elasticsearch index 之 create index（二）
创建索引需要创建索引并且更新集群index matedata，这一过程在MetaDataCreateIndexService的createIndex方法中完成。这里会提交一个高优先级，AckedClusterStateUpdateTask类型的task。索引创建需要即时得到反馈，异常这个task需要返回，会超时，而且这个任务的优先级也非常高。下面具体看一下它的execute方法，这个方法会在master执行任务时调用，这个方法非常长，主要完成以下三个功能：更新合并request，template中的mapping和setting，调用indiceService创建索引，对创建后的索引添加mapping。这一系列功能完成后，合并完成后生成新的matedata，并更新集群状态，完成了索引的创建。具体的调用方法参考上一篇。代码如下所示：
@Override public ClusterState execute(ClusterState currentState) throws Exception { boolean indexCreated = false; String removalReason = null; try {
//检查request的合法性，1.5版本主要检查index名字是否合法，如不能含有某些字符，另外就是集群节点版本 validate(request, currentState); for (Alias alias : request.aliases()) {//检查别称是否合法 aliasValidator.validateAlias(alias, request.index(), currentState.metaData()); } // 查找索引模板 List<IndexTemplateMetaData> templates = findTemplates(request, currentState, indexTemplateFilter); Map<String, Custom> customs = Maps.newHashMap(); // add the request mapping Map<String, Map<String, Object>> mappings = Maps.newHashMap(); Map<String, AliasMetaData> templatesAliases = Maps.newHashMap(); List<String> templateNames = Lists.newArrayList();
//取出request中的mapping配置，虽然mapping可以后面添加，多数情况创建索引的时候还是会附带着mapping，在request中mapping是一个map for (Map.Entry<String, String> entry : request.mappings().entrySet()) { mappings.put(entry.getKey(), parseMapping(entry.getValue())); } //一些预设如warm等 for (Map.Entry<String, Custom> entry : request.customs().entrySet()) { customs.put(entry.getKey(), entry.getValue()); } // 将找到的template和request中的mapping合并 for (IndexTemplateMetaData template : templates) { templateNames.add(template.getName()); for (ObjectObjectCursor<String, CompressedString> cursor : template.mappings()) { if (mappings.containsKey(cursor.key)) { XContentHelper.mergeDefaults(mappings.get(cursor.key), parseMapping(cursor.value.string())); } else { mappings.put(cursor.key, parseMapping(cursor.value.string())); } } // 合并custom for (ObjectObjectCursor<String, Custom> cursor : template.customs()) { String type = cursor.key; IndexMetaData.Custom custom = cursor.value; IndexMetaData.Custom existing = customs.get(type); if (existing == null) { customs.put(type, custom); } else { IndexMetaData.Custom merged = IndexMetaData.lookupFactorySafe(type).merge(existing, custom); customs.put(type, merged); } } //处理合并别名 for (ObjectObjectCursor<String, AliasMetaData> cursor : template.aliases()) { AliasMetaData aliasMetaData = cursor.value; //if an alias with same name came with the create index request itself, // ignore this one taken from the index template if (request.aliases().contains(new Alias(aliasMetaData.alias()))) { continue; } //if an alias with same name was already processed, ignore this one if (templatesAliases.containsKey(cursor.key)) { continue; } //Allow templatesAliases to be templated by replacing a token with the name of the index that we are applying it to if (aliasMetaData.alias().contains("{index}")) { String templatedAlias = aliasMetaData.alias().replace("{index}", request.index()); aliasMetaData = AliasMetaData.newAliasMetaData(aliasMetaData, templatedAlias); } aliasValidator.validateAliasMetaData(aliasMetaData, request.index(), currentState.metaData()); templatesAliases.put(aliasMetaData.alias(), aliasMetaData); } } // 合并完template和request，现在开始处理配置基本的mapping，合并逻辑跟之前相同，只是mapping来源不同 File mappingsDir = new File(environment.configFile(), "mappings"); if (mappingsDir.isDirectory()) { // first index