Si ha marcado la casilla Reposición de datos históricos, Datastream transmitirá todos los datos disponibles, además de los cambios en los datos, desde el origen al destino.
Si no ha seleccionado esta casilla, Datastream solo transmitirá los cambios en los datos. Para que Datastream transmita una captura de todos los datos de la fuente al destino, debes iniciar la reposición de los objetos que contengan estos datos. Los objetos tienen la forma de esquemas, tablas y columnas de bases de datos.
Otro motivo para iniciar el relleno retroactivo de un objeto es que los datos no estén sincronizados entre el origen y el destino. Por ejemplo, un usuario puede eliminar datos de forma accidental en el destino y, por lo tanto, se pierden. En este caso, iniciar el relleno retroactivo del objeto sirve como "mecanismo de restablecimiento", ya que todos los datos se transmiten a la ubicación de destino de una sola vez. Como resultado, los datos se sincronizan entre el origen y el destino.
Después de iniciar el relleno de un objeto, puedes detenerlo. En el ejemplo anterior, el usuario modifica el esquema de la base de datos y el esquema o los datos se dañan. No quieres que este esquema o estos datos se transmitan al destino, por lo que detienes el relleno retroactivo del objeto.
También puede detener el relleno de objetos por motivos de balanceo de carga. Datastream puede ejecutar varios retornos de datos en paralelo. Esto puede que añada más carga al origen. Si la carga es significativa, detén el relleno de los objetos y, a continuación, inicia el relleno de cada uno de ellos.
Estados de objetos
Los distintos estados del ciclo de vida de inicio y detención del relleno de un objeto son los siguientes:
Sin estado (representado en la interfaz de usuario como -): entre los motivos por los que un objeto puede tener este estado se incluyen los siguientes:
La emisión no se ha iniciado.
La casilla Rellenar datos históricos no estaba marcada (por lo que el relleno se define como manual).
El objeto se excluye explícitamente de la cobertura automática.
La secuencia se ha configurado para incluir tablas futuras. Si esto ocurre, cuando se añadan tablas nuevas a la fuente, no se creará ninguna tarea de relleno automático (porque las tablas nuevas no suelen tener datos históricos que rellenar).
Pending: aún no se ha iniciado el relleno del objeto.
Active: se está rellenando el objeto.
Completed: se ha completado el relleno del objeto.
Stopped: se detiene el relleno del objeto. Si se vuelve a iniciar el rellenado de datos de un objeto, Datastream transmitirá todos los datos asociados al objeto desde la fuente al destino.
Failed: no se ha podido rellenar el objeto y debe iniciarse de nuevo.
Iniciar reposición
Ve a la página Streams (Streams) de la Google Cloud consola.
Haz clic en el flujo que contenga los objetos de los que quieras iniciar el relleno.
Haz clic en la pestaña OBJETOS.
Marque la casilla de cada objeto para el que quiera iniciar el relleno.
Haz clic en INICIAR BACKFILL.
Si has seleccionado un solo objeto, haz clic en INICIAR RELLENO DE DATOS DEL OBJETO en el cuadro de diálogo. De lo contrario, si has seleccionado varios objetos, haz clic en INICIAR RELLENO DE OBJETOS.
Datastream empezará a rellenar los objetos que hayas seleccionado y el estado de cada objeto cambiará de Pending a Active y, después, a Completed. Cuando un objeto tiene el estado Completed, significa que Datastream ha leído todos los datos del objeto, pero es posible que los datos aún se estén cargando en el destino.
Detener el relleno
Ve a la página Streams (Streams) de la Google Cloud consola.
Haz clic en el flujo que contenga los objetos de los que quieras detener el relleno.
Haz clic en la pestaña OBJETOS.
Marque la casilla de cada objeto para el que quiera detener el relleno.
Haz clic en DETENER RELLENO.
