Friday, November 8, 2013

Snap shot autodelete

Snap shot autodelete

1.0.1 The purpose of this policy is to provide guidance as to use the Netapp Default functioning of snapshot auto delete. Snapshot autodelete is a policy based space management feature that is implemented in Data ONTAP 7.1 in order to get back space in volume for user data. It allows the user to define a policy to automatically delete snapshots when the volume is nearly full. It keeps the amount of unused spaced in a filer to a minimum level. The snapshot autodelete is available for Flexible volumes only.

1.1 Scope

1.1.1 This policy applies to all the volumes in Netapp Filer which are upgraded to Data ONTAP 7.1 ( and subsequent Data ONTAP versions)

1.2 When Does Snapshot Delete Happen

1.2.1 When the “trigger” criteria is near full.
Volume: The volume is near full
Snap_reserve: The reserve is near full
Space_reserve: The Space reserved is near full.

1.2.2 What defines “near full” ?
By Default near full is defined as 98%
Default is controlled by the flag “wafl_reclaim_threshold”

1.2.3 How to change the Default “wafl_reclaim_threshold” value ?
Priv set diag
Setflag wafl_reclaim_threshold 

1.3 When Does Sanpshot Autodelete Stop ?
1.3.1 Once autodelete starts, explicit conditions need to be met to stop autodelete

1.3.2 When the free space in the “trigger” criteria reaches a particular user specified percentage, the snashot autodelete stops.

1.3.3 This percentage is controlled by the value of “target_free_space”. Default value of target_free_space is 20%

1.4 What Snapshots is autodelete allowed to delete ?

1.4.1 The user can protect certain kinds of snapshots by deferring their delete.
1.4.2 The “commitment” option defines this:
“try”: Delete snapshots which are not being used by any data mover, recovery or clones. (NOT LOCKED)
“disrupt”: Delete snapshots locked by datamovers – snap mirror, dump/restore ( Mirror, Dumps are aborted )
1.4.3 Snapshots locked by clones, snap restores, cifs shares are not deleted.
1.4.4 “disrupt” snapshots are only deleted if there are no “try” snapshots left to delete.

1.5 In what order are the snapshots deleted ?

1.5.1 The “Delete_order” option defines the age order. If the value is set to:
“oldest_first”: Delete older snapshots first.
“newest_first”: Delete new snapshots first.

1.5.2 The “Defer_delete” option defines the order on ‘name’. If defers the deletion
Of the snapshots to the end. If the value is set to:
“scheduled”: Delete the scheduled snapshots last. ( Identified by the scheduled snapshot naming convention)
“User_created”: Delete the user created snapshots last. Note: User_created here refer to everything-that-is-not-snap-sched-created-by-Data ONTAP itself.
“prefix”: Delete the snapshots with names matching the “prefix” string last.

1.5.3 The “Prefix” option value pair is only considered when “defer_delete” is set to
“prefix”. Otherwise it is ignored.

1.6 The Hierarchy for selection snapshots on different criteria is as follows:

1.6.1 commitment
1.6.2 defer_delete
1.6.3 delete_orader


Auto delete will first delete a “try” snashot which does not lie in the “defer_delete”
Criteria and is the oldest such snapshot. If such a snapshot is not found, the
“defer_delete” criteria would be ignored to find a snapshot. Lastly we would move
To the “disrupt” criteria (if commitment=disrupt) and attempt to find a snapshot.

1.7.1 snap autodelete vol1 on --- enables autodelete
1.7.2 snap autodelete vol1 off --- disables autodelete
1.7.3 snap autodelete vol1 show - shows the current settings
1.7.4 snap autodelete vol1 reset - resets all autodelete options to default for vol1
1.7.5 snap autodelete vol1 help - shows help commands


1.8.1 snap autodelete
snap autodelete [on off show reset help]
snap autodelete

No comments:

Post a Comment