Skip to end of metadata
Go to start of metadata

 

The ZFS file system allows the user to take snapshots (copy-on-write transactional object model) of the archive and journal. These snapshots can then be used to recover the database to a consistent state. 

To use ZFS with NuoDB, first setup and configure ZFS, create a ZFS dataset for NuoDB, and then start a NuoDB Storage Manager and point the both archive directory and journaling directory to the same ZFS dataset. A ZFS snapshot is taken on a dataset. Hence, the restriction of placing both archive and journal on the same dataset.

 

1  Example Native ZFS Setup on Ubuntu

 

2  Mount a ZFS Dataset

Examples to mount, unmount, remount, new mount.

 

3  Start a NuoDB Storage Manager

Launch the NuoDB Manager on the NuoDB Host where the storage manager will run, then run the following commands.

Initialize TRUE | FALSE

 Configure the --initialize flag to true if this is a new archive directory. If an archive directory already exists, it will be overwritten if --initialize is set to true. Therefore, if you are connecting to an existing ZFS archive directory, the --initialize flag should be set to false.

 

 

  • No labels