Umfriend

Hi thomas4d, it would help if you could show some screenprints because it is unclear to me what your settings are right now. In principle, I would say that you could: 1. Manage Pool -> File protection -> Pool file duplication -> Disable (I assume you have x2 duplication right now) 2. Manage Pool -> File protection -> Folder Duplication -> Select the folder you want x3 duplication for and set it as such. Now, duplication is not backup but especially if you disable duplication then I would even more advise to ensure you have backups somewhere. Just a heads up.

As I understand it though, nowadays it does not make that much of a difference. Sure, a 35W TDP CPU may draw more power than a 15W TDP (although with Intel it will be way more if it uses turbo as the TDP is based on base clock speed, extremely confusing that), but it will do so for a shorter period of time. Effectively it won;t matter much. On the other hand, once you need/want that extra bit of oompf, it's there. I *think* nowadays, the number of cores is more important as they all draw some in idle. So two powerful cores may be better than two weaker ones (even if the latter has a lower TDP), but a 4-core 35W TDP CPU will use more than a 2-core 35W TDP CPU as its idle power draw will be higher (that of course assumes equality for a significant number of properties of the CPUs).

But what about an AMD Athlon 200GE? TDP of 35W but like 6 times more powerful than an E-350 and you can purchase a MB that allows for a lot of SATA or SAS cards and whatnot. https://www.guru3d.com/articles_pages/amd_athlon_200ge_review,5.html - Max consumption of 52W for system https://www.guru3d.com/articles-pages/amd-brazos-platform-tested-the-e350-apu-review,10.html - Max consumption of 46W for system Granted, the test setups may not be entirely comparable but still. In IDLE it measures a difference of 9W to the advantage of the 200GE...

I am wondering whether with the low consumption modern CPUs have when Idle, you might be better off with a regular CPU. You might get by with underclocking and undervolting, although it is a bit of a hassle to set up right. But no, I can not easily find a real server solution (i.e motherboard with PCIe slots) with, say, 15W TDP CPUs.

You are running WHS. Surely you have got clients with access to shared folders? Just connect the failed HDD, via a USB thingy for instance, and copy to the appropriate shared folders for instance. Having said that, I would first copy from one HDD to another, just as it is fast and requires the smallest uptime for the failing HDD.

Nope. Scanner just checks whether the HDD can be read, not whether a bit has flopped.

So that is not my expertise (as if I have any) but if that failed drive would come to life, I would consider to not use it in your WHS box. I would want the strain on that disk to be minimal. So I would connect it to a PC, perhaps through an USB enclosure, and simply copy as much of the date of it as possible. I think I would try to use a copy program that won't halt on issues with a file but continue with the next. Perhaps Robocopy can do this for you. PS I was running WHS2011 until Christmas last year, still love it.

That is Date modified. You can add a column for date created and that is what I am looking for. Date modified will always be the last date you ran a backup and that is fine. TBH, I don't think there is a way to recover.

Open diskmgmt and design a drive letter to the BUP HDD. If by renaming you mean changing the labels, then no, I don't think that would be an issue.

We would need a bit more info. Did you have duplication? What amount of data did you have on the Pool? I fear though that your backups are lost. You have two 2TB HDDs that you backup (I think) to a 2TB HDD. Server Backup does try to retain backups but once the backup HDD becomes full, it will wipe and do a new fresh full backup. As you have added a new HDD and that HDD will have had data on it, chances are that the BUP HDD got full and wiped. This is speculation on my part. What you could try to do is to locate the large .vhd files on the BUP HDD and have explorer show the "Date created". My guess is that it will show August 31st, around 1 AM... This is (one of the) the reason(s) why my BUP HDD is far larger than the amount of data to be backed up and why I have rotating BUP HDDs...

I think it best to open a ticket. You may neede to run the/a troubleshooter. Never done that so I don't know. Removing seems hard as you don't have the space unless you unduplicate but barring having a backup ready I would not like to go that way.

OK, so this does seem weird to me. It is not right but I do not know how to fix this. I hope Christopher will join in soon. One thing, and this is a bit embarrassing, but I have found that just rebooting and trying again does the job quite often. No clue why and successful with many things.

What does the bar say when you hover over it? There is a way to "optimize" this by creating partitions and I strongly advise against it. Doing that would negate the whole idea behind duplication as you might possibly have the two duplicates of the same file located on one physical HDD which may fail. I think you either need some way to attach more drives or now replace the smallest with a larger one. I see you have a Server backup HDD, what OS are you running and what do you backup? I have to say that the Disk Space Equalizer seems weird to me. As far as I can tell I have two Pools with the same settings, both only Scanner and Disk Space Equalizer but they behave differently. What I would try if I were you is to turn off Disk Space Equalizer and turn on Duplication Space Optimizer instread (but always Scanner as the #1).

There is a bar just below in the GUI called Pool Organization. I suspect that if you hover above it it will say something like file placement not optimal or somesuch. To the right there is a small arrow through which you can Re-balance. That, I think, would help. Mind you, in your current setup you will always have slightly over 1TB unusable at least. This is because with x2 duplication. DP will not write duplicates to the same disk. Let's say one duplicate is placed on either E or J, then the other will be placed on D. E+J equals 4.43TB, so that is the max you can store on these. J would then have the other duplicates for 4.43. The remaining 1TB can not be used. This can be solved by as little as adding a 1TB HDD. One other thing, if you have Scanner then I would _always_ have Scanner as the #1 balancer.

