Skip to main content

Backup management

You can use eloqctl to create and manage backup of current EloqKV cluster.

Create backup​

Create a cluster backup and save it at given path on a specified node.

eloqctl backup ${cluster_name} start [OPTIONS] --path /path/to/backup

Options:

  • --path:
    The full path to where the backup is stored. (required)
  • --dest-user:
    User of the destination node where the backup is stored. (default: current user)
  • --dest-node:
    Node address where the backup is stored. If you want to convert backups to AOF or RDB later, this node must be on of the tx server nodes. (default: current node)
  • --password:
    Cluster password if set (default: "")

List backup of a cluster​

List current available backups of a cluster.

eloqctl backup ${cluster_name} list

Cleanup backup of a cluster​

eloqctl backup ${cluster_name} remove [OPTIONS]

If no option is provided, remove will delete all backups of the current cluster.
Options:

  • --until <PERIOD>:
    Deletes all snapshots older than the specified period. Accepted formats: - '2 days' - '15h' - '1 week' - '3 months' - '1y 6mo 2w 4d 3h 5m 7s' See https://docs.rs/humantime/latest/humantime/fn.parse_duration.html for more details.
  • --before>TIMESTAMP>: Deletes all snapshots created before this timestamp. Accepted formats: - RFC 3339: '2024-11-14T15:01:00Z' - 'YYYY-MM-DD HH:MM:SS' (assumed local time zone) - 'YYYY-MM-DDTHH:MM:SS' (assumed local time zone)

Convert existing backup to AOF file.​

eloqctl backup ${cluster_name} dump-aof [OPTIONS] --rocksdb-path <ROCKSDB_PATH> --output-file-dir <OUTPUT_FILE_DIR>

eloqctl will convert a previous backup in this cluster to AOF files. AOF files will be written to the same node where the backup is stored.
Options:
---rocksdb-path:
Path to the backup location. Must match one of the backup path returned in eloqctl backup list.
---output-file-dir:
Path where the AOF files will be written to.
---thread-count:
Worker thread count for converting backup to AOF. Each worker will consume 1 vcpu on the target node. (default:1)

Convert existing backup to RDB file.​

eloqctl backup ${cluster_name} dump-rdb [OPTIONS] --rocksdb-path <ROCKSDB_PATH> --output-file-dir <OUTPUT_FILE_DIR>

eloqctl will convert a previous backup in this cluster to RDB files. RDB file will be written to the same node where the backup is stored.
Options:
---rocksdb-path:
Path to the backup location. Must match one of the backup path returned in eloqctl backup list.
---output-file-dir:
Path where the RDB file will be written to.
---thread-count:
Worker thread count for converting backup to RDB. Each worker will consume 1 vcpu on the target node. (default:1)

Example of Dumping Data from EloqKV and Importing to Other Servers​

  1. Dump data:

    eloqctl backup eloqkv-cluster start --path /data/backup

  2. After the backup is created, check available backups.

     eloqctl backup eloqkv-cluster list
    available snapshots: [
     (
         "eloqkv-cluster",
         2024-12-04T10:02:36.165807800Z,
         "/data/backup/eloqkv-cluster/2024-12-04-10-02-36",
         "172.31.42.205",
         "ubuntu",
     ),
    ]

  3. Convert backup to AOF file.

    eloqctl backup eloqkv-cluster dump-aof --rocksdb-path /data/backup/eloqkv-cluster/2024-12-04-10-02-36 --output-file-dir /home/workspace/output_aof

  4. Check AOF files

    redis-check-aof /home/workspace/output_aof/0.aof

    The output will look like:

    AOF analyzed: size=411068632, ok_up_to=411068632, diff=0
    AOF is valid

  5. Import the AOF files to another server using redis-cli:

    redis-cli --pipe < /home/workspace/output_aof/0.aof

    After importing, the output will look like this:

    All data transferred. Waiting for the last reply...
    Last reply received from server.
    errors: 0, replies: 6567541

  6. Remove previous snapshot

 eloqctl backup eloqkv-cluster remove --until 1min