NDB Cluster 8.4 的新特性What is New in NDB Cluster 8.4

NDB 8.x 创新版本的更改Changes in NDB 8.x Innovation Releases

ndbinfo transporter_details 表。 transporter_details 表提供有关 NDB 集群中使用的各个传输器的信息。它在其他方面类似于 ndbinfo transporters 表，该表提供汇总形式的此类信息。ndbinfo transporter_details table.  The transporter_details table provides information about individual transporters used in an NDB cluster. It is otherwise similar to the ndbinfo transporters table, which provides such information in aggregate form.

与 NDB 8.0 中引入的版本相比，NDB 8.4.0 提供了额外的列。这些新列及每列的简要说明如下：NDB 8.4.0 provides additional columns as compared to the version introduced in NDB 8.0. These new columns, along with brief dscriptions of each, are listed here:

NDB 8.4.1 添加了一个 type 列，用于显示传输器的连接类型（TCP 或 SHM）。NDB 8.4.1 adds a type column, which displays the transport's connection type (TCP or SHM).

请参阅第 25.6.15.65 节，“ndbinfo transporter_details 表”，了解更多信息。See Section 25.6.15.65, “The ndbinfo transporter_details Table”, for more information.

NDB 复制：未使用更新的过滤。 以前，当执行二进制日志记录的 SQL 节点使用 log_replica_updates=OFF 时，任何在副本 NDB 集群上应用的复制更新都会被发送到执行二进制日志记录的 SQL 节点。这些更新实际上并未被应用或用于其他用途；这会导致不必要的网络流量和资源消耗。NDB Replication: Filtering of unused updates.  Previously, when SQL nodes performing binary logging used log_replica_updates=OFF, any replicated updates which were applied on a replica NDB cluster were sent on to the SQL nodes performing binary logging. These updates were not actually applied or used for any other purpose; this entailed unnecessary network traffic and consumption of resources.

在 NDB 8.4.0 及更高版本中，应用于副本 SQL 节点的更新会在该节点被过滤掉，并且不再发送到任何其他 SQL 节点。不触发任何日志记录的更新也不再由副本发送。In NDB 8.4.0 and later, updates applied on the replica SQL node are filtered out on this node, and are no longer sent onward to any other SQL nodes. Updates that do not trigger any logging are also no longer sent by the replica.

每会话二进制日志事务缓存大小。 NDB 8.4.3 新增了 ndb_log_cache_size 服务器系统变量，可以设置用于写入二进制日志的事务缓存大小。这使得可以使用较大的缓存来记录 NDB 事务，并且（使用 binlog_cache_size）为记录其他事务使用较小的缓存，从而更有效地利用资源。Per-session binary log transaction cache sizing.  NDB 8.4.3 adds the ndb_log_cache_size server system variable, which makes it possible to set the size of the transaction cache used for writing the binary log. This enables use of a large cache for logging NDB transactions, and (using binlog_cache_size) a smaller cache for logging other transactions, thus making more efficient use of resources.

Ndb.cfg 文件弃用。 使用 Ndb.cfg 文件来设置 NDB 进程的连接字符串，文档支持不完善。自 NDB 8.4.3 起，该文件的使用已正式弃用；你应预期在未来的 MySQL Cluster 版本中对其支持将被移除。Ndb.cfg file deprecation.  Use of an Ndb.cfg file for setting the connection string for an NDB process was not well documented or supported. As of NDB 8.4.3, use of this file is now formally deprecated; you should expect support for it to be removed in a future release of MySQL Cluster.

微秒级节点日志时间戳。 在 NDB 8.4.6 及更高版本中，节点日志时间戳可以以微秒精度打印。数据节点可以通过数据节点 --ndb-log-timestamps=UTC 选项启用该功能；管理节点可以通过 ndb_mgmd 选项 --ndb-log-timestamps=UTC 实现。为保持向后兼容性，默认值是 LEGACY，使用系统时区并以秒为单位显示分辨率，与之前的版本相同。Microsecond node log timestamps.  Node log timestamps in NDB 8.4.6 and later can be printed with microsecond resolution. Data nodes can enable this feature using the data node --ndb-log-timestamps=UTC option; management nodes can do this using the ndb_mgmd option --ndb-log-timestamps=UTC. For backwards compatible behavior, the default is LEGACY, which uses the system time zone and resolution in seconds, as in previous releases.

SQL 节点可以使用大致等效的选项 --log-timestamps；您应注意，此 mysqld 选项不接受 LEGACY 作为值。SQL nodes can use the roughly equivalent option --log-timestamps; you should be aware that this mysqld option does not accept LEGACY as a value.

有关 NDB Cluster 节点日志的更多信息，请参见 第 25.6.2 节，“NDB Cluster 日志消息”。For more information about NDB Cluster node logs, see Section 25.6.2, “NDB Cluster Log Messages”.