level File indexMappingsDir = new File(mappingsDir, request.index()); if (indexMappingsDir.isDirectory()) { addMappings(mappings, indexMappingsDir); } // second is the _default mapping File defaultMappingsDir = new File(mappingsDir, "_default"); if (defaultMappingsDir.isDirectory()) { addMappings(mappings, defaultMappingsDir); } } //处理index的配置（setting） ImmutableSettings.Builder indexSettingsBuilder = settingsBuilder(); //加入模板中的setting for (int i = templates.size() - 1; i >= 0; i--) { indexSettingsBuilder.put(templates.get(i).settings()); } // 加入request中的mapping，request中设置会覆盖模板中的设置 indexSettingsBuilder.put(request.settings());
//处理shard，shard数量不能小于1，因此这里需要特殊处理，如果没有则要使用默认值 if (request.index().equals(ScriptService.SCRIPT_INDEX)) { indexSettingsBuilder.put(SETTING_NUMBER_OF_SHARDS, settings.getAsInt(SETTING_NUMBER_OF_SHARDS, 1)); } else { if (indexSettingsBuilder.get(SETTING_NUMBER_OF_SHARDS) == null) { if (request.index().equals(riverIndexName)) { indexSettingsBuilder.put(SETTING_NUMBER_OF_SHARDS, settings.getAsInt(SETTING_NUMBER_OF_SHARDS, 1)); } else { indexSettingsBuilder.put(SETTING_NUMBER_OF_SHARDS, settings.getAsInt(SETTING_NUMBER_OF_SHARDS, 5)); } } } if (request.index().equals(ScriptService.SCRIPT_INDEX)) { indexSettingsBuilder.put(SETTING_NUMBER_OF_REPLICAS, settings.getAsInt(SETTING_NUMBER_OF_REPLICAS, 0)); indexSettingsBuilder.put(SETTING_AUTO_EXPAND_REPLICAS, "0-all"); } else { if (indexSettingsBuilder.get(SETTING_NUMBER_OF_REPLICAS) == null) { if (request.index().equals(riverIndexName)) { indexSettingsBuilder.put(SETTING_NUMBER_OF_REPLICAS, settings.getAsInt(SETTING_NUMBER_OF_REPLICAS, 1)); } else { indexSettingsBuilder.put(SETTING_NUMBER_OF_REPLICAS, settings.getAsInt(SETTING_NUMBER_OF_REPLICAS, 1)); } } } //处理副本 if (settings.get(SETTING_AUTO_EXPAND_REPLICAS) != null && indexSettingsBuilder.get(SETTING_AUTO_EXPAND_REPLICAS) == null) { indexSettingsBuilder.put(SETTING_AUTO_EXPAND_REPLICAS, settings.get(SETTING_AUTO_EXPAND_REPLICAS)); } if (indexSettingsBuilder.get(SETTING_VERSION_CREATED) == null) { DiscoveryNodes nodes = currentState.nodes(); final Version createdVersion = Version.smallest(version, nodes.smallestNonClientNodeVersion()); indexSettingsBuilder.put(SETTING_VERSION_CREATED, createdVersion); } if (indexSettingsBuilder.get(SETTING_CREATION_DATE) == null) { indexSettingsBuilder.put(SETTING_CREATION_DATE, System.currentTimeMillis()); } indexSettingsBuilder.put(SETTING_UUID, Strings.randomBase64UUID()); //创建setting Settings actualIndexSettings = indexSettingsBuilder.build();
// 通过indiceservice创建索引 indicesService.createIndex(request.index(), actualIndexSettings, clusterService.localNode().id()); indexCreated = true; //如果创建成功这里就可以获取到对应的indexservice，否则会抛出异常 IndexService indexService = indicesService.indexServiceSafe(request.index());
//获取mappingService试图放置mapping MapperService mapperService = indexService.mapperService(); // 为索引添加mapping，首先是默认mapping if (mappings.containsKey(MapperService.DEFAULT_MAPPING)) { try { mapperService.merge(MapperService.DEFAULT_MAPPING, new CompressedString(XContentFactory.jsonBuilder().map(mappings.get(MapperService.DEFAULT_MAPPING)).string()), false); } catch (Exception e) { removalReason = "failed on parsing default mapping on index creation"; throw new MapperParsingException("mapping [" + MapperService.DEFAULT_MAPPING + "]", e); } } for (Map.Entry<String, Map<String, Object>> entry : mappings.entrySet()) { if (entry.getKey().equals(MapperService.DEFAULT_MAPPING)) { continue; } try { // apply the default here, its the first time we parse it mapperService.merge(entry.getKey(), new CompressedString(XContentFactory.jsonBuilder().map(entry.getValue()).string()), true); } catch (Exception e) { removalReason = "failed on parsing mappings on index creation"; throw new MapperParsingException("mapping [" + entry.getKey() + "]", e); } } //添加request中的别称 IndexQueryParserService indexQueryParserService = indexService.queryParserService(); for (Alias alias : request.aliases()) { if (Strings.hasLength(alias.filter())) { aliasValidator.validateAliasFilter(alias.name(), alias.filter(), indexQueryParserService); } } for (AliasMetaData aliasMetaData : templatesAliases.values()) { if (aliasMetaData.filter() != null) { aliasValidator.validateAliasFilter(aliasMetaData.alias(), aliasMetaData.filter().uncompressed(), indexQueryParserService); } } // 以下更新Index的matedata， Map<String, MappingMetaData> mappingsMetaData = Maps.newHashMap(); for (DocumentMapper mapper : mapperService.docMappers(true)) { MappingMetaData mappingMd = new MappingMetaData(mapper); mappingsMetaData.put(mapper.type(), mappingMd); } final IndexMetaData.Builder indexMetaDataBuilder = IndexMetaData.builder(request.index()).settings(actualIndexSettings); for (MappingMetaData mappingMd : mappingsMetaData.values()) { indexMetaDataBuilder.putMapping(mappingMd); } for (AliasMetaData aliasMetaData : templatesAliases.values()) { indexMetaDataBuilder.putAlias(aliasMetaData); } for (Alias alias : request.aliases()) { AliasMetaData aliasMetaData = AliasMetaData.builder(alias.name()).filter(alias.filter()) .indexRouting(alias.indexRouting()).searchRouting(alias.searchRouting()).build(); indexMetaDataBuilder.putAlias(aliasMetaData); } for (Map.Entry<String, Custom> customEntry : customs.entrySet()) { indexMetaDataBuilder.putCustom(customEntry.getKey(), customEntry.getValue()); } indexMetaDataBuilder.state(request.state()); //matedata更新完毕，build新的matedata final IndexMetaData indexMetaData; try { indexMetaData = indexMetaDataBuilder.build(); } catch (Exception e) { removalReason = "failed to build index metadata"; throw e; } indexService.indicesLifecycle().beforeIndexAddedToCluster(new Index(request.index()), indexMetaData.settings()); //更新集群的matedata，将新build的indexmatadata加入到metadata中 MetaData newMetaData = MetaData.builder(currentState.metaData()) .put(indexMetaData, false) .build(); logger.info("[{}] creating index, cause [{}], templates {}, shards [{}]/[{}], mappings {}", request.index(), request.cause(), templateNames, indexMetaData.numberOfShards(), indexMetaData.numberOfReplicas(), mappings.keySet()); //阻塞集群，更新matadata ClusterBlocks.Builder blocks = ClusterBlocks.builder().blocks(currentState.blocks()); if (!request.blocks().isEmpty()) { for (ClusterBlock block : request.blocks()) { blocks.addIndexBlock(request.index(), block); } } if (request.state() == State.CLOSE) { blocks.addIndexBlock(request.index(), MetaDataIndexStateService.INDEX_CLOSED_BLOCK); } ClusterState updatedState = ClusterState.builder(currentState).blocks(blocks).metaData(newMetaData).build(); if (request.state() == State.OPEN) { RoutingTable.Builder routingTableBuilder = RoutingTable.builder(updatedState.routingTable()) .addAsNew(updatedState.metaData().index(request.index())); RoutingAllocation.Result routingResult = allocationService.reroute(ClusterState.builder(updatedState).routingTable(routingTableBuilder).build()); updatedState = ClusterState.builder(updatedState).routingResult(routingResult).build(); } removalReason = "cleaning up after validating index on master"; return updatedState; } finally { if (indexCreated) { // Index was already partially created - need to clean up indicesService.removeIndex(request.index(), removalReason != null ? removalReason : "failed to create index"); } } } }); }
