<html><head></head><body><div>I may have run into this issue before too, but in my case the VMs ran fine with only 1 vcpu so I booted them that way and left them like that. It was just a few small legacy VMs and I didn't spend any more time on it.<br></div><div><br></div><div>Did you try booting with only 1 vcpu?</div><div><br></div><div>Chris<br></div><br>--<br>
Chris Wik<br>
Anu Internet Services<br>
<a href="http://www.anu.net">www.anu.net</a> | <a href="http://www.cwik.ch">www.cwik.ch</a><br><br><br><div><strong>
From:
</strong>
isdtor <isdtor@gmail.com>
<br>
<strong>
To:
</strong>
<centos-virt@centos.org>
<br>
<strong>
Sent:
</strong>
24/04/2020 11:53 AM
<br>
<strong>
Subject:
</strong>
[CentOS-virt] VM migration problems
<br><br><blockquote class="mori" style="margin:0 0 0 .8ex;border-left:1px solid #CCC;padding-left:1ex;">I have migrated KVM VMs from a CentOS 6 to a CentOS 7 host. All work fine post-migration, CentOS 3 (don't ask ...), CentOS 6, CentOS 7, Windows. But the CentOS 5 VMs failed. At some point during the boot process, they became unpingable and also inaccessible.
<br>
<br>I have correlated this to the start of the irqbalance service and was wondering if it is generally considered best practice to turn it off. The VMs in question all have at least 2 vcpus, and the C5 VMs are the odd ones out.
<br></blockquote></div></body></html>