Umfriend

Also, if you use duplicate later, consider removing one disk at a time.

How large is the data you're not backing up?

Funny that. I imagined it was because for most users, DP simply works. Anyway, AFAIK, use of trial versions is easy. Moreover, IIRC, if you deactivate a license, DP will still work, just not write to the Pool.

So to be sure, duplication *is not a backup*. But I think I get what you mean. Yeah, Hierarchical Pooling could help in the sense that you could have one underlying pool with just clean/new/trustworthy disks and be certain that they contain exactly one duplicate of (more than x1) duplicated files. The thing is, once a Pool grows larger and has more disks, the "clean/new/trustworthy" property becomes opaque over time so such a setup would require management (not that much, possible to plan but still). If what you need safeguarded fits on a single HDD then I would suggest to consider actual backup-solutions. You could even rotate such backups offsite. Then, not only do you have pretty good protection against drive failure but theft, fire etc. as well.

You *could* use Hierarchical Pooling but as the new drive ages as well and so at some stage you'd want another HDD to serve this purpose, I don;t think it is tenable long-term. Not sure what you mean by "(either in existing machine or elsewhere)". If you want transportability of certain files or drives then it becomes even harder I think. I would say the whole idea of duplication is that you need not worry about which drives contain which duplicate. Just have a decent backup solution and it should be almost perfect.

I have to say I agree with klepp0906 on this one. The only caveat is that users should not expect Stablebit support to be online there often.

Funny that, over here enterprise drives (like the Toshiba MG08/09, right?) are cheaper than the NAS-rated drives

I only let Scanner run a check. I don't see why a full format would do anything unless Scanner actually finds something (but with a new HDD, RMAing might be most sensible).

Does DP say anything on _why_ it won't remove the HDD? Is the HDD to be removed almost empty already or does DP not move anything off of it? Pretty sure your Pool will be fine. Just provide a bit more info on what DP is saying/doing.

I've had the issue of locked files on a HDD that I was trying to remove. A simple reboot fixed that.

I can't remember when I had to run the repair backup tool last, I *think* that was when I was running WHS2011. One question though: Do you make actual backups, e.g. do you run Server Backup as well? Cause if you do, then there is no issue deduplicating client backups temporarily or even definitively (if up-time is nit that high of a priority). Duplication isn't backup and that becomes clear when you have to run the repair tool to bring things back.

Uninstalling Scanner does not sound like a good idea to me. AFAIK, DP will only evacuate a drive marked by Scanner if you have the StableBit Scanner balancer add-in active in DrivePool. And if you have that add-in, it allows for a number of options, for instance, to limit the HDD to duplicated files only.

I see. Yes, it takes a but of engineering but I think it is the best way to go. Edit: I would consider to do: Pool A: HDD 1, 2, 3, x1 duplication Pool B: CloudDrive 4+5, x1 duplication Pool C: Pool A + Pool B, x2 duplication And then, for files that you do not want duplicated, do not store them in Pool C but in Pool A. So for a "user" it would seem his files are on two (virtual) drives, Pool A for undup and Pool C for duplicated files. I am guessing here that the distinction between Pool A and Pool C is clear cut (e.g., work files in Pool C, downloaded media in Pool A). Otherwise this won't work well for a user.

You mean your Stablebit licenses? So best is to deactivate on WHS2011 and then activate on the W10 machine. And yes, install W10 and DrivePool, shut down, connect HDDs, power on and you should be good to go.

Yes, it does get more complicated, especially if you want only parts duplicated. Perhaps to simplify you could (a) duplicate all, (b) duplicate nothing and have real backups or (c) have separate Pools for duplicated files and unduplicated files?

There are file placement rules that may help but I never used them. I do use hierarchical pooling to ensure one duplicate is on one and the other on another group of disks so that I only need to backup one set of disks. Maybe that can help you. But afaik, within one pool, there is no way to ensure one and only one duplicate ends up at a specific subset of HDD and the other duplicate on another subset.

Ah yes, you are right. Hadn't considered your data is duplicated.

There is the Drive Usage Limiter Balancer that allows you to set individual drives to store only duplicated or unduplicated files. So you could set drive 2 to only hold unduplicated files. It would mean that if you have duplicated files, none of the duplicates would arrive there.

First, if the Pool has 12TB free and one HDD has 8.3TB used, then you are not even able to endure the loss of a 14TB drive. The fullest drive has 5.7TB free and if that drive fails, you lose that 5.7TB as well. So then you'd have 6.3TB free on the other drives to cover for the 8.3TB on that 14TB HDD. That won't work. So if it fails slowly, such that Scanner may trigger an evacuation, but you don't add a HDD before it fails completely, you may lose data (assuming no backups). I use duplication for resiliency/up-time and as such always ensure that any Pool has more free space than the largest HDD in that Pool.

Drive replacement seems simplest. Instead of a new MB + CPU with 8 slots, you might consider a 2nd hand SAS HBA card (like IBM M1015 / Dell Perc H310). But in all these scenarios there is the placement of addiotnal HDDs.

AFAIK, NTFS an ReFS (but maybe limited)

https://storage.microsemi.com/en-us/support/sas/sas/aha-1100-24i/ Yes. I have never heard of a SAS card that did not work with SATA drives. I guess you need cables like these: https://www.onlinekabelshop.nl/mini-sas-hd-sff-8643-4x-sata-0-50-meter-40520 but cheaper/better models may exist.

No way a spinner like that reads 300MB/s. Suggest you restart scans on both and report how much they've progressed after like 10 minutes .

So effectively to not use the SSD Optimizer?

Sorry but to me that seems like a weird piece of advice. Managing a collection of torrents might just be one of the use-cases you'd like a Pool for. And how do torrent files work differently from other normal files?