bland328

Members
  • Posts

    105
  • Joined

  • Last visited

Everything posted by bland328

  1. Ah! I think this explains the hard link struggles I'm experiencing when attempting to de-dupe files using rmlint. So, @limetech, do I understand correctly that: If I make a hard link in a User Share (e.g. in /mnt/user/test), it will indeed make a hard link on the Device (e.g. in /mnt/disk3/test), but... The shfs FUSE file system generates "pseudo" inode values, and therefore... There's no way to detect hard links in User Shares--I can only detect them by examining the inodes of files on Device(s)?
  2. I'm also using 6.6.6, for the benefit of anyone else reading this.
  3. Oh, I get your point. However, I did that long ago, since I just hate for things like that to mismatch—so, my VM name (the <name><name/> in the XML) and directory name do match, and I still run into this problem. Odd.
  4. In this 2017 post: @llonca13 says: That sounds promising! Though I haven't been able to find the original @SpaceInvaderOne (formerly @gridrunner) post referenced, I have attempted to change the slot number in the XML as follows: ... <address type='pci' domain='0x0000' bus='0x01' slot='0x05' function='0x0'/> ... However, when I make just the single-digit change and then click Update in the XML View, I immediately get this error pop-up from Unraid: VM creation error XML error: Invalid PCI address 0000:01:05.0. slot must be <= 0 Any thoughts on what I'm doing wrong? Do I need to somehow update the definition of PCI bus 1, or otherwise "create" slot 5? Or am I just all confused because @llonca13's original post above is only about passing through a physical NIC, and wouldn't apply to a virtual bridge?
  5. Aside from an Ethernet issue I haven't solved, I've got what seems to be a rock-solid Mojave VM running under Unraid 6.6.6. My Ethernet issue is simply that often (but not always) after I boot the VM, I don't have Ethernet access. In this case, System Preferences > Network says that my Ethernet cable is unplugged. If I then issue $ sudo ifconfig en0 down && sudo ifconfig en0 up from the Terminal, Ethernet starts working, and I've never seen it then fail again until I reboot the VM. My network configuration looks like this (MAC address censored): ... <type arch='x86_64' machine='pc-q35-3.0'>hvm</type> ... <interface type='bridge'> <mac address='99:88:77:66:55:44'/> <source bridge='br0'/> <model type='e1000-82545em'/> <address type='pci' domain='0x0000' bus='0x01' slot='0x00' function='0x0'/> </interface> ... Any thoughts on how I might fix this? Or, at least, gather more data?
  6. I see where you are coming from there, and I have indeed changed the name of my VM at some point, but the current situation is that I have manually selected the correct img file, and have to re-manually select it every time I attempt to submit any changes made in Form View. So, feels like a straightforward GUI bug, but I'm not sure exactly what is triggering it. Or, do I misunderstand your point, @user2352?
  7. Thanks very much, @SpaceInvaderOne! I will. Also, I'm a little embarrassed that I figured out the source of the problem--I hadn't disabled sleep under Mojave. So, though I'd like to think that mouse support would survive sleep, I'm not shocked that it doesn't, and setting sleep to Never fixed my mouse woes. I hope this post helps some other macOS KVM noob
  8. tl;dr: macOS VM virtual mouse eventually stops responding. I have macOS Mojave 10.14.1 (I haven't quite figured out how to successfully update to 10.14.2 yet) working very smoothly, thanks to the excellent @SpaceInvaderOne guide at: I'm accessing this VM purely through VNC, and I'm passing through no USB ports, nor even any audio or graphics. My <input>, <graphics> and <video> XML looks like this: ... <input type='tablet' bus='usb'> <address type='usb' bus='0' port='1'/> </input> <input type='mouse' bus='ps2'/> <input type='keyboard' bus='ps2'/> <graphics type='vnc' port='-1' autoport='yes' websocket='-1' listen='0.0.0.0' keymap='en-us'> <listen type='address' address='0.0.0.0'/> </graphics> <video> <model type='qxl' ram='65536' vram='65536' vgamem='16384' heads='1' primary='yes'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x0'/> </video> ... and my usb-mouse and use-kbd XML looks like this: ... <qemu:commandline> <qemu:arg value='-usb'/> <qemu:arg value='-device'/> <qemu:arg value='usb-mouse,bus=usb-bus.0'/> <qemu:arg value='-device'/> <qemu:arg value='usb-kbd,bus=usb-bus.0'/> <qemu:arg value='-device'/> ... After some time, some number of VNC connections/disconnections (I sometimes use RealVNC viewer and sometimes the built-in "noVNC" VNC Remote), or some other factor I haven't considered, the mouse stops responding, but the keyboard continues to work. At this point, I can't figure out any way to get it going again, except to reboot the VM. The mouse always works after a fresh boot. Any thoughts on what I might try, or other information that might be valuable to help others help me? Thanks, all!
