summaryrefslogtreecommitdiff
path: root/src/mongo/db/s/sharding_util.cpp
blob: 18f28d5041d885052b188bac0b1ad2199179d146 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
/**
 *    Copyright (C) 2021-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/db/s/sharding_util.h"

#include <fmt/format.h>

#include "mongo/db/catalog/index_builds_manager.h"
#include "mongo/db/catalog_raii.h"
#include "mongo/db/commands.h"
#include "mongo/db/concurrency/exception_util.h"
#include "mongo/db/index_builds_coordinator.h"
#include "mongo/db/s/shard_authoritative_catalog_gen.h"
#include "mongo/logv2/log.h"
#include "mongo/s/catalog/type_collection.h"
#include "mongo/s/catalog/type_index_catalog.h"
#include "mongo/s/cluster_commands_helpers.h"
#include "mongo/s/request_types/flush_routing_table_cache_updates_gen.h"

#define MONGO_LOGV2_DEFAULT_COMPONENT ::mongo::logv2::LogComponent::kSharding

namespace mongo {

namespace sharding_util {

using namespace fmt::literals;

void tellShardsToRefreshCollection(OperationContext* opCtx,
                                   const std::vector<ShardId>& shardIds,
                                   const NamespaceString& nss,
                                   const std::shared_ptr<executor::TaskExecutor>& executor) {
    auto cmd = FlushRoutingTableCacheUpdatesWithWriteConcern(nss);
    cmd.setSyncFromConfig(true);
    cmd.setDbName(nss.dbName());
    auto cmdObj = CommandHelpers::appendMajorityWriteConcern(cmd.toBSON({}));
    sendCommandToShards(opCtx, DatabaseName::kAdmin.db(), cmdObj, shardIds, executor);
}

std::vector<AsyncRequestsSender::Response> processShardResponses(
    OperationContext* opCtx,
    StringData dbName,
    const BSONObj& command,
    const std::vector<AsyncRequestsSender::Request>& requests,
    const std::shared_ptr<executor::TaskExecutor>& executor,
    bool throwOnError) {

    std::vector<AsyncRequestsSender::Response> responses;
    if (!requests.empty()) {
        // The _flushRoutingTableCacheUpdatesWithWriteConcern command will fail with a
        // QueryPlanKilled error response if the config.cache.chunks collection is dropped
        // concurrently. The config.cache.chunks collection is dropped by the shard when it detects
        // the sharded collection's epoch having changed. We use kIdempotentOrCursorInvalidated so
        // the ARS automatically retries in that situation.
        AsyncRequestsSender ars(opCtx,
                                executor,
                                dbName,
                                requests,
                                ReadPreferenceSetting(ReadPreference::PrimaryOnly),
                                Shard::RetryPolicy::kIdempotentOrCursorInvalidated,
                                nullptr /* resourceYielder */);

        while (!ars.done()) {
            // Retrieve the responses and throw at the first failure.
            auto response = ars.next();

            if (throwOnError) {
                const auto errorContext =
                    "Failed command {} for database '{}' on shard '{}'"_format(
                        command.toString(), dbName, StringData{response.shardId});

                uassertStatusOKWithContext(response.swResponse.getStatus(), errorContext);
                const auto& respBody = response.swResponse.getValue().data;

                const auto status = getStatusFromCommandResult(respBody);
                uassertStatusOKWithContext(status, errorContext);

                const auto wcStatus = getWriteConcernStatusFromCommandResult(respBody);
                uassertStatusOKWithContext(wcStatus, errorContext);
            }

            responses.push_back(std::move(response));
        }
    }
    return responses;
}

std::vector<AsyncRequestsSender::Response> sendCommandToShards(
    OperationContext* opCtx,
    StringData dbName,
    const BSONObj& command,
    const std::vector<ShardId>& shardIds,
    const std::shared_ptr<executor::TaskExecutor>& executor,
    const bool throwOnError) {
    std::vector<AsyncRequestsSender::Request> requests;
    for (const auto& shardId : shardIds) {
        requests.emplace_back(shardId, command);
    }

    return processShardResponses(opCtx, dbName, command, requests, executor, throwOnError);
}

std::vector<AsyncRequestsSender::Response> sendCommandToShardsWithVersion(
    OperationContext* opCtx,
    StringData dbName,
    const BSONObj& command,
    const std::vector<ShardId>& shardIds,
    const std::shared_ptr<executor::TaskExecutor>& executor,
    const CollectionRoutingInfo& cri,
    const bool throwOnError) {
    std::vector<AsyncRequestsSender::Request> requests;
    for (const auto& shardId : shardIds) {
        requests.emplace_back(shardId, appendShardVersion(command, cri.getShardVersion(shardId)));
    }
    return processShardResponses(opCtx, dbName, command, requests, executor, throwOnError);
}


