SpaceInvaderOne

Yeah it finds my efi partitions ok. I find if i want to install kexts though, I must delete all folders in the clover kext folder except the "other" folder and put them in there. Ed

Glad you have got it working. I have not got the screen correctly showing more than 800x600. You can edit the resolution in the config.plist but it seems to mess up the screen for me. But you may have better luck? The reason 800x600 works is because that is the setting in the bios of the vm. It has to match the resolution of the osx when using vnc. If using gpu passthrough this isnt an issue. If you change the resolution in the bios aswell it may work. But you would have to do so every boot. I havent tried yet as i can never get into the bios screen quick enough!! I will have a look at this problem and see if i can figure a way around this. This isnt a problem on seabios so maybe boot with that. You can still use same disk image just setup another xml and point it to the vdisk but with the seabios settings and bootloader file.

Hi Peter sorry for late reply. Yes if you want to edit clover. Open the efi partition then goto clover and click open file then goto efi partition then open the config.plist in the clover folder. It doesnt find it itself for some reason. clover installer settings as below.

I followed the video exactly and the same has happened to me. I start the new OVMF VM and i get the same error message ("Guest has not initialized the display yet"). I'm running 6.2 stable. EDIT: Changed the VNC to Cirrus and now get the clover boot menu. However, the VM reboots (repeatedly) as soon as the Apple logo is displayed. This can happen when the cpu is not set to core2duo in xml. can you post your xml file here please (just remove the osk key from it though) Here is my XML -- <domain type='kvm' xmlns:qemu='http://libvirt.org/schemas/domain/qemu/1.0'> <name>MOMAC02</name> <uuid></uuid> <metadata> <vmtemplate xmlns="unraid" name="Windows 10" icon="/mnt/disks/VM02/MOMAC02/OSX-10.11.png" os="OSX"/> </metadata> <memory unit='KiB'>4194304</memory> <currentMemory unit='KiB'>4194304</currentMemory> <memoryBacking> <nosharepages/> </memoryBacking> <vcpu placement='static'>2</vcpu> <cputune> <vcpupin vcpu='0' cpuset='8'/> <vcpupin vcpu='1' cpuset='18'/> </cputune> <os> <type arch='x86_64' machine='pc-q35-2.5'>hvm</type> <loader readonly='yes' type='pflash'>/usr/share/qemu/ovmf-x64/OVMF_CODE-pure-efi.fd</loader> <nvram>/etc/libvirt/qemu/nvram/2397b50d-5f80-4c96-a331-1962018355cb_VARS-pure-efi.fd</nvram> </os> <features> <acpi/> <apic/> </features> <cpu mode='host-passthrough'> <topology sockets='1' cores='1' threads='2'/> </cpu> <clock offset='utc'> <timer name='rtc' tickpolicy='catchup'/> <timer name='pit' tickpolicy='delay'/> <timer name='hpet' present='no'/> </clock> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <devices> <emulator>/usr/local/sbin/qemu</emulator> <disk type='file' device='disk'> <driver name='qemu' type='raw' cache='writeback'/> <source file='/mnt/disks/VM02/MOMAC02/MOMAC02.img'/> <target dev='hdc' bus='sata'/> <boot order='1'/> <address type='drive' controller='0' bus='0' target='0' unit='2'/> </disk> <controller type='usb' index='0' model='ich9-ehci1'> <address type='pci' domain='0x0000' bus='0x00' slot='0x07' function='0x7'/> </controller> <controller type='usb' index='0' model='ich9-uhci1'> <master