以上就是创建index的create方法，方法中主要进行了两个动作：合并更新index的matadata和创建index。更新合并matadata的过程都在上面的代码中体现了。创建索引是调用indiceSerivice构建一个guice的injector，这个injector包含了Index的所有功能（如分词，相似度等）。同时会将其存储到indiceService中，以一个map的格式存储Map<String, Tuple<IndexService, Injector>> indices。运行中的集群每次对某个索引的操作都首先从indice中获取对应的IndexService。这一部分代码如下所示：
public synchronized IndexService createIndex(String sIndexName, @IndexSettings Settings settings, String localNodeId) throws ElasticsearchException { if (!lifecycle.started()) { throw new ElasticsearchIllegalStateException("Can't create an index [" + sIndexName + "], node is closed"); } Index index = new Index(sIndexName);
//检测index是否已经存在 if (indices.containsKey(index.name())) { throw new IndexAlreadyExistsException(index); } indicesLifecycle.beforeIndexCreated(index, settings); logger.debug("creating Index [{}], shards [{}]/[{}]", sIndexName, settings.get(SETTING_NUMBER_OF_SHARDS), settings.get(SETTING_NUMBER_OF_REPLICAS)); Settings indexSettings = settingsBuilder() .put(this.settings) .put(settings) .classLoader(settings.getClassLoader()) .build(); //构建index对应的injector ModulesBuilder modules = new ModulesBuilder(); modules.add(new IndexNameModule(index)); modules.add(new LocalNodeIdModule(localNodeId)); modules.add(new IndexSettingsModule(index, indexSettings)); modules.add(new IndexPluginsModule(indexSettings, pluginsService)); modules.add(new IndexStoreModule(indexSettings)); modules.add(new IndexEngineModule(indexSettings)); modules.add(new AnalysisModule(indexSettings, indicesAnalysisService)); modules.add(new SimilarityModule(indexSettings)); modules.add(new IndexCacheModule(indexSettings)); modules.add(new IndexFieldDataModule(indexSettings)); modules.add(new CodecModule(indexSettings)); modules.add(new MapperServiceModule()); modules.add(new IndexQueryParserModule(indexSettings)); modules.add(new IndexAliasesServiceModule()); modules.add(new IndexGatewayModule(indexSettings, injector.getInstance(Gateway.class))); modules.add(new IndexModule(indexSettings)); Injector indexInjector; try { indexInjector = modules.createChildInjector(injector); } catch (CreationException e) { throw new IndexCreationException(index, Injectors.getFirstErrorFailure(e)); } catch (Throwable e) { throw new IndexCreationException(index, e); } IndexService indexService = indexInjector.getInstance(IndexService.class); indicesLifecycle.afterIndexCreated(indexService); //将Indexservice和IndexInjector加入到indice map中 indices = newMapBuilder(indices).put(index.name(), new Tuple<>(indexService, indexInjector)).immutableMap(); return indexService; }
以上方法就是具体创建索引的过程，它是在master上操作的，同时它是同步方法。这样才能保证集群的Index创建一致性，因此这也会导致之前所说的大量创建创建索引时候的速度瓶颈。但是创建大量索引的动作是不常见的，需要尽量避免。创建一个索引对于一个集群来说就是开启对于该索引的各种操作，因此这里通过guice将索引的各个功能模块注入，并获得index操作的接口类Indexservice。如果这个方法执行成功，则可以合并template及request中的mapping，并且向刚创建的索引添加合并后的mapping，最后构建新的matadata，并将集群新的matadata发送给各个节点完成索引创建。
总结：索引创建的过程包括三步：更新集群matadata，调用indiceService中创建索引，向新创建的索引中放置（合并到Index对应的mappingService中）mapping。这三步都在以上的两个方法中。完成这三步，集群中就保存了新索引的信息，同时索引配置和mapping放置也完成。索引就可以正常使用。