// TODO SERVER-67593: Investigate if DBDirectClient can be used instead.
Status createIndexOnCollection(OperationContext* opCtx,
                               const NamespaceString& ns,
                               const BSONObj& keys,
                               bool unique) {
    try {
        // TODO SERVER-50983: Create abstraction for creating collection when using
        // AutoGetCollection
        AutoGetCollection autoColl(opCtx, ns, MODE_X);
        const Collection* collection = autoColl.getCollection().get();
        if (!collection) {
            CollectionOptions options;
            options.uuid = UUID::gen();
            writeConflictRetry(opCtx, "createIndexOnCollection", ns.ns(), [&] {
                WriteUnitOfWork wunit(opCtx);
                auto db = autoColl.ensureDbExists(opCtx);
                collection = db->createCollection(opCtx, ns, options);
                invariant(collection,
                          str::stream() << "Failed to create collection "
                                        << ns.toStringForErrorMsg() << " for indexes: " << keys);
                wunit.commit();
            });
        }
        auto indexCatalog = collection->getIndexCatalog();
        IndexSpec index;
        index.addKeys(keys);
        index.unique(unique);
        index.version(int(IndexDescriptor::kLatestIndexVersion));
        auto removeIndexBuildsToo = false;
        auto indexSpecs = indexCatalog->removeExistingIndexes(
            opCtx,
            CollectionPtr(collection),
            uassertStatusOK(
                collection->addCollationDefaultsToIndexSpecsForCreate(opCtx, {index.toBSON()})),
            removeIndexBuildsToo);

        if (indexSpecs.empty()) {
            return Status::OK();
        }

        auto fromMigrate = false;
        if (!collection->isEmpty(opCtx)) {
            // We typically create indexes on config/admin collections for sharding while setting up
            // a sharded cluster, so we do not expect to see data in the collection.
            // Therefore, it is ok to log this index build.
            const auto& indexSpec = indexSpecs[0];
            LOGV2(5173300,
                  "Creating index on sharding collection with existing data",
                  logAttrs(ns),
                  "uuid"_attr = collection->uuid(),
                  "index"_attr = indexSpec);
            auto indexConstraints = IndexBuildsManager::IndexConstraints::kEnforce;
            IndexBuildsCoordinator::get(opCtx)->createIndex(
                opCtx, collection->uuid(), indexSpec, indexConstraints, fromMigrate);
        } else {
            writeConflictRetry(opCtx, "createIndexOnConfigCollection", ns.ns(), [&] {
                WriteUnitOfWork wunit(opCtx);
                CollectionWriter collWriter(opCtx, collection->uuid());
                IndexBuildsCoordinator::get(opCtx)->createIndexesOnEmptyCollection(
                    opCtx, collWriter, indexSpecs, fromMigrate);
                wunit.commit();
            });
        }
    } catch (const DBException& e) {
        return e.toStatus();
    }

    return Status::OK();
}

Status createShardingIndexCatalogIndexes(OperationContext* opCtx,
                                         const NamespaceString& indexCatalogNamespace) {
    bool unique = true;
    auto result = createIndexOnCollection(opCtx,
                                          indexCatalogNamespace,
                                          BSON(IndexCatalogType::kCollectionUUIDFieldName
                                               << 1 << IndexCatalogType::kLastmodFieldName << 1),
                                          !unique);
    if (!result.isOK()) {
        return result.withContext(str::stream()
                                  << "couldn't create collectionUUID_1_lastmod_1 index on "
                                  << indexCatalogNamespace.toStringForErrorMsg());
    }
    result = createIndexOnCollection(opCtx,
                                     indexCatalogNamespace,
                                     BSON(IndexCatalogType::kCollectionUUIDFieldName
                                          << 1 << IndexCatalogType::kNameFieldName << 1),
                                     unique);
    if (!result.isOK()) {
        return result.withContext(str::stream()
                                  << "couldn't create collectionUUID_1_name_1 index on "
                                  << indexCatalogNamespace.toStringForErrorMsg());
    }
    return Status::OK();
}

Status createShardCollectionCatalogIndexes(OperationContext* opCtx) {
    bool unique = true;
    auto result =
        createIndexOnCollection(opCtx,
                                NamespaceString::kShardCollectionCatalogNamespace,
                                BSON(ShardAuthoritativeCollectionType::kUuidFieldName << 1),
                                !unique);
    if (!result.isOK()) {
        return result.withContext(str::stream() << "couldn't create uuid_1 index on "
                                                << NamespaceString::kShardCollectionCatalogNamespace
                                                       .toStringForErrorMsg());
    }

    return Status::OK();
}

}  // namespace sharding_util
}  // namespace mongo