  9. Ah-ha! I may have found the problem: If I change the Primary vDisk Location from Auto to Manual, I can then Update the VM. But when I Edit it again, the Primary vDisk Location reverts to Auto.
  10. I'm also seeing this problem. I created a new Windows 10 VM today, and made no XML edits at any point. I now cannot successfully submit any further changes from the Form View—when I try, I get the "Updating..." text in the button, seemingly forever.
  11. Thanks again for a great and thorough response! I get the APFS bit--I think a lot of people were initially so confused that they managed to confuse other people with their postings on the topic. Interestingly, you say: .I'd sure like for that to be true, and perhaps under the right circumstances, it is, but none of my new SMB shares have been recognized by Time Machine, and even when I've used the macOS tmutil command along these lines... sudo tmutil setdestination -ap smb://[email protected]/timemachineshare ...I've had no luck. If you know of some magic, I'd love to hear it! I have managed to make a sparsebundle locally, using a command like this... hdiutil create -size 2000g -fs HFS+j -volname "Time Machine" ComputerName.sparsebundle ...then copy that out to an SMB share on my Unraid server, then mount that from the share, then hand the resulting mountpoint (e.g. '/Volumes/Time Machine') over to Time Machine using the tmutil command. That all seems to work, except it looks to me like Time Machine has then been fooled into thinking it is backing up to a local disk. Which is either a problem, or it isn't. I don't know if Time Machine uses any more fault-tolerant approaches or different read/write patterns when it knows it is dealing with a network destination, so I'm not sure whether to consider this approach a high-quality solution or a dangerous stunt. As for your question about why I'm trying to get away from AFP--well, at the moment, I'm honestly not really trying to solve a particular problem. So perhaps I shouldn't even be attempting it! But, since I was making some changes and interested in dropping (again, for no particular reason) dependence on that aging protocol, I figured I'd take a shot. That all said, it looks like Samba version 4.8 has robust Time Machine support, as discussed here: https://macosx.com/threads/smb-samba-for-time-machine-backup.324958/ and a variety of places online, thanks to the addition of (what I understand to be) new Global configuration options such as: fruit:time machine = yes fruit:advertise_fullsync = true I don't know if there's an easy/safe/friendly way to inject Samba 4.8 into Unraid 6.5.3, but given that Unraid 6.6.0-RC1 already has Samba 4.8.4 in it, I imagine I'll just be patient. Unless, of course, someone wants to tell me about that easy/safe/friendly way to inject Samba 4.8 into Unraid 6.5.3 😅
  12. +1 I'm also trying to get Time Machine working over SMB between Unraid 6.5.3 and High Sierra, so if anyone has insight, it's much appreciated.
  13. First of all, that's a great response, and I really appreciate it--it does give me some more insight. Secondly, I feel sorta dumb about my question, because I was a little short on sleep and not properly considering (or remembering ) that only my cache drive is BTRFS--my actual data drives are XFS, so I think that makes my question pretty much moot. That said, I'm currently drowning in hints and tips here and there about getting Time Machine working over SMB between HIgh Sierra and Unraid 6.5.3, so if you have any more to say about "increasing support for SMB," I'd love to hear it. It looks to me like Samba 8.0 tidies Time Machine issues up quite a lot, but I realize Unraid doesn't yet include it, so I'm open to suggestions. And perhaps I should be starting a new thread EDIT: In fact, I have jumped over to join the conversation at:
  14. Should I disable Copy-on-Write (CoW) for a Time Machine share on a BTRFS disk? Given that I can't conceive of how or why I'd benefit from any BTRFS-level snapshotting on Time Machine file sets, and given the admittedly little bit I understand about potential CoW performance penalties, I suspect I should disable it. Or, at the very least, that there's probably not a downside to disabling it. Can anyone with a deeper-than-mine understanding of Time Machine and/or BTRFS Copy-on-Write provide any insight? Thanks very much for any help!