I am assuming you are copying everything from the failing drive. Merging the structure is easy. From the (copy of) the Poolpart.* folder, copy the contents to the Pool and it should work out fine. e.g., if you have a folder in the Pool, say, P:\SomeData, you may have on the backup/copy drive (let's say that drive is Y:) a folder like Y:\PoolPart.*\SomeData. You can simply select everything in Y:\PoolPart.*\ (and I mean "in", do not select the PoolPart folder itself, just the contents) and copy tp P:\. Removing drives typically does well, except possibly for locked files or unduplicated files that are corrupted. Personally, if a HDD is starting to show serious wear _and_ you have no duplication _and_ at the first attempt it does not run smoothly then I would: 1. Stop the DP service (to ensure no file replacement due to a possible rebalancing occurs); 2. Copy the data from the failing drive to another drive; 3. Physically remove the failing HDD 4. Start DP 5. Copy from another drive to the Pool. I have an IBM 1015 SAS controller in IT mode and Scanner reads SMART fine. But I did have to select a setting (something like Unsafe) in the Advanced Settings. Funnily enoughm that option is not active right now and it still reads my SAS connected HDDs. Not sure how that came about. But there are threads here that discuss this.

Yup. No worries and good luck.

So moving the files from the old HDD to the other remaing HDDs in the Pool (of which the HDD you just added would be one) is done automatically by DP once you click on Remove in the GUI for the failing HDD (this is what I call "Let DP do its magic"). It may take a while. The reason I am recommending to copy to a HDD outside the Pool first is just that, given you do not have duplication, I would like to safeguard the data as simply as possible. I think (but do not know) that reading from it to move through a Remove (in the GUI) may put a bit more stress on the failing HDD then a simple copy first.

So when you add a 6TB HDD to that setup, and assuming you have not tinkered with the balancing settings, any _new_ files would be stored on that 6TB HDD indeed. A rebalancing pass, which you can start manually, will fill it up as well. With default settings, DP will try to ensure that each disk has the same amount of free space. It would therefore write to the 6TB first until 4TB is fee. Then equally to the 6TB and 4TB until both have 3TB free etc. The 500GB HDD will see action only when the others have 500GB or less available. This is at default settings and without duplication.

With failing drives I think it is always a good idea to copy/backup that drive first, just to be sure. Bear in mind though that a new re-balancing pass may place files on the suspect HDD again. This is one of the reasons I recommend Scanner because it will try to (tell DP to) evacuate a suspect HDD and prevent new files being written to it. In any case, the default way of dealing with this is, I think (and how I would do and have done it): 1. Add a new HDD to the Pool; 2. Through the GUI, remove the suspect HDD; 3. Let DP do its magic (be patient); and, 4. Once done, remove the physical HDD. You're 2nd option seems fine as well except that when you remove through the GUI, DP will already try to move files from the HDD-to-be-removed to the other disks in the Pool. Option 1 does not work, even though a new HDD may have a PoolPart.xxx folder, DP will (typically?) not actually recognize that as part of a Pool as the GUID of the HDD is different.

Tricky this. Are some of the HDDs connected through USB? What you could do, if you have some spare machine somewhere: 1. Disconnect the faulty drive 2. Connect to another machine, see if you can copy data from it. If you can, then at least the data is safe. 3. If you could, then you could remove the drive that DP will say is missing. Then the Pool would be alright again but miss data. 4. Copy/move from (2) to the Pool This is what I would do if I really though DP measuring and moving files might put to much stress on the HDD. Not sure if this is recommended by Stablebit though.

So DP will run as a trial for 30 days on the new machine. I guess you need to contact Stablebit to work this out. Perhaps Christopher can help out pursuant to this thread but you might want to open a ticket. Can't help you with that one (and if I could, I'd be selling licences! ).

The new build, is that an OS-reinstall or a new machine? And the "new drives", are they really new or just Pooled HDDs transferred from one system to the other. I'll give a few pointers: 1. De-activate your license on the current machine. Do write down the license code somewhere, you'll need it on the new machine. 2. Settings are stored in the Pool itself and DP on another machine will pick them up once it finds the Pooled HDDs. It should be easy BUT: 3. If you use a balancer add-in that is not installed by default then I would first build the machine while not connecting the Pooled HDDs. Install DP, activate and then install the add-in(s). Only then (and a reboot first makes sense to me) connect the Pooled HDDs. You should be good to go.

Yes. I have never tried it but DP should not need drive letters. You can also map drives to folders somehow so that you can still easily explore them. Not sure how that works but there are threads on this forum.

Do you have Scanner? And yeah, even though I have a far smaller Pool (6 HDD in a 9 HDD setup), I label them with a sticker.

Oh, and I am not convinced about energy/heat. 25 500GB SSDs is just 12.5TB. I would think they use more than one 14TB spinner.