I need some advice with regards to using S3 for offsite.
We have a single SOBR on Enterprise Edition with roughly a total of 20TB of backups, with the biggest single vbk of 8TB.
Jobs are forever forward incremental with a 30 day retention
We behind a 100mbps synchronous fiber.
A total 30TB of local storage for backups so no local disk space issues
We want to consolidate to Veaam, in order get rid of our file based legacy offsite backup with Livevault because the contracts up end of Jan, and replace it with the either:
Amazon VTL Gateway Appliance (already have a file gateway for another purpose) and provision a massive local cache disk.
The new Veeam V10 copy to object storage function?
A Capacity Tier - accepting the potential for 30 day old restores on object storage and non 3-2-1 compliance - (I am aware that the Capacity Tiers primary functions is to solve the expensive local storage problem and not DR)
I believe should wait for the V10 copy functions because it looks really simple instead of going through the effort of configuring a VTL appliance, but my colleague disagrees. He seems to think the V10 copy function is too simple and won't perform as well with the large backup files, compared with the trickle capabilities of the VTL Gateway?
VTL is a terrible solution for copying backups to S3 from every perspective. Even just because it will require sending periodic fulls to the cloud (and storing them there), so think bandwidth requirements and storage costs. And I'm not even talking how long restores will take, if you ever need to restore from those cloud copies. Not only you will need much time to copy all required backups back to on-prem before you can even start restores, but you will also need enough disk space to stage them. Might be a huge problem to find this much disk space when we're talking about dealing with a disaster that took away your primary backup repository!
In “Copy Mode" once a backup is copied to the Capacity Tier what happens to those files? Retention rules inheriting from the job or have some funcitonality act like copy jobs or no retention rules?
Yes, the retention is still managed by whatever job created those backups.
No, you don't need to add a synthetic full to your jobs to seal backup chains. Copy mode processes all restore points as soon as they appear in the Performance Tier, including incremental restore points. As such, backup mode doesn't matter and periodic fulls are not required. Sealing remains the requirement for Move mode only, which is designed for archiving older backups such as GFS restore points.