All Products
Search

This Product

    ApsaraDB RDS:Download backups

    Document Center

    ApsaraDB RDS:Download backups

    Last Updated:Mar 28, 2026

    Download ApsaraDB RDS for MySQL backups for long-term retention, querying, migration, or auditing.

    Important

    Downloaded backup files cannot be used to restore an ApsaraDB RDS instance directly. To restore an instance, see Overview of data restoration solutions.

    Choose a download method

    Before downloading, identify your scenario to pick the most efficient method.

    ScenarioRecommended approach
    Migrate to another RDS instance or a self-managed databaseRestore to a new instance, restore from a physical backup, or migrate data using Data Transmission Service (DTS)
    Query data at a specific point in timeRestore to a pay-as-you-go or Serverless instance, query, then release the instance. See Restore full data. Alternatively, download the backup as described in this topic and restore it to a self-managed database.
    Save a local copy or download for auditingDownload the backup as described in this topic, export data using Data Management (DMS), or use mysqldump
    Upload a backup to Object Storage Service (OSS)For cloud disk instances, create an Advanced Download task and save directly to OSS. For other instances, download the backup first and then upload it to OSS. Alternatively, use Database Backup (DBS) and select OSS as the storage type.
    Long-term backup retentionSee Long-term backup retention
    Automate backup downloadsUse DBS for logical backup and configure automatic downloads
    Download binary logsFor cloud disk instances, download the log backup for the target point in time. For instances with high-performance local disks, see Download a local disk backup.

    Usage notes

    • Failed tasks are not charged.

    • A download task cannot be canceled after it starts.

    • Each cluster or instance supports only one download task at a time. A new task cannot start if a previous task is running or has failed.

    • Downloading a backup does not use the current instance's bandwidth.

    Billing

    Network traffic fees

    Download typeFee
    Internal network downloadFree
    External network downloadFree for the first 500 GB per instance per month; pay-as-you-go beyond that

    To view your external download volume, go to the ApsaraDB RDS instance list, click the instance ID, and check Backup Download Volume in the Instance Resources section on the Basic Information page.

    Purchase a network plan to cover external download traffic at a discounted rate.

    If you download backups to OSS, charges are based on your actual OSS usage.

    Backup set conversion fees

    This fee applies when you use the Advanced Download feature for a cloud disk instance to convert a snapshot backup into CSV, SQL, Parquet, or qb.xb (physical backup) format. No free quota applies.

    RegionConversion fee
    Public cloudUSD 0.03125/GB

    No conversion fee is charged if the conversion task fails.

    After decompressing a qb.xb (physical backup) file converted from a snapshot backup, the file size may be smaller than the Backup Conversion Traffic displayed in the console.

    Region

    Conversion fee (USD/GB)

    Public cloud

    0.03125

    Download backups from a cloud disk instance

    Cloud disk instances support Advanced Download tasks for a specific point in time or a specific backup set. Convert snapshot backups to CSV, SQL, Parquet, or qb.xb (physical backup) format and download to a URL or directly to an OSS bucket.

    Prerequisites

    Before you begin, make sure that your instance meets all of the following conditions:

    • Instance status: Running

    • Storage type: Enterprise SSD (ESSD) or Premium Performance Disk

    • Database version: MySQL 8.0 or 5.7

    • Supported regions: China (Chengdu), China (Shenzhen), China (Guangzhou), China (Qingdao), China (Beijing), China (Zhangjiakou), China (Hangzhou), China (Shanghai), Hong Kong (China), Malaysia (Kuala Lumpur), Indonesia (Jakarta), Japan (Tokyo), Singapore, US (Silicon Valley), US (Virginia), and Germany (Frankfurt). The feature is rolling out to additional regions — check the console for the current list.

    • Minor engine version: 20201031 or later. If your instance does not meet this requirement, upgrade the minor engine version.

    • Backup encryption: Disabled. Advanced Download is not supported for encrypted instances that use cloud disks.

    • Data archiving: Disabled. Advanced Download is not supported for Premium Performance Disk instances with data archiving enabled.

    • RAM permissions: The Resource Access Management (RAM) user must have backup file download permissions. See Grant permissions to a read-only RAM user to download backup files.

    Important

    The Download Instance Backup button may be unavailable even when the minor engine version meets the requirement. This can happen when the database proxy is enabled on instances with hardware limitations. To resolve this, disable the database proxy, then upgrade the minor engine version. After the upgrade, re-enable the proxy if needed.

    Limitations

    • Unsupported objects: Downloaded backup files do not include expression indexes, foreign keys, generated columns, hidden columns, views, functions, stored procedures, system variables, or triggers.

    • Unsupported databases: System databases are excluded: information_schema, mysql, performance_schema, sys, and __recycle_bin__.

