Hi Ana, thank you for chiming in, see my response inline:

> On 1. Sep 2022, at 13:38, Ana Emília M. Arruda <emiliaarr...@gmail.com> wrote:
> 
> On Tue, Aug 30, 2022 at 3:43 PM Justin Case <jus7inc...@gmail.com 
> <mailto:jus7inc...@gmail.com>> wrote:
> Greetings,
> 
> I am interested in the new V13 storage groups feature.
> If I wanted a given pool to use file folumes on several externally connected 
> USB HDDs:
> 
> -the devices would be connected all the time, would the director fill the 
> devices one ofter the other, or would it scatter the jobs randomly across the 
> devices?
> 
> This will depend on the storage group policy you use. The default 
> "ListedOrder" will always use the first storage in the storage group list 
> unless there is a connection failure. The LeastUsed policy will load balance 
> the storage usage between the storages defined in the list.
> 
> It will not spread jobs randomly across the devices or it will not skip a 
> storage if it gets full.
> 
> -may I extend the list of storages in a given pool to extend the available 
> storage space by adding USB HDD devices to the storage? (I mean distinct HDD 
> devices, not enlarging the storage of a RAID!)
> Yes.
>  
> 
> -This would mean that I don’t need to use vchanger vor this?
> vchanger and storage groups have different purposes, but none will skip a 
> device if it gets full.
> 

So to me it seems Storage Groups could actually be a good way to dynamically  
extend USB storage (although requiring manual interaction). when a device is 
full, and using ListedOrder, I could move the full device  to the end of the 
list and could I still do restore jobs from all devices in the list?

I was thinking vchanger would let me use a bunch of individual disks and fill 
them up one after the other… without me needing to manually interact with it 
after initial correct setup… weird.


_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users

Reply via email to