Set storage inventory

Updated at：2025-11-03

Overview

Storage inventory is a feature that assists users in managing objects within their buckets. Based on the user's configured inventory tasks, BOS can conduct daily or weekly scans of specified objects in the bucket and generate an inventory report. This report is saved as a CSV file in a user-defined bucket, listing objects, their metadata, and the required attribute information configured by the user.

Description:

The inventory function is charged from December 8, 2022 based on the number of listed objects (files). For pricing details, refer to BOS Price Description.

This feature is currently fully supported in the North China (Beijing), East China (Suzhou), and South China (Guangzhou) regions.

The efficiency of inventory exports is influenced by the number of objects in the bucket. For buckets with a large volume of files, a 1-2 day delay in generating the inventory is typical.

Application scenarios

The inventory feature helps users manage bucket objects systematically, serving as an alternative to object storage list API calls. Key uses of the inventory function include but are not limited to:

Conducting regular audits and generating file lists along with their associated attributes.
Streamlining and accelerating workflows through periodic, script-free scanning. Simple configurations enable routine output of file reports.

Prerequisites for use

Before configuring the inventory, you need to understand two concepts:

Source bucket: The bucket to enable the inventory function.
- Listing the objects included in the inventory.
- Providing inventory configuration details.

Target bucket: The bucket storing the inventory.
- Offering inventory list files.
- Including a Manifest file that describes the location of the inventory list files.

Set storage inventory

Sign in to the Baidu AI Cloud Object Storage (BOS) Management Console.
From the bucket list on the left, select the bucket for which you want to set permissions, then click its name to enter the bucket management directory.
Select the Configuration Management tab in the top navigation bar.
On the Configuration Management page, select Advanced Configuration, and click Add Configuration in the Inventory Configuration area to create an inventory rule.

Set inventory rules

Configuration name	Configuration description
Inventory status	Enable: Inventory rules take effect immediately upon creation, with BOS periodically scanning and generating reports based on rule configuration . Disable: The inventory rule is only created but not enabled. You can enable it later .
Inventory name	Enter the inventory rule name, which can only contain uppercase/lowercase letters, numbers, and -_ .; it must start with a letter, and be 1-64 characters in length .
Inventory storage position	The storage position for the output inventory report must be the bucket in the current region
Report inventory prefix	Enter the name for the inventory prefix. You can customize a name as the inventory report prefix generated under this strategy
Include file scope	You can set to apply to all files in the bucket or only to files with a specific prefix .
Inventory export cycle	You can set to export the inventory every three days or weekly
Inventory export format	Currently support to export inventory reports in CSV format

Note:

A single bucket can support up to 20 inventory rules simultaneously. If the rule limit is exceeded, existing rules must be deleted before creating new ones.

Edit or delete inventory rules

You can modify existing inventory rules, such as reactivating a disabled rule. You can also delete inventory rules. Once a rule is deleted, BOS will stop generating inventory reports based on that rule.

Inventory report storage path

Inventory report storage location: The inventory report will be published in the target bucket, and the inventory report will be published in the following path:

Plain Text

1destination-prefix/[source-bucket]/[config-ID]/[YYYY-MM-DDTHH-MMZ]/xxx.csv

Manifest-related files storage location: The relevant Manifest files will be published in the target bucket in the following path:

Plain Text

1destination-prefix/[source-bucket]/[config-ID]/[YYYY-MM-DDTHH-MMZ]/manifest.json
2destination-prefix/[source-bucket]/[config-ID]/[YYYY-MM-DDTHH-MMZ]/manifest.checksum

Meaning of the path field

desitination-prefix: It is the “Report Inventory Prefix” set when you configure the inventory, which can be used to group all inventory reports in the public location of the target bucket.
source-bucket: It is the name of the source bucket corresponding to the inventory report. This folder is added to avoid possible conflicts when multiple source buckets send their respective inventory reports to the same target bucket.
config-ID: It is the “Inventory Name” set when you configure the inventory. When multiple inventory reports are set for the same source bucket and sent to the same target bucket, config-ID can be used to distinguish different inventory reports.
YYYY-MM-DD-HH-MM: Timestamp, which includes the time and date when the bucket scanning starts when generating the inventory report.
manifest.json: Refers to the Manifest file.
manifest.checksum: Refers to the MD5 of the content of the manifest.json file.

Inventory content

The generated inventory includes a list of inventory CSV files and the corresponding Manifest files.

Contents of inventory CSV file list

The output inventory file list includes the following:

Plain Text

1</tbody>

Field description	Description
Bucket	Bucket name
Key	File name
Size	File data length
Etag	Object hash. It may be the base64-encoded MD5 digest of the object data, depending on object creation mode .
StorageClass	File storage class
CreateTime	File creation time. Time format: yyyy-mm-ddThh:mm:ssZ . Such as: 2016-09-07T00:00:00Z（UTC）

Manifest file content

Manifest file content:

Field description	Description
sourceBucket	Source bucket name corresponding to the inventory report
destinationBucket	Name of the target bucket for storing exported files
version	Exported version. The format is "year-month-day", e.g., "2021-11-03"
fileFormat	Report format
fileSchema	Report structure
files	Actual list of exported report files in the target bucket. Include key (inventory file name), size (inventory file size), and md5Checksum (md5 value)

Example of a Manifest report:

Plain Text

1{
2   "sourceBucket": "src-bucket-name",
3   "destinationBucket":"destination-bucket-name",
4   "version":"2020-11-03",
5   "fileFormat":"CSV",
6   "fileSchema":"Bucket,Key,Size,etag,StorageClass,PutDate,CreateTime",
7   "files":[
8        {
9            "key":"destination-prefix/[source-bucket]-[config-ID]-[YYYY-MM-DDTHH-MMZ]-index.csv",
10            "size":1603959,
11            "md5Checksum":"2177e775870def72b8d84febe1ad3574"
12        }
13      ]
14}

API reference

PutBucketInventory: It is designed to create an inventory task in the source bucket. You can name the task and then use this request to create an inventory task.
GetBucketInventory: It is designed to retrieve a specific inventory configuration of a specified bucket through the GET operation.

Set server-side encryption

Managing Bucket Tag