  15. I humbly suggest/request an unRAID-friendly Docker based on the Raspberry Pi Server (piserver) at: https://github.com/raspberrypi/piserver piserver is an official project of the Raspberrry Pi Foundation (https://www.raspberrypi.org/blog/piserver/), which can run on a Raspberry Pi, but is primarily intended to run on a Debian-based OS on an x86 computer. Its purpose is pretty slick: To provide easy-to-manage network-based bootability to Raspberry Pis without SD cards (or with minimal, read-only, corruption-proof SD cards for older models). I took a shot at building a VNC-friendly Docker based on the brilliant https://hub.docker.com/r/jlesage/baseimage-gui/ base images created by @DJoss, but my Docker-fu is still too weak to overcome a couple D-Bus struggles...and maybe I didn't really need to be using the bigger Debian baseimage-gui in the first place! I painlessly built it and got it running on a local Debian 8 VM, but just can't...quite...make it a good Docker. Many thanks to anyone who might make this magic happen.
  16. @Squid, this plug-in is fantastic! I love the ability to execute a script both At Startup of Array, but I now I have a script that modifies /root/.bash_profile, and I really need to execute it at boot time, regardless of whether the array has started yet, and even if the array is configured so as not to automatically start at all. So, I'm requesting an "At Boot" (or "After Boot") scheduling option, if it is workable and ever fits into what you are working on. Thanks for your consideration!
  17. Fantastic. I took your advice, and got in on the Focus 550W Gold for $54.99 after rebate. Thanks very much for your help, @unevent!
  18. Thanks very much for the advice. Are you saying that 550W sounds like enough for my specific configuration?
  19. Finally, it is time to build a modest new box so I can do some light virtualization with unRAID 6 (but no gaming on this new box)--I'm loathe to admit I'm still running unRAID 5.x, but pleased that it has served me well for so long! I'm piecing this together: ASUS M5A97 R2.0 AMD FX-6300 32GB DDR3 1600 RAM IBM M1015/LSI SAS9220-8i PCIe controller (1) SSD (2) WD Green drives (10) WD Red drives The number of drives is unlikely to go beyond 13, since I'm putting this in a Rosewill RSV-L4412 12-bay hotswap case. Some back-of-envelope calculations (that I'm not particularly qualified to be doing) suggest I might need as much as 250W sustained if everything is running full-out, so my gut is telling me to go with a Seasonic FOCUS FM 450W 80+ Gold (SSR-450FM) or Seasonic FOCUS Plus 550W 80+ Platinum (SSE-550PX). Or, how about a Corsair TX650M 80+ Gold, which I can get through Newegg for just $65 after rebate right now? Any thoughts on whether I'm even close to reasonable on my PSU thoughts are greatly appreciated!
  20. It is set to automatically reinstall upon boot in unMENU, and openssh-5.1p1-i486-1.tgz.auto_install is right where it should be in /boot/packages . If I telnet in after boot and manually run openssh-5.1p1-i486-1.tgz.auto_install , ssh then works. I assume it is probably this old issue: http://lime-technology.com/forum/index.php?topic=12706.msg123281#msg123281 but since that post is 2+ years old and I don't see anyone else talking about this, I'm checking in here before altering auto_install to cure it. Can anyone please advise? Thanks very much for the help.
  21. Figured I'd have to reboot the box or restart a service or some such thing, but not needed...WeeboTech's suggestion worked perfectly (under admittedly limited testing) and immediately. Slick! Thank you. Edit: For future visitors, I'll point out that nothing has been renamed--instead a SMB "boot" alias has been created that also points to the boot drive.
  22. Thanks, gentlemen! I agree that better finger (or brain?) training is probably the best answer...but now I'm very curious about whether the smb-extra.conf approach will do the trick. I'll try it and report back...
  23. It is possible/practical/not too terrible of an idea to rename the "flash" share as "boot" under unRAID 5? I ask only because when I bounce back and forth between administrative and development tasks via SSH and SMB, I'm forever getting "\flash\" vs. "/boot/" wrong on one side of the fence or the other. I know better, but apparently my fingers don't Thanks for any thoughts on this. B
  24. I'm using unRAID 4.7 Plus, and would like to create some folders or shares in which Windows clients have permission to read files and create new files, but not to delete or modify existing files. Can someone give me a push in the right direction on this?
  25. I certainly didn't wait longer than five minutes, so if it happens again, I'll give it more time to come back on its own. Has anyone else experienced this?