[SOLVED] New Proxmox install- 1st VM wont start- Error- start failed:QEMU exited with code 1

[WallyV]

Installed Proxmox last night, just did an apt update & upgrade then I created a Debian12 VM- 32GB RAM - Processors 20( 2 sockets,10 cores) Haswell- 100GB virtual disk from a 500GB disk. The HPE DL380 Gen9 server has 2-Intel Xeon E5-2687W V3-Total Cores 10 Total Threads 20x2 =20 Cores 40 Threads with 448 GB RAM so I didn't think 32 GB ram Processors 20 would cause any problems.
It shows the VM was created OK but every time I start the VM, it exits with Error:start failed: QEMU exited with code 1
This is meant to just be a temporary VM until I am able to add additional hard drives hence the 100GB virtual drive.
I am not sure how to view logs or errors on Proxmox.
Any other information you need?
Any idea why the VM won't start?
Thanks

 

[Pifouney]

Hi,

Check with only 1 socket activated. I think you need to enable numa with this kind of VM with more than 1 socket processor

 

[WallyV]

Hi,

Check with only 1 socket activated. I think you need to enable numa with this kind of VM with more than 1 socket processor

Thanks,
I edited the VM in the GUI, changing the Processors from 2 sockets- 20 cores , to 1 socket-10 cores, then restarted and got the same error message.
Any other ideas?
Thanks

 

[WallyV]

Hi,

Check with only 1 socket activated. I think you need to enable numa with this kind of VM with more than 1 socket processor

I experimented with various options in the Processors section. The only thing that has worked so far is- 1 socket- 10 cores, Haswell-noTSX ( Haswell & Haswell-IBRS didn't work)
In advanced-
VCPUs 10 CPU units 20 & Enable NUMA checked.
I haven't checked console yet.
Memory says 32.00 Mib . I will shut it down and try 32 GB

 

[Pifouney]

Not experienced yet, but i know that your problem is "because" your ve multi sockets hardware

 

[Pifouney]

Not experienced yet, but i know that your problem is "because" your ve multi sockets hardware

Cannot be more precise sry

 

[bbgeek17]

Did you try to double click on the failed task to see if there is more information in details?
You can also run "journalctl -n 500" immediately after error to see if system log provides more information.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[WallyV]

Thanks. I did get the VM working. I was able to get the VM started, installed Debian12, and now I am installing the updates on the VM.
Memory- 32GB
Processors- 10 ( 1 socket, 10 cores) Haswell-noTSX
In Advanced-
Numa=1 cpuunits=20

I also got another VM working with nearly the same options-
Memory- 32GB
Processors- 10 ( 1 socket, 10 cores) Haswell-noTSX-IBRS
In Advanced-
this area is grayed out so I don't actually know if these options are active
VCPUs:10 CPU units: 100 Enable NUMA: Checked(YES)


I will experiment with more VMs and various Types and options to find out other variations that work and others that give me the old errors again.
Thanks everyone.

 

[Ela@Blach]

Generally when I have issues with a VM starting, I will revert to some basic and well known settings... CPU type KVM, Ballooning Off, Single socket, cpu flags in default positions, iscsi controller: LSI 53C...., Disk mounted as IDE*, NIC: Intel E1000, Passthrough devices disabled.

Most of the time... virtual machine will boot with these settings and then you can start introducing things back in and find the issue. If that does not work, I usually start playing with CPU flags. (Oh and double click the tasks where it fails to start so you can see more details)

 

[WallyV]

Did you try to double click on the failed task to see if there is more information in details?
You can also run "journalctl -n 500" immediately after error to see if system log provides more information.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

Thanks. I couldn't figure out how to find a terminal for Command line in the Proxmox Dashboard, so I opened a terminal in my Debian12 laptop and did ssh root@192.168.0.169 ( the local IP of Proxmox VE) and ran the command you listed. That allowed me to see the last 500 lines of messages. I also just doubled clicked on "Error: start failed: QEMU exited with code 1" in the Status column on the bottom of the Dashboard page and saw this-
kvm: warning: host doesn't support requested feature: CPUID.07H:EBX.hle [bit 4]
kvm: warning: host doesn't support requested feature: CPUID.07H:EBX.rtm [bit 11]
kvm: Host doesn't support requested features
TASK ERROR: start failed: QEMU exited with code
I have since gotten that VM to run but at least now I have 2 ways to get information on error messages.
Thanks

 

[crimildom]

"kvm: -drive file=/dev/pve/vm-100-disk-0,if=none,id=drive-virtio0,format=raw,cache=none,aio=io_uring,detect-zeroes=on: Could not open '/dev/pve/vm-100-disk-0': No such file or directory
TASK ERROR: start failed: QEMU exited with code 1. "
when a try to start a vm a get the error message above, then when i go to "/dev/pve" are only displeyed "root and swap", i cannot see the virtual disks.
please, how can i solve this problem?

 

[LnxBil]

how can i solve this problem?

Look at the output of dmesg if there are errors and provide the ouput of pvs and lvs (all in CODE tags)

 

[crimildom]

Look at the output of dmesg if there are errors and provide the ouput of pvs and lvs (all in CODE tags)

thank you, but how can i get the dmesg and pvs/lvs?

 

[bbgeek17]

thank you, but how can i get the dmesg and pvs/lvs?

One way is to SSH into the host and execute those commands.
Another way via GUI > Datacenter>Nodes>NODE>shell (top right corner). However, its less practical as there is limited scroll back and no copy/paste.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox