• Posts

  • Joined

  • Last visited

Everything posted by TreyH

  1. I’ve managed up till now passing through all my USB ports via IOMMU groups except the one the boot flash was plugged into. But adding a new device has scrambled this, and I don’t know how to proceed. I’ve been operating fairly happily for four months now with Unraid as my home NAS and as the hypervisor for my Win 10 gaming/dev machine. Few problems, none on the Unraid side, just on the Windows side. But getting it set up right required Putting my boot flash on a USB-A → USB-C adapter and plugging it into my mobo’s lone USB-C port Passing through all the USB busses except for the USB-C one. This meant I couldn’t use USB-C devices in Windows, but that wasn’t a big deal. The alternatives were to only passthrough two USB 2 or three USB 2.1 ports plus a USB-C port, and that just wasn’t sufficient for my Windows installation. I’ve now run into a more difficult issue. I have an Intel WiFi/Bluetooth PCI card that has a USB header—basically, the WiFi goes through PCI but the Bluetooth goes through the USB (though it can’t be powered via USB, so the card has to sit in a PCI slot even if you aren’t using it for WiFi). This was weird, but fine—the USB bus that saw the Bluetooth adapter was one of the ones I was already passing through. But, I needed to move the card to a different PCI slot to accommodate a new SATA HBA since I was adding a couple new disks to my pools—the Intel wireless card was in a full-width slot even though it was just an x1 card. Unfortunately, when I moved the card to the free x1 slot, the Bluetooth’s USB device moved, too—to the same bus the USB-C port is on. I don’t understand this—I’d think that the header would determine the bus, not the PCI slot, but I don’t really understand how this odd hybrid PCI/USB card works. I’d already broken up the IOMMU groups as much as possible before finding my only choice for the boot flash was the USB-C port. So what now? Are either buying a Bluetooth card that doesn’t require USB, or else an actual Bluetooth USB dongle I can plug into (a hub attached to) one of the currently passed-through buses, my only options (other than not using the new HBA)?
  2. I created a small `C:` drive vdisk, installed Windows 10, installed some apps like Steam that allow further installation on network drives, used shares for that, and no problem. But now I need to install some things from the Windows Store (some of which are quite large). Unfortunately, if I go into Settings > System > Storage > Change where new content is saved: I'm not allowed to change any of these--they are all locked to the C drive only. I have mapped drive letters for my shares, and they work as storage locations for other contexts, just not this one. If there's a way to change the install location of new apps without using this screen, that's fine too, I just don't know what that is? Thanks for any help!
  3. Is there a release notes version giving 6.9.0-rc2 → 6.9.1 changes? With a year of betas trying to figure out what changed from rc2 by teasing out what’s already changed from 6.9.0 → ~30 builds... → 6.9.0-rc2 isn’t straightforward! I’m probably just missing the obvious place to look... Edit: Uh, all those numbers are one off, obviously, in one direction or the other. No coffee yet, sorry. Y’all know what I mean, just the changes from rc2 to the brand-new stable release.
  4. Could you elaborate on what “this” is? Your link is to the release notes for 6.9.0-beta25 and it’s not clear what “the emulated version” refers to in that context. What did you change, settings-wise?
  5. I’m a newbie so didn’t have a previously-performant Windows 10 VM, but for me I tried to set mine up first with a GPU vBIOS grabbed from the online database—which didn’t work at all—and then pulled off of my GTX 1080 with the vBIOS tool. The latter worked but was noticeably _slower_ than not passing in a vBIOS at all. Insofar as the changes to the VM manager I’ve read in the changelog since 6.8, one thing I might check is ensuring you have a) CPU isolation and b) no “odd hyperthreads”—cases where you’ve assigned one hyperthread of a core and not the other(s).
  6. Oops, sorry—I missed this message last month or i would have replied sooner. It turns out the checkbox is unclickable, but the particular colors used for a dimmed checkbox compared with an active checkbox on this device (Safari/iPadOS) wasn’t clear. Sorry for the confusion.
  7. I just executed a command I very much did not want to execute, because it turns out that for some reason ⌃C gets interpreted by the Webterminal’s shell as a ⏎, even though NAS# stty -a intr = ^C [...] is this a known issue? I’ve tried it on different devices and it seems to work this way on all of them. I’ve never seen this before on any Linux distro; universally (until now) I’ve found ⌃C aborts the current command and gives you a new clean command line, it doesn’t execute it! Can anyone shed light on what’s going on? Can this be worked around (`stty` was already set correctly as I wrote, but altering it didn’t seem to matter)? Should I submit a bug report? (addendum: I just tried other commands that tell me what the shell is seeing when I press ⌃C, and it’s seeing a ⌃J. But ⌃J and ⏎ (aka ⌃M) also send a ⌃J. Sooo.... weird.) (This is on 6.9.0-rc2.)
