Cannot check QEMU binary /usr/libexec/qemu-kvm

[boubouX]

Hello,

I’ve run into an issue while migrating a VM from a CentOS KVM cluster to a new Ubuntu KVM cluster. From what I’ve found, this seems to be a generally known behaviour, and the usual workaround involves adding a symlink to match Ubuntu’s directory layout.

2025-10-28 22:21:50,814 ERROR [o.a.c.e.o.VolumeOrchestrator] (Work-Job-Executor-29:ctx-4a00989d job-51554/job-51555 ctx-decac023) (logid:266dbad9) Failed to migrate VM [VM instance {"id":109,"instanceName":"i-16-109-VM","state":"Migrating","type":"User","uuid":"6f0b89d7-7f4c-45f6-90c5-ec8ca25f51ad"}] along with its volumes due to [com.cloud.utils.exception.CloudRuntimeException: Copy volume(s) to storage(s) [{volume: "221", from: "10", to:"42"},{volume: "142", from: "10", to:"21"}] and VM to host [{vm: "109", from: "32", to:"59"}] failed in StorageSystemDataMotionStrategy.copyAsync. Error message: [Exception during migrate: org.libvirt.LibvirtException: Cannot check QEMU binary /usr/libexec/qemu-kvm: No such file or directory].].

That said, I haven’t come across any mention of this from CloudStack users. Surely this has happened to others… but I can’t seem to find any reference to it anywhere. Before I make the simple exception fix, I just wanted to confirm whether this path is actually hardcoded or if there’s a cleaner solution I’ve missed.

Regards,

[weizhouapache]

@boubouX
I do not recommend the live migration from centos to ubuntu, or vice versa, because centos does not use vanilla qemu

stop the vm, and start it on ubuntu server, would be the better way.

[boubouX]

HI Wei,

I should have been clearer - I was migrating from a stop state.

[boubouX]

My response makes no sense - let me try to reproduce it

[boubouX]

I have recreated the issue. It seems that I was indeed trying to do a live migrate from the CentOS cluster to the Ubuntu Cluster. Sorry about the confusion.