A tool for syncing Csasandra SSTables out to various destinations.
Example saving to Cloudfiles:
bin/cassandra-syncer --config config.json --target cloudfiles://bucket-name --source /var/lib/cassandra/data
Example saving to another local path, which could be a network volume like EBS:
bin/cassandra-syncer --target directory:///mnt/network-volume --source /var/lib/cassandra/data
Example restoring from the backup:
bin/cassandra-syncer-restore --source directory:///mnt/network-volume --target /var/lib/cassandra/data
This is not just a copying of all available SSTables into the Cassandra data directory -- we keep track of what the current set of SSTables are when we created the backup, meaning we only restore the minimum set of SSTables needed.
Another feature this enables is the ability to prune old backups accurately, based on what is actually needed.
Example pruning backups older than 21 days:
bin/cassandra-syncer-fsck --older-than 21
# TODO: this isn't available on NPM yet (!) npm install -g cassandra-syncer
- fsck utility