  8. Yes, definitely the usbip-gui package. Hmm.... strangely, after trying to open its settings from the Plugins tab, I tried to visit it via Settings -> USB Settings, which just took me to /USBIP_Settings with the two options (Enable USBIP and Enable USB on Dashboard) both "Disabled", but doing this--just visiting that page without making any changes--resulted in the USB tab appearing in my interface, with that URL /USB. I just tested it to mount a USB flash drive in Windows 10, and it worked fine. So I'm not sure what happened here to delay its working, but it does seem to be working now.
  9. Does this work on 6.9, or without the usbip kernel extensions installed? I installed this via Community Apps and when I click its settings, it takes me to the URL /USB, which is totally blank. I haven't installed any of your other plugins.
  10. For the record, this may have been a “Windows 10 is still settling with updates, etc.” issue—starting about 12 hours after I posted the earlier message, my keyboard has been waking up the display. In that time, I: activated Windows, installed more virtio drivers, installed the driver for my LG monitor (which I had to download directly from LG’s support site; Windows couldn’t autodetect the monitor as anything but a general plug-and-play monitor, which gave me 800 × 600 resolution on a 3440 × 1440 screen), and installed essential updates (and reboots as needed). So if you’re having this issue, you may want to try this.
  11. In the 6.9 interface, it’s offering me the chance to bind the USB port my Unraid boot flash is on. Is this normal and correct behavior? What will happen if I try it? Will I be able to see the flash device both in vfio vm’s and in Unraid, or will Unraid keep that one to itself? (I have worked on Linux kernel internals this present century—just barely!—so I’m guessing if this has something to do with the vagaries of USB Storage, such that the flash device can remain mounted by Unraid while the underlying USB port appears present but unavailable for any functions to the VM, then this would seem to not cause any problems for the Unraid kernel’s VFS—except pulling the flash while Unraid is running wouldn’t allow it to be remounted by Unraid if it were plugged back in to that port and a VM could grab it. But I assume that removing the boot flash during operation isn’t safe anyway.)
  12. (See below for possibility that this was transient.) I have Windows 10 with a GTX 1080 and USB keyboard and mouse passed-through. I went through and changed all the power settings (never sleep, etc.), but because my display is very bright even when displaying all black (it's a 40-inch ultra wide, so the LED's have a lot of bleed through) I enabled display sleep. It seemed to work, as the screen turned off but Windows network services were still available from other devices... but then I couldn't get it to wake the monitor back up, via keyboard or mouse. From other threads I've searched, it appears this is expected behavior with the only real solution being to get lucky passing through USB controllers (I've already passed through the ones Unraid itself doesn't need). But here's the thing... just playing around, before I did all this, I used the VM manager to pause the VM. I then used a different device to unpause it. When I came back to this machine to look for docs that might help, the old VM manager was still there, still showing the status as "paused". So just for giggles I clicked it and tried to resume—and it worked, the display blinked back to life and I could log back into Windows! But the other device didn't have a 'resume' because it didn't think the VM was sleeping. So... there is absolutely a way to wake from display sleep, maybe not with the keyboard, but it can be done. The Web UI just doesn't expose it in normal operation, since it doesn't show 'resume' when it doesn't think the VM itself is asleep. I'd really like to be able to hit that sleep button and have the monitor stop sucking up watts and bathing the room in light when I'm not using it. Even if I have to use another device to re-wake it. Does anyone have a suggestion for how I might be able to do it? (I guess I could always pause and then resume again, but it seems silly since obviously the pause step isn't strictly necessary.)
  13. How do you pass those through to a VM, then? (Would I have been unable to do the same thing to /dev/sde, if I hadn't clicked the settings and switched on passthrough on the disk that did have settings?) I'm not sure why it thinks this disk and only this disk has a filesystem whose type is "isw_raid_member", when it's actually an NTFS filesystem striped across two other disks as well: root@NAS:~# lsblk -fM NAME FSTYPE FSVER [...] sdb isw_raid_memb 1.3.00 sde isw_raid_memb 1.3.00 └─sde1 isw_raid_memb 1.3.00 sdf isw_raid_memb 1.3.00 [...]
