Remind me about matching Tape/Disk RP numbers?

[pkelly_sts]

I remember there was some kind of "restriction" around tape copies having to match the RPs on disk but can't remember the specifics, could someone remind me please?

If e.g. you have a backup job on disk with 60 RPs, and you copy it to tape, then the RPs retained on tape have to *at least* match those retained on disk (e.g. if you delete an early tape which contained RPs that are still on disk, they'll be copied to tape again on the next pass?)

Is that right?

One way to avoid this scenario is to use a lower-retention disk job as the source job for copy-to-tape jobs, but it does effectively mean requiring to have at least two running (on-disk) repositories if you did want to keep more RP's on disk.

Am I remembering correctly?

[HannesK]

Hello,
hmm no, that kind of restriction does not exist. Maybe you remember a different product

Best regards,
Hannes

[pkelly_sts]

Hmm, I definitely remembered /some/ sort of restriction along these lines & having had another search I think I've found it at https://helpcenter.veeam.com/docs/backu ... l?ver=95u4

(However it's specific to forward & reverse incremental, not forever forward)

"Tape Retention and Disk Retention

For forward incremental and reverse incremental backup chains, it is strongly recommended that you set the retention period for the tape archive at least twice longer than the retention period for source backups on disk. The tape jobs analyze the existing tape archives and synchronize them with disk backups. If you have some restore points on disk and the tape archive misses them (for example, if the media pool retention allowed overwriting of these tapes), the tape job will re-write all missing restore points. For more information, see Retention Policy.

For example, there are 14 backup files on the backup repository that are kept for 14 days. The backup to tape job archives files once a week. The retention policy for the media pool is set to 7 days. In this case, Veeam Backup & Replication will first write 14 backup files from the backup repository to tape. After a 7-day interval, Veeam Backup & Replication will start recording the whole set of backup files from the backup repository to tape anew, overwriting backup files on tape with their copies from the backup repository.

When the source job produces a forever forward incremental backup chain, you can set any retention period for the tape archive. To back up such chains, the virtual full mechanism is enabled automatically. Keep in mind that the virtual full backup must be always newer than the full backup on disk. Otherwise, the tape job will copy the full backup from disk and synthesize the virtual full within one backup set. For more information, see Virtual Full Backup."

[soncscy]

Hey Paul,

I've read this a few times and I never saw it as a restriction, more a precaution about re-backing up files on accident.

>If you have some restore points on disk and the tape archive misses them (for example, if the media pool retention allowed overwriting of these tapes), the tape job will re-write all missing restore points.

To me this is more about the selection algorithm occasionally repacking up points that you already allowed to age out on tape.

To be honest, I've never understood this part of the guide since for our shop, tape is about Years of retention, not days, and I just don't know who is really relying on tape for rapid short-term recovery. We keep our weeklies, sure, but Daily has never been exciting for us when it comes to tape.