            builder->appendNumber("elapsedMillis", elapsedMillis);
+        }
+    }
+
+    for (auto&& collection : collectionStats) {
+        BSONObjBuilder collectionBuilder(builder->subobjStart(collection.ns));
+        collection.append(&collectionBuilder);
+        collectionBuilder.doneFast();
+    }
+}
+
+}  // namespace repl
+}  // namespace mongo
diff --git a/src/mongo/db/repl/tenant_database_cloner.h b/src/mongo/db/repl/tenant_database_cloner.h
new file mode 100644
index 00000000000..eaa681118f2
--- /dev/null
+++ b/src/mongo/db/repl/tenant_database_cloner.h
@@ -0,0 +1,130 @@
+/**
+ *    Copyright (C) 2020-present MongoDB, Inc.
+ *
+ *    This program is free software: you can redistribute it and/or modify
+ *    it under the terms of the Server Side Public License, version 1,
+ *    as published by MongoDB, Inc.
+ *
+ *    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.  See the
+ *    Server Side Public License for more details.
+ *
+ *    You should have received a copy of the Server Side Public License
+ *    along with this program. If not, see
+ *    <http://www.mongodb.com/licensing/server-side-public-license>.
+ *
+ *    As a special exception, the copyright holders give permission to link the
+ *    code of portions of this program with the OpenSSL library under certain
+ *    conditions as described in each individual source file and distribute
+ *    linked combinations including the program with the OpenSSL library. You
+ *    must comply with the Server Side Public License in all respects for
+ *    all of the code used other than as permitted herein. If you modify file(s)
+ *    with this exception, you may extend this exception to your version of the
+ *    file(s), but you are not obligated to do so. If you do not wish to do so,
+ *    delete this exception statement from your version. If you delete this
+ *    exception statement from all source files in the program, then also delete
+ *    it in the license file.
+ */
+
+#pragma once
+
+#include <vector>
+
+#include "mongo/db/repl/base_cloner.h"
+#include "mongo/db/repl/tenant_collection_cloner.h"
+
+namespace mongo {
+namespace repl {
+
+class TenantDatabaseCloner final : public BaseCloner {
+public:
+    struct Stats {
+        std::string dbname;
+        Date_t start;
+        Date_t end;
+        size_t collections{0};
+        size_t clonedCollections{0};
+        std::vector<TenantCollectionCloner::Stats> collectionStats;
+
+        std::string toString() const;
+        BSONObj toBSON() const;
+        void append(BSONObjBuilder* builder) const;
+    };
+
+    TenantDatabaseCloner(const std::string& dbName,
+                         InitialSyncSharedData* sharedData,
+                         const HostAndPort& source,
+                         DBClientConnection* client,
+                         StorageInterface* storageInterface,
+                         ThreadPool* dbPool);
+
+    virtual ~TenantDatabaseCloner() = default;
+
+    Stats getStats() const;
+
+    std::string toString() const;
+
+protected:
+    ClonerStages getStages() final;
+
+    bool isMyFailPoint(const BSONObj& data) const final;
+
+private:
+    // TODO(SERVER-48816): implement unit tests
+    // friend class TenantDatabaseclonerTest;
+
+    class TenantDatabaseClonerStage : public ClonerStage<TenantDatabaseCloner> {
+    public:
+        TenantDatabaseClonerStage(std::string name,
+                                  TenantDatabaseCloner* cloner,
+                                  ClonerRunFn stageFunc)
+            : ClonerStage<TenantDatabaseCloner>(name, cloner, stageFunc) {}
+
+        bool isTransientError(const Status& status) override {
+            // Always abort on error.
+            return false;
+        }
+    };
+
+    /**
+     * Stage function that retrieves collection information from the donor.
+     */
+    AfterStageBehavior listCollectionsStage();
+
+    /**
+     * The preStage sets the start time in _stats.
+     */
+    void preStage() final;
+
+    /**
+     * The postStage creates and runs the individual TenantCollectionCloners on each database found
+     * on the sync source, and sets the end time in _stats when done.
+     */
+    void postStage() final;
+
+    std::string describeForFuzzer(BaseClonerStage* stage) const final {
+        return _dbName + " db: { " + stage->getName() + ": 1 } ";
+    }
+
+    // All member variables are labeled with one of the following codes indicating the
+    // synchronization rules for accessing them.
+    //
+    // (R)  Read-only in concurrent operation; no synchronization required.
+    // (S)  Self-synchronizing; access according to class's own rules.
+    // (M)  Reads and writes guarded by _mutex (defined in base class).
+    // (X)  Access only allowed from the main flow of control called from run() or constructor.
+    // (MX) Write access with mutex from main flow of control, read access with mutex from other
+    //      threads, read access allowed from main flow without mutex.
+    const std::string _dbName;                                                // (R)
+    std::vector<std::pair<NamespaceString, CollectionOptions>> _collections;  // (X)
+    std::unique_ptr<TenantCollectionCloner> _currentCollectionCloner;         // (MX)
+
+    TenantDatabaseClonerStage _listCollectionsStage;  // (R)
+
+    Stats _stats;  // (MX)
+};
+
+
+}  // namespace repl
+}  // namespace mongo
\ No newline at end of file