Keeping a second copy
Each backup copied, as it is stored, to a second drive or your NAS after it passes its check. Either copy restores on its own.
On this page
A backup on one drive is lost with that drive. Keep a second copy somewhere else copies each backup to a second place: another drive, a USB disk you keep in a drawer, your NAS (Home and Pro), or an S3 bucket off-site (Pro).
Turning it on#
- Open the job (Edit) and go to the Where step.
- Under Keep a second copy somewhere else, switch it on.
- Choose Another drive or USB, Network share / NAS or Cloud (S3), and the folder (for the cloud: the saved S3 profile, the bucket and the folder in it). A drive other than the backup's is best.
- Choose what the copy keeps (below), then Test second copy.
- Save the job.
The job page then shows two cards: Backup and Second copy.
What it copies#
- Each backup that completes and passes its check is copied as it is stored: compressed and encrypted, nothing read from your PC again.
- On a drive or a share, every file is read back from the copy and compared. In a bucket, every upload carries a checksum the bucket checks before it keeps a byte, and the backup's files are then listed there with their sizes. A file that does not match is sent once more; if it still does not match, that backup is not kept on the copy (
E-COPY-002). - A backup that failed, or failed its check, stays on the backup only.
- A copy is never half there: a backup appears on the copy only when all of it has been copied and checked.
What the copy keeps#
| Choice | The copy keeps |
|---|---|
| The same as the backup | What the job's own retention keeps. A mirror's copy keeps the newest. |
| The last 30 days / the last year | Backups from that period, whole chains only. |
| The last 10 backups | That many, whole chains only. |
| Every backup | Everything, until you change it. |
A mirror job can keep several backups on its copy: a single up-to-date copy on the drive, and some history on the NAS.
When it is made#
- Right after each backup that completes and passes its check. It runs on its own line in Activity, and the backup's report says Second copy: copied to ….
- When the place comes back. If the drive is unplugged or the NAS is off, the copy waits. GBM AI looks again every few minutes and catches up by itself. A drive is recognised by its volume serial, whatever letter it gets.
- Copy now, on the job page, copies whatever the copy does not have yet.
Pausing scheduled backups pauses catching up too. A copy that is running can be paused or stopped like a backup; a stopped copy keeps the backups it finished.
Restoring from the copy#
Restore lists each backup once, with a Held on column: E: J: for a backup on both, copy waiting for one the copy does not have yet.
- If the backup's drive is not there, the point restores from the copy. Nothing to choose.
- Backups the backup's own retention removed, but the copy still keeps, are listed too.
- An encrypted backup needs the same password from either copy.
- On another PC, Restore ▸ Open backups… ▸ A folder opens the copy's folder like any backup.
What the job page says#
| The copy | Means |
|---|---|
| Up to date | The copy has every backup it should. |
| 3 points to copy | Backups are waiting to be copied; the next backup or Copy now copies them. |
| Waiting for J: (Travel SSD) | The drive or share is not there. Plug it in or switch the NAS on. |
| Copy failed | The last copy did not finish. Open its report to see why. |
A copy in an S3 bucket#
- Each backup goes up as it is stored, under the folder you chose in the bucket, so Restore ▸ Open backups… ▸ S3 bucket opens the copy like any cloud backup: on this PC, or on a new one with nothing but the S3 account.
- Restore names the bucket on the points it holds (S3). When the backup's drive is gone, open the copy from the bucket as above.
- What the copy keeps works as on a drive, whole chains only.
Not yet#
A second copy of a backup that itself goes to S3 comes in a later version.