  14. Hi, I'm just setting up my first array and VM's planning to migrate data from my old pre-Unraid disks. In this thread in another topic, I was advised to passthrough all three 3 TB drives in an Intel RST RAID 5 set, but as shown below, only one of the there (marked in green) has the gears while the other two drives in the set (marked in red) do not: How do I pass them through? Addendum: Sorry, just as I sent this I got a reply in the other thread: I was under the impression that "normally" as of the 6.9 UI is via the gears? The passthrough switches have gone away, including for the lines without gears, as you can see in the screenshot. I wonder what's the difference between Dev 2 (with the gears) and Devs 3/4 (without), anyway?
  15. Thanks, but if you look at the screen grab above, only one of the three 3 TB drives has the gear icon—what do I do about the other two?
  16. Thanks, that makes sense, but following the guide for 6.9, I went to Tools → System Devices, and all my disks—my array disks and these Intel RST disks—are in the same IOMMU group. Which makes sense, as I don’t have a separate storage card but the built-in mobo one. So how does one go about hiding devices rather than the entire IOMMU group?
  17. So I tried this, first with an UEFI ArchLinux boot USB to verify I could still mount and access the data on my RAID 5 set okay (I could), and then booting 6.9.0-rc2 (created with the Unraid USB Creator with UEFI checked) on the same USB port. The Intel RST status on the BIOS screen showed everything green, but when I got into the UD settings, I got this (with the green highlighting the three drives in the RAID 5 set): They showed up as separate drives (which, in itself, isn't necessary worrying, since ArchLinux sees them as four drives, each of the 3 raw 3TB disks, and then the 6TB logical disk), but only for the raw disks, not the RAID set. Only the one marked even had a "Mount" button, and it didn't work. The log messages were: Feb 14 10:20:03 NAS unassigned.devices: Adding disk '/dev/sde1'... Feb 14 10:20:03 NAS unassigned.devices: Mount drive command: /sbin/mount -t isw_raid_member -o rw,auto,async,noatime,nodiratime '/dev/sde1' '/mnt/disks/WDC_WD30EFRX-68E************RTHYY' Feb 14 10:20:03 NAS unassigned.devices: Mount of '/dev/sde1' failed. Error message: mount: /mnt/disks/WDC_WD30EFRX-68E************RTHYY: unknown filesystem type 'isw_raid_member'. Feb 14 10:20:03 NAS unassigned.devices: Partition 'WDC_WD30EFRX-68E************RTHYY' cannot be mounted. I went into the Linux console and checked, and unlike on Arch Linux where you can find these under /dev/md*, no /dev/md* devices exist on Unraid. Do I need to pass some kernel boot options or something?
  18. (tl:dr: I’ve boldfaced my questions below so you can see what I’m asking without having to read all the background info.) I am in the process of trying to use Unraid to replace a bare-metal Windows 10 Pro install. I used it as a (light-duty) Windows development workstation and for some gaming with a GTX 1080. I also used this Windows host OS for an SMB server for other devices in the house, as a Hyper-V hypervisor for (very light-duty) Docker images, and a Ubuntu development workstation. Most of my “real work” was in Ubuntu, so the heaviest load was when I was interactively using Ubuntu (usually via Mosh/SSH—I didn’t use it for graphics) or when I was Windows gaming. My Windows boot disk installation is now corrupt beyond repair after a botched Windows Update. OS damage only, though—no physical or filesystem damage on the SSD I can detect with Windows Rescue, SystemRescue, and some third-party tools—so I intend to use the Unassigned Devices plugin to move application data into the VM. I have some questions, which I don’t expect any one person to be able to answer all of, but if you know the answers to any, I’d appreciate the advice: Can I use my original OEM Windows install DVD? I’ve read about various methods for using a DVD/Blu-Ray writer like I have to import/export data to the array, but can I use it to install Windows, or must I download an ISO? (Since it’s a 3-year-old disc, I don’t actually know how much network-download time I’d save from doing this versus starting with a more recent ISO—so if it doesn’t affect licensing it may not be useful to bother.) Can I transfer my Windows 10 Pro license to the VM, or will I have to purchase a new one? Right now my husband and I both use the Windows side interactively, tied to our online Windows accounts, with Microsoft password and PIN “security” for login. Can we continue to use our online Windows accounts with Unraid—either through some mechanism to use Windows authentication, or just by setting up users with the same names and passwords