Edit Table

Edit a table

1. Log in to the OpenSearch console. On the instance details page, click Table Management in the left-side navigation pane.

2. Find the table you want to modify, then click Edit in the Actions column.

3. In the Basic Table Information step, set the number of data shards.

   Note

   The shard count must be a positive integer and can be up to 256. Set this value to no more than three times the number of Searcher workers.

4. Review the data source, then click Next.

   The behavior differs by data source type:

   Data source	Steps
   API	Click Next. No changes are needed.
   MaxCompute	By default, the existing configuration is retained. To update it, select the modify option, enter the new configuration, click Data Source Verification, then click Next.
   Object Storage Service (OSS)	By default, the existing configuration is retained. To update it, select the modify option, enter the new configuration, click Data Source Verification, then click Next.

5. Update the index schema, or skip this step.

   • No changes needed: Select Do Not Change and click Next.

   • Changes needed: Select Change and update the schema. Both table mode and developer mode are supported.

6. Update table configurations, or skip this step.

   • No changes needed: Select Do Not Change and click Next.

   • Changes needed: Select Change, then update the settings below as needed.

   Data processing configuration

   Each data source gets two free data processing resources by default. To handle large-volume data updates in a short time window, increase process_partition_count. After updating this parameter, click Billing Documentation to review the additional resource costs.

   Dictionary configuration

   Configure a custom dictionary if the built-in tokenizer does not meet your query tokenization requirements.

7. Select a rebuild method, complete the full data source details, then click Confirm.

   Important

   Choose your rebuild method carefully. The available options depend on your data source type, and some options are irreversible.

   Rebuild method	Available for	What it does	Risk
   Re-import full data	MaxCompute, OSS, API	For MaxCompute data source tables, the system reindexes by pulling the configured partition data and incremental data based on the specified timestamp. The Timestamp field sets how far back (up to 3 days) the new index can trace API incremental data.	Low — existing data is preserved until the new index is ready
   Empty data	API only	Clears all existing data and starts tracing real-time data from the specified timestamp.	High — this cannot be reversed
   Use current index data	MaxCompute, OSS, API	Rebuilds from a selected data version, avoiding data loss and reducing the complexity of pre-rebuild data restoration.	Low — select the restoration version carefully

   For details on selecting a restoration version for Use current index data, see Recover data from indexes.

8. Monitor progress on the Change History page. After the table modification completes, two finite-state machines (FSMs) are created — one to deploy the updated configuration, and one to manually initiate full indexing. Your changes take effect after the FSM for data source modifications finishes. Check the Change History page for the current FSM status to confirm when your changes are live.