ilarion
Members
-
Joined
-
Last visited
Everything posted by ilarion
-
So what is your card model ? Like all the letters and numbers....
-
It is exactly say in the VM log what device it is not present and you are trying to attach to VFIO. As for automatic part. It is pci address that you attach that mean if you swap a card it will be attached automatically. For unraid this is a slot that you redirect what is on it is for vm os concern. And if it is nothing for most case is, mess because when you remove device and it is alone in the ioumu group this group number is bean taken with the next inline. So all VM has to be reedited.
-
53 is the port for dns... if it is shared between unraid and adguard one of them will not have dns i suppose that is why it has to put it on external network bridge not on internal docker one.So that you can assign it ip of the unraid network not of the docker one.
-
Btw did you install qemu guest tools ? This is part of it.
-
#Attach graphics card without turn off?? - Where do you reed this ? Upload xml and diagnostics and tell my the exact model of the card. Bios looks legeit when open it with hex editor maybe is only legacy or uefi. Maybe that is why it doesn`t start if it is only legacy. Btw from what i remember with one gpu that has only legacy bios it is not showing windows boot process (tinycore)(because OVMF is uefi emulation) but when windows boot it was ok and showing desktop (just not the part before desktop), but if you don`t have nvidia drivers in windows installed it wont show when booted and only black screen. Did you try with vnc virtual card and 970 as a second card ? PS: Are you seeing tinycore logo on UBUNTU VM when booting or first thing after start is desktop of ubuntu ?
-
So you are trying to install windows or is it already installed one ? If you trying to install it, best practice is to install it with vnc and after installation enable RDP (remote desktop, or team viewer or etc) , after that attach the card to already installed system. PS: the vbios that you are using is with different names(for ubuntu and windows from the screenshots)? Is it the same ? If it is bios from techpowerup and not one that you dump it from your card it wont work without editing. Sometimes bios that doesn't work with your card just being ignored and works but....
-
Put audio part of the gpu too in the VM template, don't pin 0/8 treads to VM leav them for unraid use other couple.
-
hit ESC on boot.
-
post your xml
-
Why are some of the menus in the "Nvidia control panel" missing? For those of you that do gpu passthrough with NVIDIA cards and pull your hair why half of the menus in the "Nvidia control panel" are missing. - problem that google didn't help at all. I have the luck to have nvme passtrought and could boot baremetal so i could see that baremetal menus are visible so for them to be visible in vm put : <kvm> <hidden state='on'/> </kvm> in the "<features>" section of your xml.
-
-
I think it is, it is just the diference between allocated space and block space. In main page in used space shows block space and in file manager shows block space. In smb share from windows explorer shows block space too. And on top of that there is the difference between GiB and GB that confuse me more at the start of the investigation because a forgot about that thing. But in the end i don`t know if this space is usable. So what i have to believe "du" or "ls" ?
-
Only opening the page for pinning brings up the banner "CPU Isolation: A reboot is required to apply changes" without any changes make. This banner stay till restart. I tried different browser after restart to exclude some cashing thing - same problem. PS: i get too an error of ntbd something about frequency :ntpd[466]: frequency error -512 PPM exceeds tolerance 500 PPM. I have 4 working ntp entered and checked that they are up and pingble. PS2: if "My Servers" plugin is used and there is no internet when the server is booting nginx demon starting takes 3 minutes. For what i think it is that cerbot try to get replay from lets encrypt or something to test the certificate for ssl. But this is too long if you testing something and have too reboot continuously. PS3: Don`t try to use the new amd_pstate driver it is unusable. Yes it is get core to 500Mhz min but there is no difference on power usage from my smart plug. And on top of that VMs are sluggish and unresponsive. PS4: I am eagerly waiting for nerd pack to be usable again so that i can start using sensors and dynamics temp plugin with new supported asus wmi.I try it on different distro with 5.17 and is working. So after so much time asus x570 boards are having temp control and fan monitoring.
-
Sorry, i use XFS. So what is actually has to happen? Doesn't it have to make allocation size same with actual(logical) so that i can use this space for other files?
-
It is all working for me with the exception of that the host dosen`t reclaim actual space. When i use "ls" the qcaw2 file is still the size before the optimization in the guest and trim on the host. Only "du" shows the trimmed size, but that is not usable space on host. So how to reclaim any tangible space on the host? Do i have to use LVM or some kind of esoteric fs as btrfs or zfs so that to work ?
