Search results

Same problem. Any workaround? I tried changing hv_* to several combinations but it didn't work.

I have already read and followed that article. Both VMs have the same Windows Revision and same virtio drivers installed. As you can see in the configuration file, virtio scsi single is used.

Windows 10 VM is too slow on Proxmox. Even dragging the window is too slow. I have two VMs with similar settings. But the second VM is not slow. The difference is, slow VM's cpu type is host , while fast VM's cpu type is qemu64 . VM1: agent: 1 balloon: 0 bios: ovmf boot...

Thank you! Using virtio-blk has improved performance! config: ``` virtio1: datastore:vm-12205-disk-0,backup=0,cache=writeback,discard=on,iothread=1,size=120G ``` ``` # dd if=/dev/vda of=/dev/null bs=1M skip=2000 count=6000 iflag=direct,sync status=progress 6090129408 bytes (6.1 GB, 5.7 GiB)...

I know dd is not a "good benchmark", but I want to make sure that sequence performance is good. ``` agent: 1 balloon: 40960 bios: ovmf boot: order=scsi0;ide2;net0 cores: 16 cpu: host,flags=+aes efidisk0: local-lvm,efitype=4m,pre-enrolled-keys=1,size=4M ide2...

The physical disks are NVMe and provisioned with lvmthin. Very fast when benchmarking a thin-volume assigned to a vm on host . But in vm it is quite slow. Please advise for DISK IO performance. VM Options: ``` -drive...

Adding xsave no longer panics! good!

I just configure CPU as added avx cpu and booted. The OS was Ubuntu 20.04/22.04. It has nothing to do with MongoDB as it causes a kernel panic during the boot process. It works when the cpu is set as the host. As leesteken said, might be able to solve it with an extra flag. I just don't know...

Theoretically, if you type exactly like that, the avx flag will be expanded on the existing kvm64. But this didn't work for me. A kernel panic occurs. https://forum.proxmox.com/threads/avx2-cause-kernel-panic.115206/

cpu-models.conf: cpu-model: avx flags +avx;+avx2 phys-bits host hidden 0 hv-vendor-id proxmox reported-model kvm64 Adding the avx2 option causes a kernel panic. Both Ubuntu 20.04 & 22.04 are the same. Is there a bug in qemu?

We can change the CPU to host to use AVX, but live migration is difficult in this case. AVX is supported by most server CPUs. Like aes, it would be nice to be able to use AVX while using kvm64. Example) AVX is required from MongoDB version 5 or later.

Are there any recent kernel or qemu modifications related to this?

Huh... I shut down pvestatd on the node that says in the error for a few hours and the error went away. However, this has happened before, and I think there will be problems again sometime. Note that all servers are synchronized with NTP, usually with an error of fewer than 0.0001 seconds.

There is only one pvestatd running on every node.

This error message keeps filling up the disk. Killing the pvestatd on node-01 stops the error, but it is not a solution. Feb 9 13:30:38 pve-node-02112 pmxcfs[1788890]: [status] notice: RRD update error /var/lib/rrdcached/db/pve2-storage/node-01/local...

First of all, if the smm=off option is added, no graphics are output, and neither ssh nor ping works. There are many important VMs on the node where the problem occurred this time, so I did not check it. I will check later. E5-2630 v3 is our latest hardware... :_( It's been working fine for...

This is before editing: /usr/bin/kvm -id 41010 -name k8s-stor-node-02063 -no-shutdown -chardev 'socket,id=qmp,path=/var/run/qemu-server/41010.qmp,server=on,wait=off' -mon 'chardev=qmp,mode=control' -chardev 'socket,id=qmp-event,path=/var/run/qmeventd.sock,reconnect=5' -mon...

Follow that comment, there is no graphic output as shown below and the VM does not turn on normally. Guest has not initialized the display (yet).

Sorry. Found the log. There is too much memory left, and there is no OOM log anywhere. # cat kern.log.1 Jan 25 12:01:56 XXX kernel: [1122677.458424] device tap41010i0 entered promiscuous mode Jan 25 12:01:56 XXX kernel: [1122677.528142] fwbr41010i0: port 1(tap41010i0) entered blocking state...

There is no sign of shutdown even inside the VM, and it was forced to stop. Both syslog/kern have no logs. Any way to find out the cause? Surely it will be again. version: pve-manager/7.1-7/df5740ad (running kernel: 5.13.19-2-pve)