Christopher (Drashna)

This is for StableBit Scanner, specifically, correct?

Just a heads up, the balancing is done with a background I/O priority. This means that if the disk(s) are being used for anything else, then it will cede priority to those other tasks. So if you do have something like backblaze running too, it may run slower. Also, even if the disk that is being written to is fast and not busy, it is still limited to the read speed from the other drive(s).

Well, it's because different things are being tracked. The Volume ID is how Windows identifies which volume gets which letter. And this shouldn't change. As for identifying the disks, that depends on information about the drives, from the USB controller, which could change. It may be best to to open a ticket at https://stablebit.com/Contact for this. But, specifically, if you can run the StableBit Troubleshooter twice, once before rebooting, and once after rebooting, where the information has gotten mixed up. https://wiki.covecube.com/StableBit_Troubleshooter And make sure the "Include disk details" option is enabled (it should be by default for StableBit Scanner, but just to make sure). And use either the Contact ID from that ticket, or "27507" for the contact ID, in StableBit Troubleshooter.

The command line option that shane mentioned is what I would recommend, as it's probably the simplest option if you're comfortable with command line.

lol. If it makes you feel any better, I'm definitely not immune to that. I've questioned stuff not working or working oddly, only to realise .... I am/was the problem. Too many times.

What Shane said. You can't replace the system disk. However, you can add the system disk and the E drive to the pool (which appears that you have done. And you can hide the E:\ drive by removing it's drive letter. But because of how the software works, and how booting and "boot drivers" work, we are not able to support booting for the pool drive.

If you're seeing a lot of drives coming up with errors, and suddenly, there is the "burst test" option. It may be worth running this on one or more of the drives. This reads a very small block of data, but reads it over and over and over. This way, it should not be reading it from the disk, but from the drives cache. BUt this should saturate the bus/communication. If you see errors coming back from this burst test, then there is some sort of communication error with the drive. If there is a connection error, it could be with the drive's pcb, the cables themselves, the connectors for the cables (eg, loose fit), the SATA controller, drivers, etc. If that's not the case, then could you open a ticket at https://stablebit.com/Contact

Some of the older versions have some .... issues with the update notifications. Newer versions, we've completely overhauled the code for handling updates, and should be more stable and reliable (especially when using StableBit Cloud).

I'm glad to hear it!

The 4077 release should also include the fix. If it's not, could you uninstall it, and reinstall it and see if that helps?

Yup, something broke on our end. This should be fixed now, and updating should get the fixed version.

Okay, thank you for letting me know. And if that's the case, one of the recently UI changes may have accidentally broken something in the code handling the UI translation or detection.

Data recovery can cause this to happen, yeah. As for fixing this: https://wiki.covecube.com/StableBit_DrivePool_F1655 This should cover everything you need to do.

Thank you for the investigation! That definitely helps a lot! And there looks like there were UI fixes in 1600, so likely something broke there, accidentally. That said, I've created an issue for this: https://stablebit.com/Admin/IssueAnalysis/28956

StableBit DrivePool should pick up the lanaguage settings from the OS and use that. Is Windows set to use japanese?

If you're not seeing any issues, this is most likely fine. Generally, these are handled issues that shouldn't have any impact on the use and performance of the software. Specifically, this looks to be related to balancing, and the balancing ratio. And likely, there is some missing data or stale data that is triggering this. So very likely that this is completely harmless. If you are experiencing issues, or are extremely concerned about this, please open a ticket at https://stablebit.com/contact

Have you tried the unsafe Direct I/O setting, as well as the "nowmi" option? Both of these are in the advanced settings: https://wiki.covecube.com/StableBit_Scanner_Advanced_Settings#Advanced_Settings

Yup! The only caveat is that you may need to mess around with permissions. But that's not a StableBit DrivePool thing, but more of a general Windows thing.

Unfortunately, no, not currently. The Windows apps (including XBOX apps) do some weird stuff that isn't supported on the pool. And adding support for it isn't trivial (it's likely some very low level file system implementation that may be very difficult to implement). That said, you *could* use StableBit CloudDrive and the local disk provider to create a drive on the pool, and install the games to that. Because it's emulated at the block level rather than the file system level (like StableBit DrivePool), it shouldn't have the issues with the Windows/XBOX Apps

For remote management, if you use Windows Server 2019 or newer, you can use "Windows Admin Center" which gives you .... pretty much full access over the system remotely. It's pretty freaking awesome, actually! https://www.microsoft.com/en-us/windows-server/windows-admin-center I HIGHLY recommend this (though it won't install on domain controllers). For file shares, you can do that with Computer Management, it's just a bit more of a pain to setup. As for backup, I also recommend Veeam. If you want centeralized backup, the Community edition of Backup & Recovery is ... a bit heavy on the initial install process, but works very well. https://www.veeam.com/products/free/backup-recovery.html?ad=menu-products-portfolio-free

If you don't see the red "real time placement limiters" arrows on the drive in question, then it may be just that the drive has been evacuated, but doesn't have enough free space to be used for new files (eg, it has less free space than some of the other drives in the pool). This is because the balancing doesn't forcibly equalize the usage (though there is a balancer that you can enable to force that behavior)

I'd have to dig, but there have been a number of driver related fixes recently, so this may work better, now. https://dl.covecube.com/DrivePoolWindows/beta/download/StableBit.DrivePool_2.3.9.1646_x64_BETA.exe

If you're using the 1.x version of StableBit DrivePool, then it's in the daskboard settings, under the StableBit DrivePool section. There should be a "Manage license" link at the bottom of the page. Click on that, and you have the option to deactivate the license. If you're using the 2.x version, in the UI, there should be a gear icon in the top right corner. Click that and select the "Manage license" option. From there, you should be able to deactivate the license. As for running Windows 10 as a home server, it depends on your needs. You can create file shares on the pool via Explorer or Computer Management. But settings get a bit tricky. However, if you're using a media server software, then this may not be necessary, and you'd just need to point that software at the data on the pool. Also, note that support for Windows 10 ends next year (Oct 2025).

Don't copy the PoolPart.xxxx folder, just the contents in it. There are some super hidden information attached to the folder which copying over may cause issues. And if the J:\ drive is already in the pool, copying the contents of F:\PoolPart.xxxx to J:\PoolPart.yyyyy should be fine. In either case, make sure that you remeasure the pool (from the "manage pool" button/menu in the UI) once you've moved the files over.

The issue is that you can have x3, x4 or higher. And you can have it different on each folder level. Calculating the correct size would be very expensive, in terms of resources. Especially as some of this has to be done in the kernel, where time is critical. The choice was to report the "raw" size rather than the adjusted size, to make things faster and simpler. Unfortunately, this does mean that the size appears to be misleading.