> For the complete documentation index, see [llms.txt](https://mariadb.com/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://mariadb.com/docs/gridgain/gridgain-9/gridgain9-management/backup-and-replication/snapshots/point-in-time-recovery.md).

# Point in Time Recovery

Restore a GridGain 9 cluster to any point in time within the low watermark window with point-in-time recovery (PITR).

{% hint style="info" %}
This feature is only available as a part of GridGain 9 Ultimate edition.
{% endhint %}

By using the point in time recovery (PITR) feature, you can restore cluster status to any point above the [low watermark](/docs/gridgain/gridgain-9/architecture/storage/low-watermark-and-gc.md) point.

GridGain 9 recovers to a point in time only within the low watermark window. There is no roll-forward of changes onto a restored snapshot. To recover to a state older than the low watermark, restore a snapshot taken at or before that point.

## Limitations

* Point in time recovery can only be performed for data above the [low watermark](/docs/gridgain/gridgain-9/architecture/storage/low-watermark-and-gc.md) (600000ms by default). Older data can only be recovered by restoring a [snapshot](/docs/gridgain/gridgain-9/gridgain9-management/backup-and-replication/snapshots/data-snapshots.md).
* Recovery of large amounts of data may require additional storage space, as both versions will be temporarily kept available in case the recovery fails.

{% hint style="warning" %}
Do not run a point-in-time recovery and a snapshot restore at the same time.
{% endhint %}

## Performing Point in Time Recovery

To start point in time recovery, use the `recovery tables start` command and specify the table or tables to recover. Table names can be schema-qualified.

```
recovery tables start --tables Person,PUBLIC.Accounts --timestamp 2024-09-10T10:53:00+01:00
```

Once replication is started, it will return the operation ID.

## Monitoring Recovery

Point in time recovery may take a long time. You can monitor current replication status by using the `recovery tables state` command and passing the ID of data recovery to it:

```
recovery tables state --id=944e4abe-b655-4283-9d78-449b5186a7ac
```