EN 
30.11.2025 Ondřej WELCOME IN MY WORLD

This website is originally written in the Czech language. Most content is machine (AI) translated into English. The translation may not be exact and may contain errors.

Tento článek si můžete zobrazit v originální české verzi. You can view this article in the original Czech version.
NetApp ONTAP práce se Snapshoty

NetApp ONTAP working with Snapshots

| Petr Bouška - Samuraj |
Storage Snapshots are an effective solution for (short-term) retention of older versions of data. In a NetApp ONTAP environment, we can schedule the automatic creation of Volume Snapshots and the retention of a defined number of their versions. Using NetApp FlexClone technology, a Snapshot can be used to create a writable clone of the volume, which can be used for testing, development, or data recovery.
displayed: 3 943x (1 233 CZ, 2 710 EN) | Comments [4]

Note: The areas described in the article were tested in version ONTAP 9.15.1P8. The article primarily focuses on SAN Volumes and Snapshots on them.

In the article Veeam Backup & Replication - Snapshot-Only Job and Recovery from Storage Snapshot we dealt with Snapshot orchestration and recovery from the Veeam side. We also covered some concepts and principles. Here we will look at working with Snapshots directly on storage with the NetApp ONTAP system. For data recovery we have various options, but all steps must be performed manually.

NetApp Snapshots Technology

NetApp was one of the first companies (possibly the very first) to introduce the feature of efficient Storage Snapshots. The key advantage of NetApp Snapshots is that they have no performance impact and are highly scalable. They efficiently utilize storage because they only occupy space for changed data blocks.

Snapshots can be used for both NAS volumes (file access - NFS export, SMB/CIFS share) and SAN volumes (block access - LUN), as well as for S3 buckets.

ONTAP Snapshots

Snapshot refers to instant capture of system or data state at a specific time. NetApp describes a Snapshot as a read-only image of a volume at a specific point in time. No data copying or duplication occurs. Snapshots do not replace backup.

WAFL and pointers

NetApp uses its WAFL (Write Anywhere File Layout) technology to create low-overhead Snapshots, which it refers to as Snapshot Copies. WAFL uses pointers to data blocks on disk. When writing, it does not overwrite existing blocks, but writes (updated) data to new (free) blocks and changes the pointers.

A Snapshot contains only pointers that were in the volume at the time the Snapshot was taken (blocks contained in the Snapshot are locked and are not deleted or overwritten when original data is deleted). Creating a snapshot is almost instantaneous, takes up little space (only deleted or changed data blocks since its creation) and does not burden the system (does not slow it down).

Volume Snapshot

Snapshots are created at the FlexVol volume (Flexible Volume) level. They are stored in the volume from which they were taken (occupying its space). Since ONTAP 9.4, we can create up to 1023 for one volume. Above FlexVol is the protocol layer (NFS, SMB/CIFS, FC/FCoE/iSCSI).

Snapshots can be created manually or automatically using a scheduler. Automatic Snapshot deletion is usually used, which removes Snapshots that have reached a certain age. Snapshots can also be replicated to another NetApp system, which may be in a different geographical location.

Data consistency in Snapshot

When we perform a SAN volume Snapshot, it is Crash-consistent. If we want to have a Snapshot in a consistent state, we must use some application with client integration. This could be NetApp SnapCenter or Veeam Backup & Replication for VMware.

NetApp FlexClone

FlexClone, also referred to as a virtual copy, is a Snapshot with write capability. It allows instant creation of a writable copy of a volume that shares data blocks with the parent. It is used for testing, development, analysis, or recovery without affecting production data.

We can clone a FlexVol volume (a Snapshot is created or an existing one is used) or a specific Snapshot which creates a FlexClone volume. Each FlexClone can be further cloned. A FlexClone volume can be separated from its parent volume at any time using the Split function.

Note: FlexClone technology can also be used for file cloning in NAS environments or LUN cloning in SAN environments.

NetApp tools and features

NetApp offers additional tools for working with Snapshots, such as

  • SnapCenter - centralized backup management, integration with applications (e.g. Exchange, SQL)
  • SnapMirror - data replication between two NetApp systems
  • SnapRestore - enables quick recovery of entire volumes or individual files from Snapshots
  • SnapVault - long-term data backup using snapshot replication to secondary storage
  • SnapLock - ensuring data immutability (Immutability / WORM)

Using Snapshots for data recovery

  • recovery of specific files - within NAS volumes we can easily restore deleted or corrupted files from specific Snapshots, these are available in the .snapshot directory in the given Share (if enabled, it is available to users)
  • recovery of entire volume (Volume Restore) - restoration (rollback) of entire volume to Snapshot state, overwrites current volume data and is irreversible, existing Snapshots remain
  • FlexClone of Snapshot - cloning creates a volume that we can mount to a server and copy data from

Working with Snapshots in NetApp ONTAP

In newer versions of the ONTAP system, we can manage Snapshots through both CLI (Command Line Interface) and GUI (Graphical User Interface) System Manager.

