Veeam Beginner Questions

[OhMyEnglishTeaBags]

Hello all,

Firstly sorry if the below is straight forward, i'm responsible for the backup solution where I work so I need to be entirely clear on how it works. I've checked the Wiki's and help guides but still unclear so hoping you can help clear it up! For context we are in the UK trialing Veeam Backup & Replication in conjunction with Wasabi. We also are trialing the Office365 backup but that seems straight forward.

1: A question on retention and immutability. I need to be clear how it works as the reseller for Wasabi has a 30 day restriction to delete files, any files deleted before those 30 days then we get charged x amount (still better than Wasabi's default 90 day limit). If I set retention to 10 days, and immutability for 30 days, does that mean the backup will be kept for 40 days, or only 10 days but protects up to 30 days?

I also read on the wiki: Veeam Backup & Replication will automatically add from 1 to 10 days to the immutability expiration date. Does this mean that my 30 day immutability will actually be 30+(1-10)?

2: Leading on from question 1, immutability is set up (and also enabled in Object Locking @ Wasabi) but when I tested, I could still delete files from local disk as the administrator? I thought the point was so that if we get hit with malware and it starts deleting everything, it wouldn't be able to delete those files?

3: If we wanted to keep local copies for 60 days but cloud copies for 30, would I have to have one SOBR job and then a separate job or is there another way to do this? If I did it that way then i'd be holding day 1-30 twice on local storage? Or I guess the other way is a script like robocopy to copy the file to another directory, that could work.

Thank you for taking the time to read my post and for any replies you may add!

[veremin]

1. The backups will be immutable for 30+ days, if that's what you are asking for. However, it's not recommended to set retention policy shorter than immutability period - the latter is designed to protect the most recent restore points.

2. Are you talking about removing backups from local storage (Performance Tier) or cloud storage (Capacity Tier)?

3. Not sure what you are trying to achieve here, but Capacity Tier does not have its own retention, so files landing on it will expire according to the source job retention.

Thanks!

[OhMyEnglishTeaBags]

Hi Veremin,

Thanks for your reply.

1: Thank you. This makes sense.

2: So I am unable to delete from Cloud Storage, which is great, that's what we want. But on local Performance Tier, using a network admin account I can go to the location, e.g. C:\Backups and I can delete the files. I was expecting Windows to say error, these files are protected. I'm worried that if some malware takes over our admin account, they can delete the local backups.

3: So the length of days is an example here, but because cloud storage is expensive we wanted to upload 30 days backup to the cloud in capacity tier (and therefore 30 days in performance tier because of SOBR) but to have a longer backup retention for performance tier. So how could we achieve uploading 30 days to Wasabi (capacity tier), but keeping 60 days in performance tier?

Thanks again!

[veremin]

2. If you want to have local backups protected, you should consider having repository that supports immutability as Performance extent - Linux Hardened Repository, for example.

3. As mentioned Capacity Tier does not have its own retention, so you will need to create a separate job with 30 day retention and point it to SOBR. However, have you already researched how much data will be present on object storage? How much you will be paying for it? And whether storing data for longer time (60 days instead of 30) will result in significant solution price increase?

By the way, you should not take local storage consumption as an example here, cause Capacity Tier is working in different, forever incremental way, receiving only blocks that have not been transferred already.

Thanks!

[OhMyEnglishTeaBags]

3: If I have a backup job that saves to repository X, if I then point it to SOBR (btw how do I point it to SOBR once it's saved to the repository?) it will save to the repositories in SOBR (e.g. Repository Y and Wasabi). Repository X and Y cannot be the same, can they?

Thank you for your patience!!!

[chris.childerhose]

3. In order to redirect your job to SOBR you will need to also copy the files from repo X in order to map the job properly once it is directed to SOBR. Other options would be -

a. Clone the job and direct the clone to SOBR which will start a new backup chain
b. Created a SOBR adding repository X to it and then it will redirect the job for you

Hope this helps.

[OhMyEnglishTeaBags]

Hi Chris.

So if I want to store 60 days local copy in my local repository but only 30 days in the cloud, what would be the best way?

If I copy the files, it will just copy everything in the folder (all 60 days) unless I made a copy job each day for each new backup? I must be missing something?

Cloning the job will just mean I have it stored twice locally, 1 for 1-60 days in local repo and 1 for 1-30 days in local repo and uploading to cloud (SOBR)?

If I add the repo to SOBR, it doesn't help me achieve the 60 day local copy?

Sorry I just want to have a clear picture in my head.

Thanks!

[chris.childerhose]

I think this forum post might help with your retention and what you need but it is based on the main job retention and then setting the Capacity Tier option to Move - object-storage-f52/different-retention- ... 78388.html

Hopefully, this helps as well.

[OhMyEnglishTeaBags]

Thank you both for your replies. I've set up a backup job to go to the SOBR, but to move backup to object storage as they age out of the operational restore window after 1 day. Then I made a backup copy job to copy it from that location with a separate retention job.

I'm trialing the move backup to object storage being set to 1 day, but I may have to set it to 7 days if the weekly synthetic fulls don't work.

[chris.childerhose]

Not a problem. Let us know how it goes with your jobs and offloading.

[OhMyEnglishTeaBags]

Hello,

Just leading on from before. I've got "move backup to object storage as they age out of the operational restore window after 0 days" enabled and then the copy job (because we want on-site storage to have longer retention that the cloud) but for some reason it's still holding all the backup copies. Although the date is all the 8th May, both repositories are 12GB in size and the whole point is we only want to store the data once, but have the local storage a longer retention.

Screenshot to show what I mean: https://ibb.co/HDZjxvJ

Hope someone can help with this!

Thanks.

Edit: Does it not delete the backup file once out of the chain? Do I have to do it manually? Looking at the restore points, only 2 restore points are local and the rest are in the cloud. Yet they're all sat there in the folder? https://ibb.co/jVF8wYG

[veremin]

It seems that the current backup chain is still active which prevents move policy from offloading restore points. In order for move policy to take place you have to wait until the chain becomes inactive. Thanks!