and keeping them sync’d up? Assuming there’s no mechanism to make Unraid play with Windows auth (I couldn’t find one but maybe my search terms are just bad), is there a guide to how to make the usernames and passwords match? Windows usernames (with all their extra backslash cruft beyond the simple “username”) and access control totally befuddles me—I’m a multi-decade Unix/Linux guy, and get stumped by this “easy” Windows stuff. Even though we have separate accounts and data, we haven’t been terribly hygienic about shared files—we just are both members of Administrators. Just from reading and not yet doing, it seems we can still do roughly the same, and maintain separate file ownership but still share some files—and perhaps do even better with Unraid by making certain folders automatically share with different permissions? We have a large LG ultrawide display attached to the GTX 1080. Assuming I successfully pass it through (and I was able to in a test install last week), there are two issues with this display: The monitor has a built-in USB KVM switch that automatically routes the USB devices plugged into it to the system the display’s input is set to (by using alternate-mode USB on DisplayPort or HDMI, or a USB A-B cable assignable to some or all ports). We do use this feature a lot, to switch the monitor from this machine to laptops and Raspberry Pis from time to time so these generally-headless devices can be controlled without a lot of unplugging and replugging. Can I expect my monitor’s USB KVM switch to just continue to work? We only attach devices that make sense to move around with the display—currently, a Logitech webcam, a webcam light controllable with a Windows or Mac app, and a keyboard with its own USB 3 bus through which a mouse is attached. Before with Windows 10 bare-metal, the iGPU went unused and disabled in the UEFI config. Will I want to enable the iGPU for Unraid console access, and attach it to another one of the display’s ports in case the network is down? Or will this cause issues if the server reboots unexpectedly because UEFI will probe the display, find it’s not connected (since it will almost certainly be set to use the GTX’s output at the time the server goes down), and then grab the GTX such that it can no longer passthrough? My motherboard doesn’t support USB passthrough on the iGPU HDMI port. That’s what the display’s USB A-B cable is for—assuming I do have the iGPU set up as a console, can I arrange things on the Unraid side so that the keyboard and mouse will connect correctly via this USB-to-USB cable when I switch to that input? This display is bright. If it unexpectedly wakes up in the middle of the night, it sometimes awakens me, even bouncing around a hallway and into my bedroom. That, I can’t do anything about—but I do want the display to properly sleep after an idle timeout with the Win 10 VM is active. Will that be a problem? Is it as simple as setting display sleep settings in Windows as if it were still the bare-metal hypervisor? Finally, a couple slightly-related questions, about my old Ubuntu VM on Hyper-V, which as I mentioned is my real workhorse OS. (Note that because of my corrupt Windows 10 install, I currently have no way of booting it—so I can’t do any preparatory steps on Linux.) Can I just migrate my Ubuntu Hyper-V VM wholesale to Unraid? Whatever the answer to #1, if I install the qemu-ga-x64 package, can I make Unraid automatically hibernate (rather than shut down) Ubuntu gracefully when Unraid is restarted? Can I mount the VHDX file (with the Ubuntu VM quiesced or not yet created, I mean)? This thread suggests the answer is no, but since it’s been a year, perhaps something has changed? There are lots of posts here asking about installing libguestfs, but the most recent is in 2018 and none have replies indicating if someone was successful in getting it to work.
  19. I have a network-based backup (though apparently—as I discovered—I can’t rebuild the boot OS from it, unfortunately, the data files are all there and fine from spot-checks)—but building a new installation from scratch based off of it would take many days (and be costly, as I’m on a storage plan where I pay for bulk transfer). I would truly rather use the media I already have installed here. That said, someone on the official Discord has suggested that all I need do is change the Unraid boot USB to UEFI rather than legacy, and then UD will address the RAID volumes fine? This would make sense, as the same was necessary for SystemRescue and the Windows Recovery USB’s to mount them. I will give this a try tomorrow and report back if that’s all it is—though if someone reads this before that and knows this to be dangerous, I’d appreciate the warning.