用于集群节点通信的 TLS。 NDB Cluster 8.3 及更高版本提供对通过传输层安全性 (TLS) 和互联网公钥基础设施 (PKI) 认证和加密 NDB 节点之间以及 NDB 管理服务器与其客户端之间连接的网络通信的支持；TLS 应用于 NDB 传输协议和 NDB 管理协议。TLS for cluster node communications.  NDB Cluster 8.3 and later provides support for network communications secured by Transport Layer Security (TLS) and Internet Public Key Infrastructure (PKI) to authenticate and encrypt connections between NDB nodes, and between the NDB management server and its clients; TLS is applied to the NDB Transporter Protocol, and to the NDB Management Protocol.

此功能使用 TLS 双向认证，其中节点自身的证书文件包含节点用于验证其对等方证书的信任链。当集群启用 TLS 时，数据节点和管理节点使用 TLS 执行以下任务：This feature uses TLS mutual authentication, in which a node's own certificate file contains the chain of trust which the node uses to validate the certificates of its peers. When TLS is enabled on the cluster, data and management nodes use TLS to perform the following tasks:

使用 MySQL 客户端协议的连接采用 MySQL 用户认证，并可能使用 TLS（包括可选的双向 TLS），如本手册其他部分所述；详情请参见 第 8.3 节，“使用加密连接”。Connections that use the MySQL client protocol employ MySQL user authentication, and may use TLS (including optional mutual TLS) as described elsewhere in this Manual; see Section 8.3, “Using Encrypted Connections”, for more information.

NDB 实现了一个新工具 ndb_sign_keys，可用于创建和管理 CA、证书文件以及密钥。您可以使用给定的配置文件，通过 ndb_sign_keys --create-key 为集群中的所有节点生成一组密钥和证书。NDB implements a new tool ndb_sign_keys which can be used to create and manage CA, certificate files, and keys. You can generate a set of keys and certificates for all nodes in a cluster with a given configuration file using ndb_sign_keys --create-key.

使用 ndb_sign_keys，节点证书可以绑定到特定主机名、设置在指定日期过期，并与特定节点类型关联，从而使客户端与服务器区分开来，管理服务器与数据节点区分开来。（每个 NDB TLS 证书都可以用于 MGM 客户端连接。）私钥会在原地创建，从而将包含私钥的文件复制降到最低。私钥和证书都标记为“活动”或“待定”；ndb_sign_keys 还提供了密钥轮换的帮助，以允许待定密钥在活动密钥过期之前替换活动密钥。Using ndb_sign_keys, a node certificate can be bound to a particular hostname, made to expire on a given date, and be associated with a given node type, so that clients are distinct from servers, and management servers from data nodes. (Every NDB TLS certificate can be used for MGM client connections.) Private keys are created in place, so that copying of files containing private keys is minimized. Both private keys and certificates are labeled as either active or pending; ndb_sign_keys also provides help with rotating keys to allow for pending keys to replace active keys before the active keys expire.

可以使用系统终端通过 ndb_mgm 客户端并带有 --test-tls 来测试节点 TLS 连接，或者在 ndb_mgm 客户端中使用 TLS INFO 命令。您可以通过检查 ndbinfo certificates 表获取集群节点使用的证书信息。Testing of node TLS connections can be done from the system shell using ndb_mgm client with --test-tls, or within the ndb_mgm client using the TLS INFO command. You can obtain information about certificates used by cluster nodes by checking the ndbinfo certificates table.

要强制要求使用 TLS，请在每个集群主机的 my.cnf 中设置客户端选项 ndb-mgm-tls=strict，然后在集群 config.ini 文件的 [mgm default] 部分中设置 RequireTls=true，并且在配置文件的 [ndbd default] 部分也设置 RequireTls=true。然后执行集群的滚动重启，使用 --reload --config-file 重启管理服务器。To enforce a requirement for TLS, set the client option ndb-mgm-tls=strict in my.cnf on each cluster host, then set RequireTls=true in the [mgm default] section of the cluster config.ini file, and set RequireTls=true in the [ndbd default] section of the configuration file as well. Then perform a rolling restart of the cluster, restarting the management server with --reload --config-file.

在 NDB 8.3 及更高版本中，NDB Cluster API 应用程序也支持使用 TLS 连接。有关 MGM API 支持的信息，请参阅 TLS 函数。NDB API Ndb_cluster_connection 类添加了 configure_tls() 和 get_tls_certificate_path() 方法，用于客户端设置 TLS 连接。Use of TLS connections is also supported in NDB Cluster API applications in NDB 8.3 and later. For information about MGM API support, see TLS Functions. The NDB API Ndb_cluster_connection class adds configure_tls() get_tls_certificate_path() methods for setting up TLS connections by clients.

