Christopher (Drashna)

StableBit DrivePool doesn't have a master and subordinate/duplicate copy. Both are equally viable, and treated as such. This is very different from how Drive Bender handles things.5 As for being documented, not really, but sort of. Eg: https://wiki.covecube.com/StableBit_DrivePool_Knowledge_base#How_To.27s That said, if the data is in the same relative path under the PoolPart folders, they're considered duplicates. Changing the duplication settings, or even remeasuring can kick off a duplication pass that will automatically prune the duplicates, as needed. Also, the "dpcmd" utility has an option to disable duplication for the entire pool, recursively. However, that kicks off a duplication pass that actually manages the files. Just have both products installed. That's it. You can fine tune settings in StableBit DrivePool, in the balancing settings, as the "StableBit Scanner" balancer is one of the 5 preinstalled balancer plugins. That should be fixed now. Though, the file system scan won't trigger the drive evacuation. And yeah, that fix was shipping in the 2.5.5 version, and the latest stable release is 2.5.6, so this definitely shouldn't be an issue anymore. (we haven't seen it in a while).

Simple answer is, you don't. The best way to do this would be be to add the 2x4TB drives to a pool, and then add that pool and the 8TB disk to another pool. Enable redundancy on the top level pool. That would get you what you want.

I'm guessing that the SMART warning is for the same disk, and that it's "Uncorrectable Sector Count" and/or "Pending Sector Count". If so, then yeah, the disk is failing, and there is an issue with the platters (or possibly the heads). The disk in question should be removed from the pool and system immediately. And it should be RMA'ed if it's under warranty. If you're running into issues/errors with removing the disk from the pool, use the "Force damaged disk removal" option, as it is likely erroring out on reading files on the failing disk. If all the data on the disk in question is duplicated, then I would recommend just physically pulling the disk, and letting DrivePool handle the reduplication, when it's able to. However, yeah, when StableBit Scanner marks a disk as failing, StableBit DrivePool will attempt to move all of the pooled data off of that drive, to help prevent data loss. Which may be/is why most of the other disks are completely full. That said, if the drive is under warranty and you have the money, it may be worth doing an "Advanced replacement" RMA. This is where they send you a new disk first, and then give you time to mail back the bad disk. This would allow you to swap out the disks and move the data.

Well, the simplest option is to limit the upload bandwidth to 80mbps, as that should get close to 750GB per day. However, you'd want to split this up between drives on the same account. As for the UserRateLimitExceeded, I suspect that's a different issue, actually. And just to make sure, could you head to https://stablebit.com/Contact and open a ticket there, so we can be sure.

Try deleting the file/soft link from the underlying disks (in the "PoolPart" folder), and see if that helps.

Not if you use Dynamic Disks. We do not support these, due to some of the complex setups that can be created. However, you can use Storage Spaces with StableBit DrivePool.

Could you install the latest beta and see if this is still occurring? http://dl.covecube.com/CloudDriveWindows/beta/download/StableBit.CloudDrive_1.2.0.1282_x64_BETA.exe

Oh, that sounds like the "reset"3.3v pin issue with the SATA power connector, for white label drives. If you look that up, you should see what I mean... and that's problem it.

Nope. The DrivePool pool is not actually a disk. It's an emulated drive, and contains no actual data. All the files and data reside on the underlying drives. If you want to do this, then I would recommend partitioning a number of the disks, enabling BitLocker on those new partitions, and then pooling those.

Having the ECC memory reduces the number of places that corruption can occur. But the CPU (if damaged or overclocking) can still cause issues. But there are also: Issues with the storage controller, Issues with the connectors, Issues with the cables, Issues with the power, Issues with the drive's controller or firmware Or even, software on the host system. StableBit Scanner does perform a surface scan of the disk, which is a sector by sector read of the entire disk. This can catch both issues with the disk platter, or even communication issues. Having ECC or parity protection, in general is to find random bit flips. So it's good for detecting outright corruption of the data. However, random bit flips are exceptionally rare (like, urban myth level of rare). And StableBit Scanner doesn't detect this version of "bitrot". It is for detecting the sections of the disk where the bit state has gotten "fuzzy". The read of these sections can cause the disk's firmware to fix them, or to remap them internally, before they become inaccessible. Or to let you know when they are no longer accessible. So it solves different problems. However, I'd still recommend the ECC memory, in general. If it's not much more expensive than non-ECC, then no reason not to.

The other potential issue is with NTFS permissions. But this applies to any drive. And if you run into issues, just reset the permissions.

This can happen, if non data is placed on the disk in question. You can force the creation of the folder by adding a file placement rule to place data on the drive in question, and then add a single file to the pool. That should create the folder on the disk in question. StableBit DrivePool places the actual files on the underlying disks. So any given file needs to be able to fit on the disk in question. It doesn't break up the files, in any way.

Not really. The option really is "Once it gets to the point to balance, then balancing everything we can". Basically, the same answer as above, actually.

No. You'd have to move the disks from one system to another. My guess is that this is an issue with the drive's partition table. Some external encloses "modify" it so that the disks will only work in the enclosure. You can repartition the drive once it's plugged into the system directly, but you'd lose all the data that is on that drive.

Yes, and no. The SSD Optimizer can, but it does so by setting the real time placement limiters, and turning a drive into a write cache for new files. It may be close to what you want, though.