    • Unsupported data types: Download tasks may fail for tables containing GEOMETRY, POINT, LINESTRING, POLYGON, MULTIPOINT, MULTILINESTRING, MULTIPOLYGON, or GEOMETRYCOLLECTION data types.

    • Name restriction: Database and table names cannot contain forward slashes (/). Tasks will fail if names include this character.

    • OSS storage class: Only OSS buckets using the Standard storage class are supported as download destinations. To save to a bucket with a different storage class, either download to a Standard bucket first and convert the storage class, or download to a local machine and upload to the destination bucket.

    Download a cloud disk backup

    1. Go to the ApsaraDB RDS instance list, select the instance's region, and click the instance ID.

    2. In the left navigation pane, click Backup and Restoration > Base Backup List.

    3. Start the download task:

      • Log backup: On the Log Backup tab, find the target log file and click Download in the Actions column. Log backup must be enabled.

      • Data backup: On the Data Backup tab, find the target backup set and click Download Instance Backup in the Actions column.

    4. In the Download Time Point and Backup Set step, select Download by Backup Set or Download by Time Point, then click Next.

    5. In the Download Instance and Database/Table step, keep the default (Instance Download) and click Next.

    6. In the Download Destination and Format step, configure the destination and format:

      Download destinationConfiguration
      OSS (recommended)Enter the bucket name and directory prefix. Set the download format to CSV (without header), CSV-with-header, SQL, Parquet, or qb.xb (physical backup). If this is your first time using this feature, click Go to Authorize > Agree to Authorization to grant ApsaraDB RDS access to OSS.
      URLSelect URL and the download format. After the task completes, manually generate a download link. The link is valid for a limited period.

    7. Accept the terms and click Finish. The console redirects you to the Backup Download List tab to monitor the task.

    8. Get the backup file:

      • OSS: Wait for the task status to become Successful. The file is automatically saved to the specified OSS bucket and directory.

      • URL: Wait for the task status to become Successful. Within 3 days, click Generate Link in the Download Destination column. Set the link validity from 5 minutes to 1 day (default: 2 hours). After the link expires, temporary data is deleted and you must create a new download task.

      Important

      The conversion process competes for computing resources. A download may fail due to temporary resource allocation issues or unsupported data formats. If a task fails, create a new download task.

    9. Download the file using wget or curl (recommended for stability and resumable downloads): wget

      ParameterDescription
      -cEnables resumable downloads
      -t 0Retries indefinitely
      -OOutput path and filename
      nohup ... &Runs in the background; the process exits after the download completes
      ParameterDescription
      -C -Enables automatic resumable downloads
      --retry 10Retries up to 10 times on failure
      -oOutput path and filename
      nohup ... &Runs in the background; the process exits after the download completes
      nohup wget -c -t 0 "<backup-download-url>" -O <local-path/filename.tar.gz> > /tmp/download.log &

      Example:

      nohup wget -c -t 0 "https://dbs-****.aliyundoc.com/****.tar.gz****" -O /backup/examplebackup.tar.gz > /tmp/download.log &

      curl

      nohup curl -C - --retry 10 "<backup-download-url>" -o <local-path/filename.tar.gz> > /tmp/download.log &

      Example:

      nohup curl -C - --retry 10 "https://dbs-****.aliyundoc.com/****.tar.gz****" -o examplebackup.tar.gz > /tmp/download.log &

      For small files, paste the external URL directly into your browser.

    If the download speed drops below 64 KB/s, the download may be interrupted. Make sure you have a stable network connection. If you mount OSS using ossfs, the default multipart_size limit is 100 GB. For backup sets larger than 100 GB, increase this value. See ossfs and Mounting options. Use only wget or curl to download backup sets — third-party tools may download the file repeatedly and generate extra outbound traffic charges.

    Downloads for instances with high-performance local disks

    For instances with high-performance local disks, no data format conversion is required. Get the download link for a physical backup or log backup immediately.

    Prerequisites

    Before you begin, make sure that:

    Download a local disk backup

    1. Go to the ApsaraDB RDS instance list, select the instance's region, and click the instance ID.

    2. In the left navigation pane, click Backup and Restoration > Base Backup List.

    3. Select the Data Backup or Log Backup tab.

    4. Find the target backup file and click Download Instance Backup or Download in the Actions column.

    5. In the dialog box, copy the internal URL or the external URL:

      • Internal URL (recommended): Fast and free. Download from an Elastic Compute Service (ECS) instance in the same region and Virtual Private Cloud (VPC). Cross-region downloads over the internal network are not supported.

      • External URL: Use this to download from a local machine or another network environment.

      The download link is valid for 1 hour. Download the file promptly. If the link expires, create a new download task. A download already in progress is not affected by the link expiration.

