SimonF

Looks like there may be other devicea in that iommu group you are not passing to the vm. What devs are in grp 18 in system devices

I guess you dont have any devices connected. If you connect a device you will see it appear and then you can create a port or device mapping. If you have enabled hotplug you can attach devices via the VM tab also. If you toggle empty ports it will show empty ports also. Example with empty ports. Click on these icons. Left is for ports, right for Device Mapping and you will see. The settings. This is for a device. This is where you can specify VM and actions. Auto connect at plugin will attach usb device when connected. Auto connect at Start will attach at VM start. Start/Resume will start/resume a VM when as device is connect and and connect at plugin is enabled. Serial only is if you need to attach as a serial port, Things like ConbeeII needs this. Default is port 4 but if you have multiple you may need to change this. Port mapping is the same as a device but will connect any device on that port to a VM. Device will only do that device but will work across any port. Hope this helps. ps usbip extra is only required if you want to use usb devices over networks from/to other ser ers pcs etc

Release 2024.04.03

You need to install 6.12.10 for the fix.

Not currently an option with rc script. I can look to add options. Have created code, but not merged and pacakged as yet. two options one for port and one for device serial_id https://github.com/SimonFair/USB_Manager/commit/552567430e0d67d6f3bb08920578ca99ad4e00f6

Yes looks correct to me.

Rust binary in /usr/bin Bash in /usr/libexec Attached php file in /usr/local/emhttp/plugins/dynamix.vm.manager/scripts chmod +x for bash and php. path in XML should be /usr/libexec/virtiofsd Should work on 6.12 but I have not tested on that release. virtiofsd.php

rename /usr/local/sbin/qemu and the VM will not be able to start. You will be able to remove from the gui. But remember to rename back.

I am adding support for the next release. my pre-process is here https://github.com/unraid/webgui/pull/1650/files#diff-4c8d5cce21f2c78e645df54b1784ac87676c0af710e0cb0c941e277f8cc823d9 This bash script needs to be added to /usr/libexec/ and made executable, The preprocessor above also has to be executable. root@computenode:/mnt/user/appdata# cat /usr/libexec/virtiofsd #!/bin/bash eval exec /usr/bin/virtiofsd $(/usr/local/emhttp/plugins/dynamix.vm.manager/scripts/virtiofsd.php "$@") root@computenode:/mnt/user/appdata# it takes options sent thru to the command from the VM start. root 1143 1 0 09:10 ? 00:00:00 /usr/bin/virtiofsd --syslog --inode-file-handles=mandatory --announce-submounts --fd=46 --shared-dir=/mnt/cache --cache=never --sandbox=chroot --xattr root 1147 1 0 09:10 ? 00:00:00 /usr/bin/virtiofsd --syslog --inode-file-handles=mandatory --announce-submounts --fd=46 --shared-dir=/mnt/user/isos --cache=never --sandbox=chroot --xattr root 21072 1 0 09:13 ? 00:00:00 /usr/bin/virtiofsd --syslog --inode-file-handles=mandatory --announce-submounts --fd=42 --shared-dir=/mnt/user/domains3 --cache=never --sandbox=chroot --xattr Rust binary need to be in /usr/bin/virtiofsd cache is set to never, other values are taken from the -o options. The following additional switches are added. --syslog' --inode-file-handles=mandatory --announce-submounts but you can override these by creating /etc/libvirt/virtiofsd.opt and each option needs to be on a new line.

Can you post the XML for the VM. Have you passed the NVME as a PCI device?

Are you runming the srvio plugim nas that will create multiple gpus.

If you have intel_gpu_top installed you can run the list, this will show vendor associated. root@computenode:~# intel_gpu_top -L card1 Intel Alderlake_s (Gen12) pci:vendor=8086,device=4680,card=0 └─renderD129 card0 10de:1fb2 pci:vendor=10DE,device=1FB2,card=0 └─renderD128 card2 1002:743f pci:vendor=1002,device=743F,card=0 └─renderD130 root@computenode:~#

The autostart process is controlled by libvirtd starting and not Unraid scripts. But will look to see if there is a way to do.

You need to install the virtioscsi driver in the guest before you change from sata to scsi

if you add a dGPU you will have to change the bios to enable the iGPU as normally it likely disable igpu if dgpu is seen

Are you still have issues, You logs are full of these errors. Tower winbindd[6662]: set_global_winbindd_state_offline: Offline requested Feb 22 09:34:10 Tower winbindd[6662]: [2024/02/22 09:34:10.370664, 0] ../../source3/winbindd/winbindd_cache.c:3465(set_global_winbindd_state_offline) Can you post current diagnostics if having the issue still.

Please add to the thread option your would like to see in VM manager. Cloning and snapshots are already in the pipeline. Possible features for review(From existing feature requests). XE Driver support when kernel supports Native SR-IOV graphics support Nvidia vGPU - Licensed support. VirGL support. Better XML persistance Allow changes of running VM to take effect at next boot.

they are normally update in major version Currently is 8.1.x for qemu and 10.x for libvirt

sorry needs to be ttyS0 not lower case s

should be ttys0 to 4 for physical s0 is com1 s3 is com4

ls /dev/tty* https://www.cyberciti.biz/faq/unix-linux-apple-osx-bsd-screen-set-baud-rate/ screen can be download from nerdtools plugin I think.

if detected would show as ttysx device where x is a number Do you have enabled in bios?

License is tied to the usb flash boot drive. So you can clear config and move to a different machine.

Have you hidden the panel. Click on padlock to unlock spanner should appear on the system panel. Click and check panels selected.

Can you elaborate on the above statement? more details on which type of vgpu support, i.e. GVT-g, sr-vio or something else?