有关更多信息，请参见第 25.6.19.5 节，“NDB 集群的 TLS 链接加密”，以及第 25.5.28 节，“ndb_sign_keys — 为 NDB 集群创建、签名和管理 TLS 密钥和证书”。For more information, see Section 25.6.19.5, “TLS Link Encryption for NDB Cluster”, as well as Section 25.5.28, “ndb_sign_keys — Create, Sign, and Manage TLS Keys and Certificates for NDB Cluster”.

二进制日志注入器内存分配。 在以前版本的 NDB Cluster 中，当 NDB 二进制日志注入器处理模式更改并跟踪二进制日志状态时，用于这些目的的内存分配的 arena 选择是通过更改线程本地指针强制执行的，从而试图捕获在 epoch 处理期间执行的所有分配。在 epoch 结束时，这些指针被重置，arena 内存被释放，arena 结构被丢弃；这释放了内存，但也需要为下一个 epoch 再次设置它。线程本地指针的更改还引入了在不同子系统中激活功能时内存分配错误的风险。MySQL NDB Cluster 8.3 对此功能进行了以下改进：Binary log injector memory allocation.  In previous versions of NDB Cluster, when the NDB binary log injector was engaged in handling schema changes and tracking the state of the binary log, the choice of arena for allocation of memory for these purposes was forced by changing thread local pointers, thus attempting to try and catch all allocations performed during epoch processing. At the end of the epoch, those pointers were reset, arena memory was released, and the arena structures discarded; this released the memory, but also required setting it up again for the next epoch. The thread local pointer changes also introduced the risk of memory being allocated wrongly when activating functionality in different subsystems. MySQL NDB Cluster 8.3 makes the following improvements to this functionality:

这些更改仅限内部使用，但通过节省内存释放和连续纪元的重新分配，应该会带来显著的改进。These changes are internal only but should provide a noticeable improvememnt by saving on memory release and re-allocation over successive epochs.

NDB API 主键更新。 以前，当尝试使用除 NdbRecord 之外的任何其他机制更新主键值时，NDB API 会返回错误 4202 不允许在元组键属性上设置值，即使设置的值与现有值相同。在 NDB 8.1 及更高版本中，执行其他方式的更新时的检查交由数据节点处理，就像使用 NdbRecord 执行更新时一样。NDB API primary key updates.  Previously, when using any other mechanism than NdbRecord in an attempt to update a primary key value, the NDB API returned error 4202 Set value on tuple key attribute is not allowed, even setting a value identical to the existing one. In NDB 8.1 and later, checking when performing updates by other means is handed off to the data nodes, as it was already when using NdbRecord to perform the update.

这意味着您现在可以使用 NdbOperation::setValue()、NdbInterpretedCode::write_attr()以及其他设置列值的 NdbOperation 和 NdbInterpretedCode 方法来执行主键更新（包括 NdbOperation 方法 incValue()、subValue()，以及 NdbInterpretedCode 方法 add_val()、sub_val() 等等）。这同样适用于 NdbOperation 接口的 OperationOptions::OO_SETVALUE 扩展功能。This means that you can now perform primary key updates using NdbOperation::setValue(), NdbInterpretedCode::write_attr(), and other methods of NdbOperation and NdbInterpretedCode which set column values (including the NdbOperation methods incValue(), subValue(), the NdbInterpretedCode methods add_val(), sub_val(), and so on). This also applies to the NdbOperation interface's OperationOptions::OO_SETVALUE extension.

改进的警告。 对警告输出进行了以下改进：Improved warnings.  Made the following improvements in warning output:

此外，当 log_error_verbosity 大于或等于 3（INFO 级别）时，我们不再每隔 ndb_metadata_check_interval（默认 60）秒将消息 Validating excluded objects 打印到 SQL 节点的错误日志，因为此类消息往往会淹没错误日志，使其难以检查，并占用过多磁盘空间，同时对用户没有任何额外的好处。In addition, we no longer print the message Validating excluded objects to the SQL node's error log every ndb_metadata_check_interval (default 60) seconds when log_error_verbosity is greater than or equal to 3 (INFO level), due ot the fact that such messages tended to flood the error log, making it difficult to examine, and using excess disk space, while not providing any additional benefit to the user.

现在，对具有非常大且可能重叠的 IN() 和 NOT IN() 列表的查询之间的下推连接进行处理，方式既正确又安全。Pushdown joins between queries featuring very large and possibly overlapping IN() and NOT IN() lists are now handled in a correct and safe manner.

ndbcluster 插件日志消息现在使用 SYSTEM 作为日志级别，并使用 NDB 作为日志子系统。这意味着 ndbcluster 插件的信息性消息总是会打印；可以通过使用 --ndb_extra_logging 来控制其详细程度。ndbcluster plugin log messages now use SYSTEM as the log level and NDB as the subsystem for logging. This means that informational messages from the ndbcluster plugin are always printed; their verbosity can be controlled by using --ndb_extra_logging.