Si has seleccionado un solo objeto, haz clic en DETENER RELLENO DE OBJETO en el cuadro de diálogo. De lo contrario, si has seleccionado varios objetos, haz clic en DETENER RELLENO DE OBJETOS.
Datastream dejará de rellenar los objetos que hayas seleccionado y el estado de cada objeto cambiará a Stopped.
Cuando un objeto tiene este estado, se detiene el relleno del objeto. Si se vuelve a iniciar el rellenado de datos del objeto, Datastream transmitirá todos los datos asociados al objeto desde la fuente al destino.
[[["Es fácil de entender","easyToUnderstand","thumb-up"],["Me ofreció una solución al problema","solvedMyProblem","thumb-up"],["Otro","otherUp","thumb-up"]],[["Es difícil de entender","hardToUnderstand","thumb-down"],["La información o el código de muestra no son correctos","incorrectInformationOrSampleCode","thumb-down"],["Me faltan las muestras o la información que necesito","missingTheInformationSamplesINeed","thumb-down"],["Problema de traducción","translationIssue","thumb-down"],["Otro","otherDown","thumb-down"]],["Última actualización: 2025-09-10 (UTC)."],[[["\u003cp\u003eDatastream can backfill historical data and stream ongoing changes from a source database into a destination.\u003c/p\u003e\n"],["\u003cp\u003eBackfill can be initiated to stream all existing data from source objects (schemas, tables, columns) into the destination, or to resync data if it's out of sync.\u003c/p\u003e\n"],["\u003cp\u003eBackfill can be stopped to prevent corrupted schema or data from being streamed, or for load balancing if multiple backfills are overloading the source.\u003c/p\u003e\n"],["\u003cp\u003eObjects in a stream have various statuses like \u003ccode\u003ePending\u003c/code\u003e, \u003ccode\u003eActive\u003c/code\u003e, \u003ccode\u003eCompleted\u003c/code\u003e, \u003ccode\u003eStopped\u003c/code\u003e, and \u003ccode\u003eFailed\u003c/code\u003e, which indicate the state of the backfill process.\u003c/p\u003e\n"],["\u003cp\u003eThe object tab can be used to initiate or stop backfill for objects, and view additional information including status, recent events, total size of data streamed, and details of the table columns.\u003c/p\u003e\n"]]],[],null,["# Manage backfill for the objects of a stream\n\nA stream in Datastream can backfill historical data, as well as stream ongoing changes into a destination. As part of creating a stream, you [configured information about the source database for the stream](/datastream/docs/create-a-stream#configuresourcedbinfo).\n\nIf you selected the **Backfill historical data** checkbox, then Datastream will stream all existing data, in addition to changes to the data, from the source into the destination.\n\nIf you didn't select this checkbox, then Datastream will stream only changes to the data. To have Datastream stream a snapshot of all existing data from the source to the destination, you must initiate backfill for the objects that contain this data. The objects are in the form of database schemas, tables, and columns.\n\nAnother reason for initiating backfill for an object is if data is out of sync between the source and the destination. For example, a user can delete data in the destination inadvertently, and the data is now lost. In this case, initiating backfill for the object serves as a \"reset mechanism\" because all data is streamed into the destination in one shot. As a result, the data is synced between the source and the destination.\n\nAfter initiating backfill for an object, you can stop backfill for it. In the preceding example, the user modifies the database schema, and the schema or data is corrupted. You don't want this schema or data to be streamed into the destination, and so you stop backfill for the object.\n\nYou can also stop backfill for objects for load balancing purposes. Datastream can run multiple backfills in parallel. This may put an additional load on the source. If the load is significant, stop backfill for the objects, and then initiate backfill for them, one by one.\n\nObject statuses\n---------------\n\nThe various statuses in the lifecycle of initiating and stopping backfill for an object include:\n\n- No status (represented in the UI as `-`): Reasons for an object receiving\n this status include:\n\n - The stream hasn't been started.\n - The **Backfill historical data** checkbox wasn't selected (so the backfill is defined as manual).