Batch Operation

Last updated: 2020-01-15 16:39:41



The batch operation feature of COS allows you to manipulate objects in a bucket in batches. Currently, you can perform the following batch operations:

  • Replicating objects

You can put the needed objects in an inventory file, which can come from the inventory report generated by the inventory feature (you need to enable the inventory feature first) or be created in the specified format on your own. The batch operation feature can operate on the objects in batches based on this inventory file. For more information, please see Batch Operation Overview.


  1. Log in to the COS Console and click Batch Operation on the left sidebar to enter the batch operation job management page.
  2. Click Create Job to start creating a batch operation job. The main configuration items are described as follows.
    • Job Region: Select a region for the job to be created. The task region should be the same as the bucket region where the objects pending operations in your inventory reside; otherwise, the job would fail. The Chengdu and Chongqing regions are currently in beta test.
    • Inventory Format: Select a format for the inventory objects as needed. There are two formats:
      Inventory Format Field Configuration Instructions
      COS inventory report - Please select this format if the inventory file comes from an inventory report generated by the inventory feature
      CSV Ignore Placeholder field, which can be used when your inventory file contains irregular fields
      Bucket Bucket name
      Key Name of an object in a bucket. If CSV file format is used, the object name is URL-encoded and must be decoded before it can be used
      VersionId Object version ID. If versioning is enabled for the bucket, COS assigns a version number to the objects added to the bucket. If you want to use an object version other than the latest version, you can select a version ID that contains an inventory object
    • Inventory Bucket: Select the bucket where the inventory is stored.
    • Inventory File Path: Enter the inventory file or CSV file path in the format of directory/manifest.json or directory/manifest.csv, respectively. For example, if you have an inventory stored in the examplebucket-1250000000 root directory, the inventory path will be manifest.json.
  3. Click Next and the inventory ETag will be displayed, reflecting that you have selected the correct inventory objects. Then, you will enter the operation configuration page.
    • Job Type: Select the type of operation you want to perform on all the objects in the inventory list.
    • Destination Bucket: Select the bucket where to store the replicas of the objects in the inventory list.
    • Storage Type: Standard or Standard_IA.
    • Server-side Encryption: Select whether to encrypt the object replicas. Options include no encryption or SSE-COS encryption.
    • Access Permission: Set access permissions to the object replicas. Options include inheriting destination bucket's permission, private read/write, and public read/private write.
    • Object Metadata: Set metadata for the object copies. You can choose to copy the original metadata or replace all metadata.
  4. Click Next to enter the "Other Configurations" page where you can configure the following items:
    • Job Description (Optional): Description of the job; optional.
    • Job Priority: A job of a higher priority will be performed first. The value can be a positive integer. The greater the value, the higher the priority.
    • Job Report: Select whether to generate a job report.
    • CAM Role: You can create a CAM role or select an existing one to authorize COS.

      To authorize COS to batch operate, you need to create a CAM role. For more information on CAM roles, please see Role Overview.

  5. Click Next to enter the information check page where you can check the batch operation job configuration items. If you need to make a change, click Modify or Previous accordingly. After confirming that everything is correct, click OK.