Manual creation and deletion of Snapshots

A Snapshot can be very easily and quickly created at the current time. Similarly, we can delete an existing Snapshot without affecting the volume or other Snapshots.

ONTAP System Manager

  • ONTAP System Manager
  • Storage - Volumes - click on the desired volume
  • switch to the Snapshot copies tab
  • creating a Snapshot
    • click on Add
    • enter a name and click Add
NetApp ONTAP - Add Volume Snapshot
  • deleting a Snapshot
    • for the selected Snapshot, click on the three dots on the right and choose Delete
    • check Delete snapshot copy and confirm Delete
NetApp ONTAP - Snapshot copies - Delete

ONTAP CLI

volume snapshot create -vserver <vserver_name> -volume <volume_name> -snapshot <snapshot_name>
volume snapshot delete -vserver <vserver_name> -volume <volume_name> -snapshot <snapshot_name>

Snapshot reserve

On a volume we can set a percentage reservation for Snapshots. The space available for data is reduced by this amount. This part of the volume capacity is reserved for storing changes caused by Snapshots. It is part of the volume itself, data and Snapshots are stored in the active file system.

Note: If the snapshot reserve is not sufficient, Snapshots will begin to consume regular data space. The volume may become full. We also have the option to enable automatic deletion of Snapshots when the Snapshot reserve is exceeded.

  • ONTAP System Manager
  • Storage - Volumes - for the selected volume, click on the three dots on the right and choose Edit - Volume
  • at the bottom is the Snapshot copies (local) settings section
  • under Snapshot reserve % enter the percentage value of the reserve, information will be displayed showing how much space this represents
NetApp ONTAP - Volume - Snapshot reserve

Snapshot policies and Snapshot scheduling

Snapshot Policies determine when snapshots should be created, how many copies should be retained, and how they should be named. To determine when Snapshots should be created, a Snapshot Job Schedule is used. We can combine almost any variants, such as regular hourly, daily, weekly, but also various special ones. Basic configurations are prepared in advance.

When creating or editing a volume, we can enable scheduling of Snapshot creation (Schedule Snapshot copies) and select a prepared policy. According to the schedule, Snapshots will be automatically created and old ones deleted (after reaching the specified number).

ONTAP System Manager

Schedule management

  • ONTAP System Manager
  • Protection - Overview - find and expand the Local policy settings section
  • click on the arrow next to the Schedules panel
NetApp ONTAP - Protection - Schedule Snapshot copies

Snapshot policies management

  • ONTAP System Manager
  • Protection - Overview - find and expand the Local policy settings section
  • click on the arrow next to the Snapshot policies panel
NetApp ONTAP - Protection - Snapshot policies

Setting policy on volume

  • ONTAP System Manager
  • Storage - Volumes - for the selected volume, click on the three dots on the right and choose Edit - Volume
  • at the bottom is the Snapshot copies (local) settings section
  • check Schedule Snapshot copies and select Snapshot Policy
NetApp ONTAP - Volume - Snapshot copies (local) settings

ONTAP CLI

job schedule cron create ... 
volume snapshot policy create ... 
volume create ...
volume modify ...

Information about Snapshots and their size

We can see the list of Snapshots for a specific volume. It is easy to find out how much capacity all Snapshots occupy in total within a given volume. More complex is obtaining a detailed overview of the size of individual Snapshots. To obtain this, it is often necessary to use the command line (CLI). Through CLI you can display details such as Snapshot size, differential data (delta) or volume that can be freed (reclaimable).

ONTAP System Manager

  • ONTAP System Manager
  • Storage - Volumes - click on the desired volume
  • Snapshot list
    • switch to the Snapshot copies tab
  • occupied space
    • on the Overview tab we see the Capacity panel
    • we see information about the volume, occupied space, reserve for Snapshots
    • we can click in the upper part and a more detailed overview will be displayed
NetApp ONTAP - Volume - Capacity (Snapshots size)

ONTAP CLI

Information about volume utilization. We can see how much Snapshots occupy.

