Affinity rules are able to block maintenance reboots

E

edmeades

New Member
Proxmox Subscriber
Jul 8, 2026
3
0
1
I configured 3 VMs in a negative affinity group on my 3-node cluster. When I rebooted a node today for kernel updates, the migration process stalled silently because the rule took precedent over maintenance. It would be helpful to add an option that would allow PVE to bend the rules for maintenance and/or allow manual migration contrary to the ruleset with a warning. Further, the user should be notified if a rule is interfering with migration/reboot of a node.
 
It seems that this may have put my cluster in a weird state. All the VMs are running, but trying to manually migrate anything says "Task OK", but as far as I can see nothing actually happens.
 
dmesg output from node0:


Code: 
[3039114.139059] kvm_pr_unimpl_wrmsr: 278 callbacks suppressed
[3039114.139065] kvm: kvm [71900]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039114.167096] kvm: kvm [42492]: vcpu1, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039114.191054] kvm: kvm [71900]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039114.207093] kvm: kvm [42492]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039114.219100] kvm: kvm [42492]: vcpu1, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039114.243061] kvm: kvm [71900]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039114.271095] kvm: kvm [42492]: vcpu1, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039114.295054] kvm: kvm [71900]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039114.299039] kvm: kvm [71900]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039114.323097] kvm: kvm [42492]: vcpu1, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039119.159022] kvm_pr_unimpl_wrmsr: 287 callbacks suppressed
[3039119.159027] kvm: kvm [42492]: vcpu1, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039119.182976] kvm: kvm [71900]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039119.211018] kvm: kvm [42492]: vcpu1, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039119.234974] kvm: kvm [71900]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039119.258964] kvm: kvm [71900]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039119.263018] kvm: kvm [42492]: vcpu1, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039119.279017] kvm: kvm [42492]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039119.286986] kvm: kvm [71900]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039119.315016] kvm: kvm [42492]: vcpu1, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076
[3039119.319012] kvm: kvm [42492]: vcpu0, guest rIP: 0xffffffff8102b3e3 Unhandled WRMSR(0xc0010000) = 0x530076

Nodes 1 and 2 do not have these errors
 
You must log in or register to reply here.
Top Bottom
Back