DCS Backup and Restoration Overview

Video Tutorial

This video demonstrates how to back up data of a DCS Redis instance. As product functions evolve, the GUI may vary. This tutorial is for reference only.

Backup Modes

DCS instances support the following backup modes:

• Automated backup

  You can create a scheduled backup policy on the DCS console. Then, data in the chosen DCS instances will be automatically backed up at the scheduled time.

  Automated backups are scheduled on a daily basis. You can choose to back up data once or multiple times a week depending on your needs. Backup data is retained for a maximum of seven days, after which it is automatically deleted.

  The primary purpose of automated backups is to create complete data replicas of DCS instances so that the instances can be quickly restored if necessary.

• Manual backup

  Backup requests can be issued manually. Data in the chosen DCS instances will be backed up to OBS.

  Before performing high-risk operations, such as system maintenance or upgrade, back up DCS instance data.

Impacts on Instances During the Backup

• Backup tasks run on replicas. During the backup, the instance can provide services properly.
• During a full synchronization between the master and replica nodes or when the instance is under high load, data synchronization can take longer to complete. If a backup is initiated before synchronization is complete, the backup data will lag behind the latest data on the master node.
• If new data is written to the master node during the backup of the replica node, the backup file will not include the data changes that occurred during the backup.

Additional Information About Data Backup

• Supported instance types
  • Redis: Only master/standby, Proxy Cluster, Redis Cluster, and read/write splitting instances can be backed up and restored, while single-node instances cannot. To back up a single-node instance, you can export its data to an RDB file using redis-cli. For details, see How Do I Export DCS Redis Instance Data?
  • Memcached: Only master/standby instances can be backed up and restored, while single-node instances cannot.
• Backup mechanisms

  DCS for Redis 3.0 (discontinued) persists data to AOF files. You can persist data to RDB or AOF files in manual backups, and to RDB files in automated backups. Instances of the Enterprise (storage) Edition can be persisted only to RDB files.

  • To export RDB backup files of DCS Redis 3.0 instances, run the redis-cli -h {redis_address} -p 6379 -a {password} --rdb {output.rdb} command in redis-cli.
  • For single-node DCS Redis 3.0 instances that allow the SYNC command, you can run this command to export RDB files. Proxy Cluster DCS Redis 3.0 instances do not allow the SYNC command due to their architecture, so RDB files cannot be exported.

  Backup tasks are executed on replica nodes. DCS compresses the data persistence files of the replica nodes and transfers them to OBS for storage. DCS periodically checks the backup policies of all instances per hour and starts backup tasks for instances that require backups.

  Data backup applies only to key-value data and do not cover other data such as instance configurations.

• Backup time

  You are advised to perform backups during off-peak hours.

• Storage of backup files

  Backup files are stored in OBS.

• Handling exceptions in automated backups

  If an automated backup task is triggered while the instance is restarting or being scaled up, the scheduled task will be postponed to the next cycle.

  If an instance backup fails or is postponed due to other ongoing tasks, DCS will attempt to back up the instance again in the next cycle. A maximum of three backup attempts are allowed per day.

• Retention period of backup data

  Automated backup files are retained for one to seven days based on the retention period you set. They will be automatically deleted upon expiration, but at least the most recent backup remains available.

  • By default, a maximum of 24 automated and manual backups can be retained for a DCS instance. If you need to temporarily increase the limit, contact the customer service.
  • If the number of backups exceeds the upper limit, new manual backups cannot be created. To continue creating backups, you need to manually delete old backups. In automated backup mode, the earliest automated backups are automatically deleted to make room for new ones.
  • When an instance is deleted, its backup data is also deleted. If you need to save the backup data, download it before deleting the instance.
  • Exercise caution when deleting all backup files, because this operation will compromise backup-related capabilities, such as data restoration in the event of a failure.

Data Restoration

• Data restoration process
  1. You can initiate a data restoration request on the console.
  2. DCS obtains the backup file from OBS.
  3. Data reads or writes on the instance are suspended.
  4. The data persistence file of the master instance is replaced.
  5. The new data persistence file is loaded.
  6. Data restoration is complete, and the instance starts to provide data read and write services.
• Impacts of data restoration on service systems

  Data restoration is performed on the master node using the backup file. During the restoration, data read and write services are suspended until the restoration is complete on the master instance.

• Handling data restoration exceptions

  If a backup file used for data restoration is damaged, DCS will attempt to fix it during the restoration process. If the backup file is successfully fixed, the restoration proceeds. If the backup file cannot be fixed, the master/standby DCS instance will be restored to the state before the restoration.