AFF::> df -volume VMware_vol_02

Filesystem                       kbytes         used      avail capacity  Mounted on  Vserver
/vol/VMware_vol_02/         12777527708  11627695260 1149832448      91%  ---         svm-iscsi
/vol/VMware_vol_02/.snapshot 2254857828   1057566280 1197291548      47%  ---         svm-iscsi
2 entries were displayed.

How much free space is in the reservation for Snapshots.

AFF::> volume show -volume VMware_vol_02 -fields snapshot-reserve-available

vserver   volume        snapshot-reserve-available
--------- ------------- --------------------------
svm-iscsi VMware_vol_02 1.10TB

List of Snapshots for a given volume and their size.

AFF::> volume snapshot show -volume VMware_vol_02
                                                                 ---Blocks---
Vserver  Volume   Snapshot                                  Size Total% Used%
-------- -------- ------------------------------------- -------- ------ -----
svm-iscsi
         VMware_vol_02
                  VeeamSourceSnapshot.2025-05-23_1600    43.24GB     0%    0%
                  VeeamSourceSnapshot.2025-05-24_0000    71.50GB     0%    1%
                  VeeamSourceSnapshot.2025-05-24_0800    186.9GB     1%    2%
                  VeeamSourceSnapshot.2025-05-24_1600    60.24GB     0%    1%
                  VeeamSourceSnapshot.2025-05-25_0000    69.09GB     0%    1%
                  VeeamSourceSnapshot.2025-05-25_0800     1.18GB     0%    0%
6 entries were displayed.

What space will be freed if we delete one or more Snapshots. The command performs a calculation. The value may be smaller than the Snapshot size. Some blocks may be shared by multiple Snapshots, so they are not deleted when one Snapshot is deleted.

AFF::> volume snapshot compute-reclaimable -vserver svm-iscsi -volume VMware_vol_02 -snapshots VeeamSourceSnapshot.2025-05-23_1600

A total of 1961984 bytes can be reclaimed.

Information about the difference (number of bytes that have changed) between two Snapshots or a Snapshot and the active system.

AFF::> volume snapshot show-delta -vserver svm-iscsi -volume VMware_vol_02 -snapshot1 VeeamSourceSnapshot.2025-05-23_1600 -snapshot2 VeeamSourceSnapshot.2025-05-25_0800

A total of 455055507456 bytes (111097536 blocks) are different. Elapsed time between the Snapshot copies: 1d 15h 59m 55s.

Data recovery from Snapshot

File recovery in NAS environment is simple and described in the documentation Restore files from Snapshot. Here we will look at recovery within SAN.

Volume recovery from Snapshot

Rolling back the volume to the Snapshot state. This step overwrites the current volume data and is irreversible.

ONTAP System Manager

  • ONTAP System Manager
  • Storage - Volumes - click on the desired volume
  • switch to the Snapshot copies tab
  • for the selected Snapshot, click on the three dots on the right and choose Restore
NetApp ONTAP - Restore Volume from Snapshot

ONTAP CLI

volume snapshot restore -vserver <vserver_name> -volume <volume_name> -snapshot <snapshot_name>

Volume cloning using FlexClone

From a Snapshot we can create a clone that functions as a standard volume (the LUN is also cloned). We can mount it to a server and access the data on it. This could be for example a VMware ESXi server, where it is possible to register VMs from the Snapshot and recover data inside the VM.

Note: We can create a FlexClone directly from a volume, but a Snapshot is automatically created and used.

Step-by-step procedure for VMware connection

  • on the array we create a Snapshot clone
  • on the array we map the LUN to hosts (Map to initiator group)
  • on the ESXi server we perform Rescan HBA
  • on the ESXi server we add a Datastore (Storage - New Datastore), to which we connect the VMFS volume with a new signature (Resignature unresolved VMFS volume)

ONTAP System Manager

  • ONTAP System Manager
  • Storage - Volumes - click on the desired volume
  • switch to the Snapshot copies tab
  • for the selected Snapshot, click on the three dots on the right and choose Clone volume
  • enter a name and click Clone
NetApp ONTAP - Clone volume - FlexClone

In the volume list we can see the created clone and work with it in the standard way.

NetApp ONTAP - Storage - Volumes

ONTAP CLI

volume clone create -vserver <vserver_name> -flexclone <clone_name> -parent-volume <source_volume> -parent-snapshot <snapshot_name>
Author:

Related articles:

NetApp ONTAP

Articles that relate to NetApp All Flash FAS (AFF) and Fabric-Attached Storage (FAS) disk arrays with the ONTAP operating system.