  20. I think I may have made a very costly mistake... I was aware that Unraid cannot address disks that use Intel RST RAID. But I thought that just meant it couldn’t use them directly, when I’m now reading that it needs to be turned off in the BIOS/UEFI? That truly messes up my plans to use the Unassigned Devices plugin to migrate my RAID 5 data into Unraid, after buying enough new SSD’s and HDD’s for the purpose that I thought it “should” work.... The background: my workstation recently had a catastrophic unrecoverable Windows update error. I decided to capitalize on the need to reinstall to shift to Unraid with a Windows VM. (I used that Windows install as my graphics workstation, and also as my Hyper-V hypervisor to run NAS, Docker images and an Ubuntu VM—so not having these other functions go belly-up whenever Windows crashed was another driver to go to Unraid, though Proxmox etc. might have worked for this need.) While I’ve been unable to recover my Windows (10 Pro) install, or to do a non-destructive repair/recovery install, I can still access the filesystems, using either Windows Recovery or a live CD such as SystemRescue. I have approximately 5 TB of data, split between: Boot “disk”: Two Samsung EVO 1TB SSD’s in an Intel RST RAID mirror (approximately 900 GB used) Data “disk”: Three WD Red 3TB HDD’s in an Intel RST RAID 5 parity (approximately 4 TB used) So, using spare drives I had around, I first unplugged all of the above, plugged in these random drives, and did a minimal Unraid install with a Windows VM, just to verify I could get Windows running with my GPU and other necessary peripherals passed through. This Unraid + Windows with virt passthrough test went fine. Thinking this was going great, I then bought enough new disks so that, I would have space in Unraid for all the data on my two RAID sets, with (I thought) plenty of room for the data—once I got it off the boot and data “drives”, I could reformat them and put them to other purposes (on the Unraid or elsewhere). This includes an 8 TB HDD I bought for Unraid parity, assuming that I would never need a > 8TB drive on this machine. So—I intended to use this migration process: Temporarily unplug these five disks, Plug in the new blank ones only, Set up Unraid, including shares and a Windows VM, Then plug those RST RAID disks back in, and use the Unassigned Devices plugin to migrate my data on those two logical “drives” into Unraid shares. (I worked most of my career in SRE engineering with big storage, so I was confident in my ability to get the data over, one way or another—whether using built-in or community-contributed Unraid tools, user scripts I’d write myself, or a guest VM to do the heavy lifting. I wasn’t worried about this part.) Once I had the data off and verified, remove the SSD’s to scavenge for other purposes, and reformat and add at least one of the three old HDD’s for additional storage headroom. (I planned to test running my Ubuntu VM, on which I do most of my development work, on an Unraid share, plus the other options offered by the UD plugin, such as a passed-through drive, and compare results before deciding what to do with the old disks.) But now, as I’ve done more research before getting started, I’ve read on here that Unraid shouldn’t even be run with RST active in my BIOS (UEFI, I mean)—and trying to boot Unraid with RST active and those disk sets plugged-in could result in data loss? I hope perhaps these warnings are dated? It’s possible to—with some care—break the boot RAID 0 mirror pair and just use one of the disks as an NTFS drive. So this I’m not worried about. It’s the data on the larger RAID 5 set that I’m stuck on—I don’t have any other machine or large-enough staging disk to get its data out of the RST logical drive without playing a musical-chairs game, rebooting back-and-forth between Unraid (with my RAID disks unplugged and BIOS set one way) and SystemRescue (with my RAID disks installed and BIOS set the other way), moving a chunk that can fit on the “extra” 1TB drive at a time. (My Ubuntu VHDX stored on the RAID 5—which I’d hoped to be able to use as the basis for a new Ubuntu VM—will take some special care, as it is, by itself, 1.5 TB—though I imagine I can build a custom rescue disk with the drivers necessary to “guestmount” a VHDX as a filesystem if I need to.) Is there any sort of workaround to this painstaking and dreadful-sounding process at all—short of buying yet another drive large enough to hold all this data at once, so I can use a Live CD OS to copy the data over to before turning RST off? (I’d probably have to buy another PCI SATA card at that point, too...) Could I safely run without parity long enough to use the 8TB drive as my staging area—since it could, in fact, hold everything on both logical drives with room to spare? It is by far the largest disk in the whole kit and caboodle, so it’s eventually destined to be the parity drive.
  21. (Boldface mine.) Just a small factual correction: Linux is perfectly capable of using Intel RST logical drives. Typically, each physical disk is seen as a normal block device (/dev/sd*), while the RAID drives are visible as logical drives: # lsblk -M -o NAME,FSTYPE NAME FSTYPE ┌┈▶ sda isw_raid_member ┆ ├─sda1 ntfs ┆ ├─sda2 ntfs ┆ └─sda3 ntfs └┬▶ sdb isw_raid_member ┆ ├─sdb1 ntfs ┆ ├─sdb2 ntfs ┆ └─sdb3 ntfs └┈┈md126 ├─md126p1 ntfs ├─md126p2 ntfs └─md126p3 ntfs Here, a mirrored pair across sda and sbd (note the type is “isw_raid_member”) has three partitions, available to mount via /dev/md126p{1..3}. However, just as on Windows, Linux can’t use these drives if AHCI rather than RST is selected in the BIOS or EUFI. So effectively the point is correct as far as Unraid goes.