startport='0'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x07' function='0x0' multifunction='on'/> </controller> <controller type='usb' index='0' model='ich9-uhci2'> <master startport='2'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x07' function='0x1'/> </controller> <controller type='usb' index='0' model='ich9-uhci3'> <master startport='4'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x07' function='0x2'/> </controller> <controller type='sata' index='0'> <address type='pci' domain='0x0000' bus='0x00' slot='0x1f' function='0x2'/> </controller> <controller type='pci' index='0' model='pcie-root'/> <controller type='pci' index='1' model='dmi-to-pci-bridge'> <model name='i82801b11-bridge'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x1e' function='0x0'/> </controller> <controller type='pci' index='2' model='pci-bridge'> <model name='pci-bridge'/> <target chassisNr='2'/> <address type='pci' domain='0x0000' bus='0x01' slot='0x01' function='0x0'/> </controller> <controller type='virtio-serial' index='0'> <address type='pci' domain='0x0000' bus='0x02' slot='0x02' function='0x0'/> </controller> <interface type='bridge'> <mac address='52:54:00:00:20:30'/> <source bridge='br0'/> <model type='virtio'/> <address type='pci' domain='0x0000' bus='0x02' slot='0x01' function='0x0'/> </interface> <serial type='pty'> <target port='0'/> </serial> <console type='pty'> <target type='serial' port='0'/> </console> <channel type='unix'> <source mode='connect'/> <target type='virtio' name='org.qemu.guest_agent.0'/> <address type='virtio-serial' controller='0' bus='0' port='1'/> </channel> <input type='tablet' bus='usb'/> <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-gb'> <listen type='address' address='0.0.0.0'/> </graphics> <video> <model type='cirrus' vram='16384' heads='1'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x0'/> </video> <memballoon model='virtio'> <address type='pci' domain='0x0000' bus='0x02' slot='0x03' function='0x0'/> </memballoon> </devices> <seclabel type='none' model='none'/> <qemu:commandline> <qemu:arg value='-device'/> <qemu:arg value='usb-kbd'/> <qemu:arg value='-device'/> <qemu:arg value='usb-mouse'/> <qemu:arg value='-device'/> <qemu:arg value='-smbios'/> <qemu:arg value='type=2'/> <qemu:arg value='-device'/> <qemu:arg value='ioh3420,bus=pcie.0,addr=1c.0,multifunction=on,port=1,chassis=1,id=root.1'/> </qemu:commandline> </domain> Yes osx will not boot unless it thinks it has a core2duo in it. Add this above </qemu:commandline> <qemu:arg value='-cpu'/> <qemu:arg value='core2duo,vendor=GenuineIntel'/> also remove this line or else your mouse will not work at all. <input type='tablet' bus='usb'/> also remove these lines <timer name='rtc' tickpolicy='catchup'/> <timer name='pit' tickpolicy='delay'/> <timer name='hpet' present='no'/> should work fine now if it still reboots at apple sign then change this part <cpu mode='host-passthrough'> <topology sockets='1' cores='1' threads='2'/> </cpu> to </features> <cpu mode='custom' match='exact'> <model fallback='allow'>core2duo</model> <vendor>Intel</vendor> </cpu> but if you have added the first bit above the </qemu:commandline> you shouldnt need to worry about that. You could use the xml file in the description of the video in youtube if you want and edit that. Let me know if you have any other probs!