Backup

Articles dedicated to backup (Backup), replication (Replication) and restoration (Restore) of data. That is, data protection (Data Protection) using backup copies and recovery after a crash (Disaster Recovery).

If you want write something about this article use comments.

Comments
  1. [1] fd

    To mi neda ... snapshot ktery neovlivnuje vykon je nesmysl, technicky nerealizovatelne. Vykon to ovlivnuje vzdy. Snapshot prece znamena, ze se prestane zapisovat do stavajicich dat, a nove zapisy miri do prazdneho prostoru. A to jaksi od prirody take znamena, ze cist se musi jak z tech nemodifikovanych tak z tech modifikovanych dat = automaticky vznika fragmentace.

    Dokonce ani samotne vytvoreni snapshotu neni zadarmo, jakkoli typicky mnohem drazsi je jeho odstraneni. Jednoduse proto, ze pri zalozeni snapu staci pockat na dokonceni transakce, zatimco pri jeho odstranovani je treba vsechny pouzite bloky oznacit jako volne.

    Presne proto (vykonostni dusledky) je nevhodne (a vrele nedoporucovane) na primarnim storage udrzovat snapy jakoukoli delsi dobu.

    Nevim od kdy nabizi snapovani NetApp, ale treba EMC uz nejmene 15 let.

    Wednesday, 04.06.2025 17:54 | answer
  2. [2] Samuraj

    respond to [1]fd: Nejsem odborník na NetApp a nerad bych se dopustil nepřesnosti. Někdo od NetAppu by to určitě vysvětlil hned a jasně.

    Ale celé to vychází z NetApp patentované technologie WAFL. Ta je založena na tom, že zapisuje data kamkoliv. Při modifikaci nepřepisuje data, ale zapíše do volného místa. Jako primární se využívají odkazy na bloky. A z toho vychází ten efektivní Snapshot. Minimálně dříve to ostatní výrobci dělali jinak a proto byly Snapshoty problém z pohledu výkonu.

    Já jsem první prezentaci na NetApp viděl asi před 15 lety a již tehdy tam vysvětlovali tyto výhody. A myslím, že to měli již dlouho předtím.

    • comment responded to by [3]fd
    Thursday, 05.06.2025 10:25 | answer
  3. [3] fd

    respond to [2]Samuraj: Patentuje si leckdo lecos, toto je prosta fyzika. Ani na SSD neni seek zadarmo a tvrdit ze neco zadarmo je muze leda velmi blabolici PR.

    "Ta je založena na tom, že zapisuje data kamkoliv." ... jiste, kamkoli kde je misto === fragmentace. Cim vice snapu na storage bude, tim vice fragmentace.

    "Při modifikaci nepřepisuje data, ale zapíše do volného místa" === COW. To taky stoji nemalou cast vykonu a opet to generuje fragmentaci.

    Ostatne BTRFS umi totez a na libovolnem HW. Pricemz je to zcela verejne, a tudiz pripadne opravitelne. A dovolim si dokonce tvrdit, ze spousta tech magickych a patentovanych technologii je prave vykradeny opensource. Ono to totiz dost casto vykazuje i presne stejne chyby.

    Friday, 06.06.2025 17:13 | answer
  4. [4] Penous

    respond to [3]fd: Nerad bych se pouštěl do dlouhých diskuzí ale WAFL určitě není COW ale ROW (redirect on write), srovnávat s BTRFS to jde možná lépe se ZFS ale právě spojení zároveň s RAID a NVRAM není v opensource řešeno nebo spíše nemůže když musí běžet na jakémkoliv HW. Stačí si přečíst obecně dostupné info jako je wiki k ONTAP nebo k WAFL. Ontap / WAFL vznikl někdy kolem 1991-2 a hlavní principy u WAFL už tam byly tenkrát. Snapshoty (většinou COW) umí většina výrobců storage od druhé poloviny 90tých let :-)

    Friday, 10.10.2025 13:29 | answer
Add comment

Insert tag: strong em link

Help:
  • maximum length of comment is 2000 characters
  • HTML tags are not allowed (they will be removed), you can use only the special tags listed above the input field
  • new line (ENTER) ends paragraph and start new one
  • when you respond to a comment, put the original comment number in squar brackets at the beginning of the paragraph (line)