This role attempts to create filesystem safe point-in-time snapshot images. It will ensures file handles are closed, cache is flushed, I/O is completed, and the base volume is temporarily unmounted. Once unmounted, a snapshot will be generated and the volume remounted.
See netapp_eseries.santricity collection's readme for details on how to create, use, and maintain snapshot consistency groups. The collection's role, nar_santricity_host,
WARNING! This role cannot be application aware and will require applications to be placed into a filesystem consistent state prior to execution; also be aware of any high-availability solutions that will attempt to remount unmounted devices such as the NetApp E-Series BeeGFS HA solution.
eseries_snapshot_pit_safe: true # Whether to wait for all files to be close, sync and unmount volume before taking a point-in-time snapshot # image; otherwise, a simple filesystem suspend (dmsetup suspend) will be issued (Default: True). eseries_snapshot_pit_timeout_sec: # Maximum time to wait for files to closes when I(eseries_snapshot_pits_safe==True) (Default: 600). eseries_snapshot_pits: # List of consistency group point-in-time snapshot definitions.
- group_name: # Snapshot consistency group name pit_name: # (Optional) Name for the point-in-time snapshot. pit_description: # (Optional) Description for the point-in-time snapshot. volumes: # (Optional) List of volumes that are a subset of the consistency group's volume members.
BSD-3-Clause
Nathan Swartz (@ndswartz)