I followed the video exactly and the same has happened to me. I start the new OVMF VM and i get the same error message ("Guest has not initialized the display yet"). I'm running 6.2 stable. EDIT: Changed the VNC to Cirrus and now get the clover boot menu. However, the VM reboots (repeatedly) as soon as the Apple logo is displayed. This can happen when the cpu is not set to core2duo in xml. can you post your xml file here please (just remove the osk key from it though)

Ah ok I see. So you are running the clover bootloader but with your bios as seabios. If you installed clover to the efi when you did this you only need convert the xml. if not just put clover in the efi partition as my guide. Now goto a linux template and create a new virtual machine.Give it a random name doesnt matter what. Check machine type is q35. Don't assign any vdisks (set to none) then uncheck start vm on creation. Now goto edit xml of that vm can copy whatever is in between the OS tags will be similar to this. Its important to cut and paste this into a text file. <os> <type arch='x86_64' machine='pc-q35-2.5'>hvm</type> <loader readonly='yes' type='pflash'>/usr/share/qemu/ovmf-x64/OVMF_CODE-pure-efi.fd</loader> <nvram>/etc/libvirt/qemu/nvram/bd14d6e1-92b7-3c50-a16f-329d176d2d9b_VARS-pure-efi.fd</nvram> </os> But dont just copy and paste mine in the above example, because when creating an ovmf vm it it also creates a file in the /etc/libvirt/qemu/nvram folder. That file's name is based on the UUID of the VM (ie mine above is /etc/libvirt/qemu/nvram/bd14d6e1-92b7-3c50-a16f-329d176d2d9b_VARS-pure-efi.fd'). bd14d6e1-92b7-3c50-a16f-329d176d2d9b is the uuid of my vm. So now you can paste the part you put in the text file into your osx vm as that .fd file will now exist in the correct location. (Just follow the last couple of minutes of the second video if tim not clear here with what i say about starting and creating of vm templates to get the ovmf xml code.) After doing that you can delete the vm you just created with the linux template as that was necessary only to create that .fd file. But to be 100% "proper" you should really copy the uuid into the xml aswell to match but will work without. (I guess if you dont copy the uuid there could be problems if the uuid number was ever generated again on the server same as in the .fd file but i doubt it, unless 2 vms were using the vram file at the ame time?) So now you will be using clover but with ovmf not seabios. I have found performance better in my limited testing with geekbench 4. My gpu i passed through for osx has broken so I am awaiting another compatible one so not sure if any improvements there. Hope that makes sense.

Hi Peter I am not sure i understand what you mean? probably me being stupid lol! I dont understand how you dont have ovmf or seabios in your xml?? The xml must use one or the other seabios or ovmf. you must have something like this from archedrafts guide <os> <type arch='x86_64' machine='pc-q35-2.3'>hvm</type> <kernel>/mnt/cache/vm_images/enoch_rev2795_boot</kernel> # heres the chameleon bootloader # <boot dev='hd'/> <bootmenu enable='yes'/> </os> and as it doesn't have the below, its using seabios not ovmf <loader readonly='yes' type='pflash'>/usr/share/qemu/ovmf-x64/OVMF_CODE-pure-efi.fd</loader> <nvram>/etc/libvirt/qemu/nvram/41589327-89c4-099b-b658-e63dc0641e20_VARS-pure-efi.fd</nvram> <boot dev='hd'/> At first I tried creating a usb installer for osx on an 8 gig usb flash drive. Then i installed clover to the usb install flash drive. I didnt want to dd the image (but you could do) so I passed through the usb flash directly to the vm and booted from that. That did work so you can install the osx, to a fresh image that way without a previous osx vm. But then you need to install clover onto that or always use the dd image of the usb install media with clover on it. And in the past I find using the clover installer, problematic installing it into the efi on a vm. Also it takes so long to create the usb, dd the usb to an image (as not ideal using passthough ) and the dd image takes 8 gigs. Then the install of osx then installing clover on newly installed osx. I think for most people it is far quicker and easier to, 1. Install new osx using the normal way with the enoch chameleon bootloader. (or use an existing installed osx one if they have one ) 2. When disk utility partitions and formats vdisk during the install it creates the efi partition and recovery partition at same time. 3. So all that is needed is to open the efi partition after install using EFI Mounter v3 and copy the files there manually. (the files i linked in the video. I made by putting in q35-acpi-dsdt.aml, HFSPlus.efi and then edited the config.plist so they ready to go) 4. Now the clover is in efi of the osx disk image so no other images needed. 5. So all thats left is just adding the ovmf part to the OSX xml <loader readonly='yes' type='pflash'>/usr/share/qemu/ovmf-x64/OVMF_CODE-pure-efi.fd</loader> <nvram>/etc/libvirt/qemu/nvram/41589327-89c4-099b-b658-e63dc0641e20_VARS-pure-efi.fd</nvram> and deleting the chameleon bootloader <kernel>/mnt/cache/vm_images/enoch_rev2795_boot</kernel> Hope that answers your question Peter.

