Hi I have the below jobs running to Wasabi immutable (direct)
1 hour jobs - retention 90 days
4 hour jobs - retention 30 days
Daily - retention 30 days
and weekly jobs - retention 30 days
Any advise on this retention setup ? This setup is working well buy just wanted to get your advise
Create active full backup periodically on x day is not enabled. Do I need to set this for the above jobs ?
(For jobs to a local hardened Linux repository this option is enabled by default)
Hello! Backup frequency (RPO) and retention policy are determined solely by business requirements... you should confirm with your business leaders if what you have set up is acceptable, considering any compliance requirements around data protection your company may be a subject of. For example, most businesses are required to keep certain records for at least 3 years. Asking random people on an Internet forum is just not the right thing to do as you will get totally random feedback that has nothing to deal with the actual needs of your business.
TDG wrote: ↑Jun 03, 2023 10:03 am(For jobs to a local hardened Linux repository this option is enabled by default)
That is not true actually, active full backups are not a default setting for any scenario. Unless you're a subject to internal policies or regulations requiring you to do period active full backups, usually it makes no sense to enable them.
Thanks Gostev point noted. But pls don't underestimate the random people on this forum lol they are helping me for the past 10 years. Ok about the full backups. Cheers and keep up the good work
Frankly, I'm a little surprised over direct to S3 backups in the first place. Are you using a different technology for disaster recovery (as we all should but few of us get the opportunity to implement) or is Wasabi actually fast enough to meet restore time objectives in a DR situation?
He was only asking feedback about the retention setup, otherwise my first concern would not even be restore performance but rather the fact that it does not meet the 3-2-1 rule for the number of copies/medias... restore performance becomes important only if you actually have a good backup to restore from
Hi EWMarco. Yes we have good B/W so direct to Wasabi is working fine with a local hardened Linux repository & a replication job to DC/Co-Lo form HQ (3-2-1) Also we have Azure ASR as well (Might try Zerto, Rubrik but they may be expensive & we are a happy Veeam customer) Keep it simple
The one thing I will point out about direct to object is making sure you understand the restore options available by your object vendor of choice. For example, you may not be able to do an instant VM recovery on the object vendor's infrastructure because they don't have dedicated servers that can run the compute or networking workloads from their side. It would run on the storage server and that may not have the horsepower you need for your IVMR workloads. If your preferred vendor meets this definition, then all restores would have to be downloaded to the location you target for your restore operations.
Is that a bad thing? Am I knocking object partners? Not at all! I have many of my VCSP partners leveraging object via SOBR Capacity Tier and Direct-to-Object who don't need IVMR for certain workloads. My intent is to point out the trade-offs and considerations when selecting components to include within your infrastructure, both internal and external. Other platforms will offer the ability to host heavy machine workloads for failover purposes. But they will most likely cost more to meet that infrastructure workload requirement.
This is a great time to review the extensive VCSP partner network options Veeam has available. They are able to help provide custom tailored solutions to meet the individual needs of any business's workload requirements.