OpenSearch:Configure a PolarDB for MySQL data source

PolarDB for MySQL is an online database service developed by Alibaba Cloud that is stable, reliable, and scalable, and provides out-of-box capabilities. For more information, see What is PolarDB for MySQL Enterprise Edition?

Before you purchase a PolarDB for MySQL cluster, take note of the following items:

• OpenSearch supports only PolarDB for MySQL 5.6, 5.7, and 8.0.

• You must use the Alibaba Cloud account that you use to log on to the OpenSearch console to create the PolarDB for MySQL cluster. Otherwise, OpenSearch cannot access this cluster.

• The PolarDB for MySQL cluster must reside in the same region as the OpenSearch application for which you want to configure a data source.

• By default, the binary logging feature is disabled after the PolarDB for MySQL cluster is created. In this case, you may fail to add a database in the cluster as a data source. Therefore, the binary logging feature must be enabled. To enable the binary logging feature, set the loose_polar_log_bin parameter whose default value is OFF to ON_WITH_GTID and the binlog_row_image parameter to FULL. By default, the binlog_row_image parameter is set to FULL.

• The PolarDB for MySQL cluster supports cloned instances.

• The PolarDB for MySQL cluster supports read/write splitting.

Supported features

• All the data in the specified database or table can be manually pulled or automatically pulled at the specified time.

• You can merge rows of multiple tables from one or more data sources. To merge the rows, make sure that the source tables use the same schema and that the configurations of the data processing plug-ins are the same in the data sources. In addition, the primary key values cannot be duplicated. Data with duplicate primary key values can be overwritten. This feature can be used in the following scenarios:

  • Data is synchronized from multiple source tables in a data source to a table in an application.

  • Multiple data sources are configured for a table in an application. Data is synchronized from one or more source tables in each data source to the table.

• You can use plug-ins to convert the values of source fields.

• You can specify filter conditions to filter full or incremental data.

• You can use an asterisk (*) to match tables whose names have the same prefix.

Limits

• The binlog_row_image parameter must be set to FULL for a PolarDB for MySQL cluster. To enable the binary logging feature, set the loose_polar_log_bin parameter whose default value is OFF to ON_WITH_GTID and the binlog_row_image parameter to FULL. By default, the binlog_row_image parameter is set to FULL.

• Only PolarDB for MySQL 5.6, 5.7, and 8.0 are supported.

• You must use the Alibaba Cloud account that you use to log on to the OpenSearch console to create the PolarDB for MySQL cluster.

• The PolarDB for MySQL cluster must reside in the same region as the OpenSearch application for which you want to configure a data source.

• After you configure a PolarDB for MySQL data source for a standard application, you cannot use OpenSearch SDKs or call an API operation to push incremental data to the application.

• If you configure a PolarDB for MySQL data source for a standard application, filter conditions cannot be specified to filter the data of the data source.

• REPLACE INTO statements are not supported.

• You cannot use TRUNCATE or DROP statements to delete data. To delete data, use DELETE statements.

• The password that is used to access the PolarDB for MySQL data source cannot contain the percent sign (%). Otherwise, the reindexing fails.

• You cannot concatenate field values of source tables that use different schemas.

• We recommend that you set both the loose_max_statement_time and connect_timeout parameters to 0. You can change the values based on your business requirements after reindexing or offline changes are performed and full data is synchronized.

Usage notes

• If an ApsaraDB RDS for MySQL or PolarDB data source is attached to a PolarDB-X instance, you must specify the name of the database shard that is actually used on the PolarDB-X instance when you configure the data source. After database sharding, a database on the PolarDB-X instance has one shadow database and eight database shards. Data is randomly written to the database shards.

• You can switch between the internal and public endpoints of a PolarDB for MySQL cluster. No traffic fee is generated when OpenSearch accesses the data of the PolarDB for MySQL cluster.

• OpenSearch allows you to pull full data only from the primary database. We recommend that you import full data and rebuild indexes during off-peak hours.

• If a source table from the PolarDB for MySQL cluster contains fields of the DATETIME or TIMESTAMP format, the values of these fields are converted to the number of milliseconds. You need to set the type of destination fields in your application to TIMESTAMP.

• After full data is synchronized, the documents that do not meet the filter conditions are filtered out. If specific documents have the same primary key values as those in the destination table of an application, these documents are also filtered out.

• If no incremental data is generated at the data source for 15 days or more, data synchronization may fail. In this case, you can manually perform offline changes. For more information, see Perform offline changes.

• If SSL encryption is enabled for PolarDB for MySQL, make sure that SSL certificates are valid. If the SSL certificates expire, a connection error may occur. You must update the validity periods of the SSL certificates in a timely manner.