Thanks. Yes your and archedrafts work is excellent and i am glad i can contribute something back

Thanks Jonathanm for the feedback. No i didnt mean to leave my key in xml. i am just blurring that in the video now....Thanks yeah the vnc web client is a pain. It actually works fine if you connect from virt-managers vnc client, just not our web one. I use splashtop desktop for connection to my vms as i can use that outside aswell. Ha never heard of drunk simulator prism glasses....I wonder what it would be like wearing them when drunk...?? lol

Catalina Mojave and HighSierra - new video 28 october 2019 MOJAVE 10.14 . (Update 24 October 2018) _______________________________________________________________________________________ OLDER VIDEOS ABOUT MacOS VMs _______________________________________________________________________________________ High Sierra 10.13 (Update 24 September 2017) So now High Sierra is here its time for a new video. Things are a lot better for OSX VMs now. Since qemu 2.9 we don't need to use a patched clover. We can pass through the CPU frequency to the guest from the XML. This video although for High Sierra will also work for Sierra HOW TO INSTALL HIGH SIERRA OR SIERRA ON UNRAID CREATING INSTALL MEDIA (Update 24 September 2017) Video guide here >>>>> https://www.youtube.com/watch?v=BFPiK1Ggbks INCREASING PERFORMANCE IN AN OSX VM . (march 2018) Advanced techniques for hardware passthrough etc. (Update March 2017) This video is about advanced OSX VM techniques in KVM with unRAID. You will see how to pass through physical devices to a KVM osx sierra VM such as GPU, sound and USB. You will learn to what hardware that you can use and how to edit the XML template to add the hardware. You will learn how to use a Nvidia GPU in Sierra by installing the Nvidia web drivers. You will see how to get your HDMI sound working using an HDMI audio kext. I will quickly run through using virt-manager to avoid manual edits then discuss common osx VM problems. Note if using High Sierra as of 24 September 2017 Nvidia have not released web drivers for this OS so we cant have them working in 10.13 (but soon we should) Advanced OSX VM techniques in unRAID. Get everything working!

Hi. App is looking nice. I just notice shares with longer names dont display their name properly. Also would be great if when sleep plugin is installed we could put server to sleep from app

+1 I'm also seeing the same docker bug as CHBMB.

128mb sdram pc133

nice work its really shaping up nicely. docker and vm page is great

If you UPGRADE your current VM to '10, that will work fine -- and you'll then be able to do a clean install using the same VM if you want to. But if you try a clean install on another VM with the same key it MAY not work -- not sure how the activation servers check; but I do know they'll refuse to allow a key that's already in use ... so it may not activate correctly with that key. One approach you could take is to make a complete backup copy of your VM; then do an update of the VM to '10. You'd then have both '7 and '10 installed in the same VM => indeed if you renamed one of them they'd probably both work just fine at the same time (but that would be a licensing violation). yes it should work. But you couldnt have the vms at the same time renamed. This is because the activation seems to be tied to the uuid. You cant have 2 vms in unraid with the same uuid number. What you would have to do is to disconnect the windows 10 vm image as primary disk, and swap it to the win7 image as primary disk, this way the uuid wouldnt change. It is always a good practice to make note of your uuid for each vm that needs any activation basically all Micro$oft os

It could be that your usb controller isnt being released after the vm issue. So when you restart the vm it is not being passed through. Sometimes you have to fully powerdown to release hardware so this could be why it took you a few reboots to get it back . So you are stubbing the usb card and passing it through using the host dev method. You could try the qemu:arg method instead if you continue to have problems. That is the method i have used in the past with great sucess. But i am by no means an expert. Just one thing not related to usb problem. when i was looking at your xml. I see you are isolating cpus in your syslinux append pci-stub.ids=2109:0812 isolcpus=11,12,13,14,15,27,28,29,30,31 initrd=/bzroot but your xml you pin <cputune> <vcpupin vcpu='0' cpuset='15'/> <vcpupin vcpu='1' cpuset='16'/> <vcpupin vcpu='2' cpuset='17'/> <vcpupin vcpu='3' cpuset='18'/> <vcpupin vcpu='4' cpuset='19'/> <vcpupin vcpu='5' cpuset='27'/> <vcpupin vcpu='6' cpuset='28'/> <vcpupin vcpu='7' cpuset='29'/> <vcpupin vcpu='8' cpuset='30'/> <vcpupin vcpu='9' cpuset='31'/> <emulatorpin cpuset='0,16'/> </cputune> should be <cputune> <vcpupin vcpu='0' cpuset='11'/> <vcpupin vcpu='1' cpuset='12'/> <vcpupin vcpu='2' cpuset='13'/> <vcpupin vcpu='3' cpuset='14'/> <vcpupin vcpu='4' cpuset='15'/> <vcpupin vcpu='5' cpuset='27'/> <vcpupin vcpu='6' cpuset='28'/> <vcpupin vcpu='7' cpuset='29'/> <vcpupin vcpu='8' cpuset='30'/> <vcpupin vcpu='9' cpuset='31'/> <emulatorpin cpuset='0,16'/> </cputune> to match the isocpu you defined in the syslinux

