OpenSearch:Application management

1. Instance ID: A unique identifier for the instance.

2. Instance creation time: The time the instance was created, in YYYY-MM-DD HH:MM:SS format.

3. Billing method: The billing method for OpenSearch, which can be subscription or pay-as-you-go.

4. Region: The region where the application is deployed, such as China (Hangzhou), China (Shanghai), China (Beijing), China (Qingdao), and China (Zhangjiakou).

5. Instance type: The type of the OpenSearch instance, such as Advanced or Standard.

6. Preferred cluster specifications: The instance specifications tailored for different use cases, such as General-purpose Shared, Compute-optimized Shared, General-purpose Dedicated, and Compute-optimized Dedicated.

7. Storage resource: The total size of all documents across all tables in the application. This size is calculated based on the content of the fields, which are treated as strings, and does not include the field names.

8. Computing resource: A logical computing unit (LCU) is a unit for measuring search computing power. One LCU represents the computing power of 10 millicores in a search cluster. A millicore is a unit of CPU resources, equivalent to 1/1000th of a core.

9. Instance description: A brief, customizable description of the instance.

When you first configure an application and complete the initial reindexing, you have a single online application. If you later perform an operation that triggers reindexing, such as clearing data or making an offline change, OpenSearch creates an offline application to ensure the stability of your production services. All changes are made to the offline application. After the offline application successfully builds its index, you can run tests on it. Once it passes testing, you can promote it to the new online application. This process ensures that any changes to your application instance are seamless and do not affect your users.

The Normal status indicates that the application version is online and serving traffic.

Other statuses for applications that are not in service include:

1) Pending: The application is not configured.

2) Initializing Version: The application version is being initialized.

3) Initializing Data: The data is being initialized.

4) Invalid: An application becomes invalid if its full data storage exceeds the quota, or if a data source table or field does not exist. An invalid application must be deleted and reconfigured.

5) Expired: For a subscription instance, the service is suspended after the subscription period ends. The instance is retained in the console for 7 days, during which you can only renew the service. Other features are unavailable.

6) Frozen: The application has overdue payments.

This section includes information about the current application version, such as its billing method, preferred cluster specifications, storage resource, and computing resource, as well as the instance ID, region, and application description.

Note: For more information about the billing rules for your application, see the Billing methods and pricing documentation.

An API endpoint, also known as an endpoint, varies by region. When you use an API or SDK to access OpenSearch, you must use the correct API endpoint for your application's region. Each region provides different endpoints for different network environments. Use the internal endpoint for access from ECS instances within the same region (including within the same VPC), and use the public endpoint for access over the internet. Choose the appropriate endpoint based on your deployment. Before use, ping the endpoint to ensure it is accessible.

API endpoints include two types of domains. The public endpoint is http://opensearch-cn-hangzhou.aliyuncs.com, which supports HTTPS and is used for access from the public internet. The internal endpoint is http://intranet.opensearch-cn-hangzhou.aliyuncs.com, which also supports HTTPS and is used for access from within the same region, including from a VPC.

This setting specifies the source of your data in OpenSearch. Supported data sources include MaxCompute, ApsaraDB RDS (version 8.0 is supported, while versions earlier than 5.2 and later than 8.0 are not), PolarDB, and pushing data via the API or an SDK.

1. Table: Represents the application's table structure (schema).

2. Index fields: The fields configured as indexes in the application. For example, you can use an index field named 'id' in a query.

   Example: query=id:'cloud search'. For more information on index analysis methods, see here.

3. Attribute fields: The fields configured as attributes in the application.

   Usage: Use attribute fields in filter, aggregate, sort, and distinct clauses to perform filtering, aggregation, and sorting. For example: filter=id>100000.

For each search request, you can specify the default display fields to limit which fields are returned from the application schema.

You can use CloudMonitor to view key metrics for the current application version, such as storage capacity (bytes), storage capacity usage (%), total number of documents, queries per second (QPS), throttled queries per second (QPS), query duration (ms), computing resource (LCU), computing resources consumed per query (LCU), and daily peak computing resource (LCU). For more information about these metrics and how to configure alerts, see the documentation.

OpenSearch keeps a copy of all data you upload or synchronize from a data source. You must reindex when you modify the application schema or need to import a full dataset. Two reindexing methods are available: 1) Manual reindexing, which is typically used when you modify the application schema or perform a full data import. 2) Scheduled reindexing, which is commonly used for daily full imports from data sources like MaxCompute. Data synchronization is enabled by default for ApsaraDB RDS data sources, so you do not need to configure scheduled reindexing tasks.

You can change the billing method from pay-as-you-go to subscription. For information about billing methods, pricing, and the procedure, see the documentation.

Error logs display error messages for failed reindexing operations and failed data processing from API or SDK pushes. Note that data pushes via the API or an SDK are asynchronous. A successful API response does not guarantee that the data was processed successfully. If data processing fails, an error message appears in the console. For information on what error codes mean, see the documentation.

Online application: The version of the OpenSearch application that is currently serving your search traffic.

Offline application: When you need to make configuration changes to an application, such as modifying the application schema, OpenSearch creates a new version of the current online application. This new version is the offline application, and it starts with the same configuration as the online application. An online application supports offline changes (to the data source, application schema, and index structure), as well as changes to query configurations, sort configurations, and other extended features. However, you cannot create another version from an offline application to make further changes. You must apply all configurations directly to the current offline application. Therefore, online and offline applications have an inheritance relationship but are independent of each other.

Unpublished application: When you publish an offline application, it becomes the new online application, and the previous online application becomes an unpublished application. At this point, the former online application, now unpublished, cannot be published again unless you initiate another offline change on the current online application. Its status is "Published".

To prevent accidental deletion, you must enter the correct application name followed by "online application" to confirm the deletion.

To prevent data loss from accidental deletion, OpenSearch prevents you from deleting an entire application while an online application exists. If you are certain that the application is no longer needed, follow these steps:

1. Delete the offline application: On the Application Management page, go to the Application Details page, select the offline application, and then click Delete Application.

   Important

   If no offline application exists, you can skip this step.

2. Delete the online application: On the Application Management page, go to the Application Details page, select the online application, and then click Delete Application.

3. When the application is in the Pending state, you can delete it.

   Note: You can only delete pay-as-you-go instances. You cannot delete subscription instances.

When you need to modify the application schema (add or delete fields, modify or add indexes, add or delete attributes) or change the data source, you must perform an offline change. For details, Perform offline changes.