• When you use OpenSearch to synchronize data from a PolarDB for MySQL cluster, you must add the IP address and CIDR block of the server in the region in which OpenSearch resides to the IP address whitelist of the PolarDB for MySQL cluster. The following table provides the IP address whitelists of different regions.

  Region	IP address
  China (Hangzhou)	100.104.190.128/26,100.104.241.128/26
  China (Beijing)	100.104.16.192/26,100.104.179.0/26
  China (Shanghai)	100.104.37.0/26,100.104.46.0/26
  China (Shenzhen)	100.104.87.192/26,1100.104.132.192/26
  China (Qingdao)	100.104.240.128/26,1100.104.111.128/26
  China (Zhangjiakou)	100.104.155.192/26,100.104.238.64/26
  Germany (Frankfurt)	100.104.127.0/26,100.104.35.192/26
  US (Silicon Valley) and US (Virginia)	100.104.193.128/26,100.104.119.128/26
  Singapore	100.104.58.192/26,100.104.74.192/26

Account authorization

• To connect to a PolarDB for MySQL cluster, you must obtain the permissions required to access the cluster. Exercise caution when you enter a specific account and password to connect to the data source for the first time.

• Make sure that your account has the permissions to query all the tables in the source database in Data Transmission Service (DTS). In addition, make sure that the SHOW CREATE TABLE *. * statement can be executed as expected. If your account does not have the permissions, real-time synchronization may fail.

• Make sure that the account and permissions are not changed unless necessary. Account changes may lead to failed real-time synchronization. In addition, if changes are made to the account, a new version of the application may fail to be created. If the account and password are changed, you must remove the existing data source and add the destination database as a new data source.

Troubleshooting

• After a PolarDB for MySQL data source is configured, reindexing may be suspended. In this case, create a test table in the database in which source tables reside. Then, write one or two data entries to or update one or two data entries in the test table every minute. This ensures that continuous binary logs are generated when data is reindexed.

• If an advanced application for which a PolarDB for MySQL database source is configured has overdue payments, you can manually trigger reindexing after you pay the overdue payments.

• The password that is used to access the PolarDB for MySQL cluster cannot contain the percent sign (%). Otherwise, the reindexing fails and the error message "Illegal hex characters in escape (%) pattern" appears.

• The primary key value of a table in an application must be unique. If data is synchronized from multiple tables with the same table schema, duplicated primary key values may exist. In this case, data may be overwritten. To prevent this issue, you can use the StringCatenateExtractor plug-in to concatenate the values of multiple fields. The source fields are pk and $table. Replace pk with the primary key field of a table in the PolarDB for MySQL cluster. $table is the default system variable, which indicates the table name. Connect the source fields by using a hyphen (-) or a custom character.

For example, a table in the PolarDB for MySQL cluster is named my_table_0 and the primary key value is 123456. In this case, the new primary key value obtained after concatenation is 123456-my_table_0.

• The fields of the DATE or DATETIME type in a source table are filtered. For example, if the name of a source field in the database table is createtime, the time format that is specified in the filter conditions to filter the data of the data source must be createtime>'2018-03-01 00:00:00'. If the value of the field is in the createtime>'2018-3-1 00:00:00' format, an error is reported.

Configure a PolarDB for MySQL data source

• You can configure a PolarDB for MySQL data source when you configure a new application.

• If you want to configure a PolarDB for MySQL data source for an existing application, you can go to the details page of the application and perform offline changes.

Procedure and usage notes

1. In the Select Data Source panel, select PolarDB and click Connect to Database.

2. Configure the parameters of the PolarDB for MySQL data source, and click Connect.

Then, OpenSearch attempts to connect to the data source and returns an error message if the connection fails.

3. After the data source is connected, select a table. The following figure shows the page that is displayed after the connection is established. After you select a table, click OK.

• Select or enter the name of the table to be accessed. The name is case-sensitive.

• You can use an asterisk (*) to match multiple tables with the same table schema in a database. The names of these tables have the same prefix. For example, you can enter table_* to match tables such as table_a and table_b.

4. After the connection is successful, configure field mappings. OpenSearch automatically synchronizes source fields from the table. For more information about data processing plug-ins, see Data processing plug-ins.

5. Specify filter conditions to filter the source data. You cannot specify filter conditions for a standard application. After you complete the configurations in the Data Source step, click Submit. The application schema is configured.

• You can configure multiple data sources for a table in OpenSearch, provided that the schemas and configurations of the source tables are the same.

• Only data entries that meet the filter conditions can be pulled from the PolarDB for MySQL data source. For more information, see Configure conditions to filter fields in ApsaraDB RDS and PolarDB data sources.