lionelhutz has tried changing machine type and reports activation is okay when changed

Yes changing the number of cores should make no difference for activation as you are not changing the cpu type/physical cpu Many motherboards allow you to enable and disable cores aswell. Changing the cpu from host passthrough to emulated qemu64 however would be more likely to trigger activation. One tip on upgrading from 7 to 10. During the upgrade process put the core count down to 1. Sometimes on a windows 7 vm upgrade can fail when using multi cores for some reason. After upgrade you can put the core count back up

Yes it is the uuid, im sure. But the same uuid cant already be defined on unraid. So you could not just copy the vdisk and use if for a few vms on the same server. Would work though if you had one on one server and another on another. But that would be naughty and microsoft would hate you lol !!!

I think if you copy the vdisk it will work ....... but you must use the same uuid in the xml, as was in the xml with the machine when it was activated example <domain type='kvm'> <name>Windows 10 </name> <uuid>92229e2f-9329-5786-c022-3c70e2788b32</uuid> <metadata> So when you make a new vm the uuid would be different so you would have to put back in the 92229e2f-9329-5786-c022-3c70e2788b32 from above aswell as replace the vdisk So just keep a note of the uuid and should be fine

There is a button called "Power On" if your server is off. Try that one That's right ! LOL got to be on of my most stupid posts!! Who would believe power on button would turn on server! Maybe i could try the on switch on my brain and see if a wol ping could wake me!!! But in my defensive i had not looked at the app when server has been off

What would be useful on the app would be to be able to send wake on lan ping to wake server if sleeping.

Sniper, just a word of advise, a bit offtopic from helping you with your setup. But I would delete this image file above Here is my Deluge inside settings https://drive.google.com/file/d/0BwVQ1QEehpJJbTdkRVV3bFdhTjg/view?usp=drivesdk If you value privacy then it may not be a good idea to show a screen shot of the things you are downloading with deluge.

What I'm referring to is that passing through my integrated Audio controller causes issues, but when I pass through my USB Audio device I have no problems. It looks like you are not using hyper-thread pairs <cputune> <vcpupin vcpu='0' cpuset='1'/> <vcpupin vcpu='1' cpuset='2'/> <vcpupin vcpu='2' cpuset='3'/> </cputune> Check your thread pairs in settings system devices. your pairs will be 0 ------4 physical core 1 1 ------5 physical core 2 2 ------6 physical core 3 3 ------7 physical core 4 Only use pairs, do not use one hyperthread of a core or else you will get latency. So if you use 1 also use 5. If you assign 5 to something else that physical core 2 is split between 2 different processes and will be a performance hit. It has no relationship to the load on the cores so your max load being only 50% isnt an indictation your vcpu assignments are okay. I would try to pin emulator tasks to another core not used by the vm aswell. It helps alot of people get better performance. read dlandon's excellent post http://lime-technology.com/forum/index.php?topic=49051.msg470454#msg470454

Ok yes i understand now. So assigning 8 cores but being able to scale those back on the start of another vm. Thats a very interesting idea and i didnt know that was possible. It would certainly be very useful. I had a similar idea a while back about isolcpus capability post-boot to be able to isolate and release cpus to the host without reboot. Having spoken to limetech this is something they have been investigating but it isnt in Linux's capability set and as such would be only be possible manually manipulating some things in the user space for this but it isnt something they are currently working on. So yes any info on how to implement what you are talking about here would be most welcome.