    6. Download the file using wget or curl (recommended for stability and resumable downloads): wget

      nohup wget -c -t 0 "<backup-download-url>" -O <local-path/filename.tar.gz> > /tmp/download.log &

      Example:

      nohup wget -c -t 0 "https://dbs-****.aliyundoc.com/****.tar.gz****" -O /backup/examplebackup.tar.gz > /tmp/download.log &

      curl

      nohup curl -C - --retry 10 "<backup-download-url>" -o <local-path/filename.tar.gz> > /tmp/download.log &

      Example:

      nohup curl -C - --retry 10 "https://dbs-****.aliyundoc.com/****.tar.gz****" -o examplebackup.tar.gz > /tmp/download.log &

      For small files, paste the external URL directly into your browser.

    A log backup is a copy of the binary logs. To retrieve binary log content, see mysqlbinlog. For high-availability or three-node edition instances, both the primary and secondary instances generate log backups. View primary and secondary instance IDs on the or three-node editionService Availability page. To restore data to a self-managed database, the data backup time must fall within the time range of the log backup, and both must be from the same instance (same instance ID).

    FAQ

    Can RDS backups be uploaded to OSS?

    Yes. Cloud disk instances support Advanced Download tasks that save directly to OSS. For other instance types, download the backup first and then upload it to OSS.

    Is there a bandwidth limit for downloading backups over the internet?

    No. ApsaraDB RDS does not impose a bandwidth limit. Download speed depends on your server's bandwidth. External downloads beyond the free quota (500 GB/month) are charged for outbound traffic. See Billing.

    How do I use downloaded data backups and log backups?

    Restore them to a self-managed database. See Restore an ApsaraDB RDS for MySQL physical backup file to a self-managed database or Restore a logical backup file to a self-managed database.

    Can the default `qp.xb` format for physical backups of high-performance local disk instances be converted to `.gz`?

    If the Restore to database or table feature is enabled, the backup compression format must remain qp.xb and cannot be changed. If this feature is not enabled, use the ModifyBackupPolicy API operation to change the format from qp.xb to tar.gz. This only changes the compression method and does not affect the instance.

    When I restore data downloaded from a cloud disk instance to a local MySQL database, I get `ERROR 1148 (42000): The used command is not allowed with this MySQL version`. How do I fix this?

    Run show variables like 'local_infile';. If the result is OFF, run set global local_infile = 1; to enable file import, then retry the import.

    Why is the backup size in the console different from the downloaded file size?

    The downloaded file is compressed and is generally smaller than the size shown in the console. Restore the file to verify data completeness. See Restore an RDS for MySQL snapshot backup file to a self-managed database.

    Can a backup file downloaded from the console be directly restored to another RDS instance?

    No. To move data to another RDS instance, restore the backup to a self-managed database first (see Restore from a physical backup or Restore from a logical backup), then use DTS to migrate to a new RDS instance, or use the full backup import feature to import MySQL backup data from OSS.

    Can I cancel a download task in progress?

    No. A backup download task cannot be canceled after it starts.

    Some data is missing from the backup list. What should I do?

    1. Restore the full data to a new instance.

    2. On the new instance, identify the missing tables or data.

    3. Export the missing data from the original instance.

    4. Use DTS to synchronize the exported data to the restored instance.

    When I use an internal URL to download a backup, I get `ERROR 403: Forbidden`. How do I fix this?

    Check that the ECS instance is in the same VPC as the RDS instance and can communicate over the internal network. Also check that the download command uses standard (half-width) double quotation marks around the URL. If the issue persists, use the external URL instead.

    How do I delete entries from the backup download list?

    Entries in the backup download list cannot be deleted. This does not affect instance operation.

    Can a downloaded backup file be directly imported into a self-managed database?

    A physical backup cannot be directly imported. Create a logical backup, download it to a local machine, and then import it.

    When I download using wget, I get `ERROR 504: Gateway Time-out`. How do I fix this?

    Add --read-timeout=300 --connect-timeout=300 to the wget command.

    Why does the downloaded file appear garbled?

    The backup is a compressed archive. Decompress it with:

    tar -izxvf <compressed_package_filename>.tar.gz -C <path_to_decompressed_file>

    How are downloaded backup files named?

    Backup files are named using the host ID and the timestamp of the instance.

    When downloading to OSS, I get `Unsupported bucket XXX's StorageClass is IA, Please Use bucket's StorageClass is Standard`. How do I fix this?

    The target OSS bucket must use the Standard storage class. Either download to a Standard bucket and convert the storage class, or download to a local machine and upload to the destination bucket.

    Why can't I find the download buttons in the console?

    If your instance is a primary instance, check that it meets all prerequisites for backup download. Read-only instances are not backed up — download backups from the corresponding primary instance.

    Why am I charged even though I did not download any backups?

    The charge is likely the backup set conversion fee. This fee is incurred when you use the Advanced Download feature to convert a snapshot backup into CSV, SQL, Parquet, or qb.xb (physical backup) format. View the corresponding Backup Conversion Traffic in the Backup Download List. See Backup set conversion fees.

    What's next

    References