\n - The object is excluded explicitly from being backfilled automatically.\n - The stream is configured to include future tables. If this happens, then when new tables are added to the source, there's no automatic backfill task created for them (because new tables typically don't have any \"historical\" data to backfill).\n\n | **Note:** For more information, see [Configure information about the source database for the stream](/datastream/docs/create-a-stream#configuresourcedbinfo).\n- `Pending`: backfill hasn't yet started for the object.\n\n- `Active`: backfill is in progress for the object.\n\n- `Completed`: backfill is completed for the object.\n\n- `Stopped`: backfill is stopped for the object. If backfill is initiated\n again for the object, then Datastream will stream all existing data\n associated with the object from the source into the destination.\n\n- `Failed`: backfill failed for the object and the backfill must be initiated\n again.\n\nInitiate backfill\n-----------------\n\n1. Go to the **Streams** page in the Google Cloud Console.\n\n [Go to the Streams page](https://console.cloud.google.com/datastream/streams)\n2. Click the stream that contains objects for which you want to initiate backfill.\n\n3. Click the **OBJECTS** tab.\n\n4. Select the checkbox for each object for which you want to initiate backfill.\n\n5. Click **INITIATE BACKFILL**.\n\n \u003cbr /\u003e\n\n | If an object has a status of `Pending` or `Active`, then you can't initiate backfill for the object.\n\n \u003cbr /\u003e\n\n6. If you selected only one object, then in the dialog, click **INITIATE OBJECT BACKFILL** . Otherwise, if you selected multiple objects, then click **INITIATE OBJECT BACKFILLS**.\n\n Datastream will start backfill for the objects that you selected, and\n the status of each object will change from `Pending` to `Active` to\n `Completed`. When an object has a status of `Completed`, this means that\n Datastream has read all the data for the object, but the data might\n still be loading to the destination.\n | If an object has a status of `Failed`, then backfill failed for the object, and you must initiate the backfill again.\n\nStop backfill\n-------------\n\n1. Go to the **Streams** page in the Google Cloud Console.\n\n [Go to the Streams page](https://console.cloud.google.com/datastream/streams)\n2. Click the stream that contains objects for which you want to stop backfill.\n\n3. Click the **OBJECTS** tab.\n\n4. Select the checkbox for each object for which you want to stop backfill.\n\n5. Click **STOP BACKFILL**.\n\n \u003cbr /\u003e\n\n | If an object has a status of `Completed`, `Stopped`, or `Failed`, then you can't stop backfill for the object.\n\n \u003cbr /\u003e\n\n6. If you selected only one object, then in the dialog, click **STOP OBJECT BACKFILL** . Otherwise, if you selected multiple objects, then click **STOP OBJECT BACKFILLS**.\n\n Datastream will stop backfill for the objects that you selected, and the status of each object will change to `Stopped`.\n\n When an object has this status, backfill is stopped for the object. If backfill is initiated again for the object, then Datastream will stream all existing data associated with the object from the source into the destination.\n\n| You can also use the **OBJECTS** tab to view additional information about the objects of a stream. This information includes:\n|\n| - The status of the objects.\n| - How many events Datastream processed and loaded into the destination for an object in the last 7 days.\n| - The total size (in GB) of all events that Datastream processed and loaded into the destination for an object in the last 30 days.\n| - Details about the table columns of the database schemas that are streamed from the source into the destination.\n\nWhat's next\n-----------\n\n- To learn more about streams, see [Stream lifecycle](/datastream/docs/stream-states-and-actions).\n- To learn how to view information about your stream, see [View a stream](/datastream/docs/view-a-stream).\n- To learn how to modify a stream, see [Modify a stream](/datastream/docs/modify-a-stream).\n- To learn how to monitor a stream, see [Monitor a stream](/datastream/docs/monitor-a-stream).\n- To learn how to recover a failed stream, see [Recover a stream](/datastream/docs/recover-a-stream)."]]