# \[KVM\] VM live migration with attached NIC issue

**URL:** <https://forum.opennebula.io/t/kvm-vm-live-migration-with-attached-nic-issue/7049>\
**Category:** Installation & Configuration\
**Created:** [March 20, 2019, 8:56am UTC](https://forum.opennebula.io/t/kvm-vm-live-migration-with-attached-nic-issue/7049 "2019-03-20T08:56:45Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![gray380](https://avatars.discourse-cdn.com/v4/letter/g/7993a0/32.png) [@gray380](https://forum.opennebula.io/u/gray380)\
**Post date:** [March 20, 2019, 8:56am UTC](https://forum.opennebula.io/t/kvm-vm-live-migration-with-attached-nic-issue/7049/1 "2019-03-20T08:56:45Z")

</div>

Hi there,

Server and nodes are hosted at the CentOS Linux release 7.6.1810 (Core).

> $ uname -r  
> 3.10.0-957.10.1.el7.x86\_64

When I start VM live migration the bridge interface has entered into the disabled state.  
Please find some console outputs/logs.  
Before migration:

> [sadm@one-kvm-node-02 ~]$ brctl show  
> bridge name bridge id STP enabled interfaces  
> br0 8000.0025902dd6fc no eno1  
> br1 8000.0025902dd6fd no eno2  
> one-20-0

After starting the migration:

> [sadm@one-kvm-node-02 ~]$ brctl show  
> bridge name bridge id STP enabled interfaces  
> br0 8000.0025902dd6fc no eno1

journalctl shows the following messages:

> Mar 20 10:25:40 one-kvm-node-02.tsu kernel: device eno2 left promiscuous mode  
> Mar 20 10:25:40 one-kvm-node-02.tsu kernel: br1: port 1(eno2) entered disabled state

VM log:

> Wed Mar 20 10:25:34 2019 [Z0][VM][I]: New LCM state is MIGRATE  
> Wed Mar 20 10:25:34 2019 [Z0][VMM][I]: Successfully execute transfer manager driver operation: tm\_premigrate.  
> Wed Mar 20 10:25:35 2019 [Z0][VMM][I]: ExitCode: 0  
> Wed Mar 20 10:25:35 2019 [Z0][VMM][I]: Successfully execute network driver operation: pre.  
> Wed Mar 20 10:25:40 2019 [Z0][VMM][I]: ExitCode: 0  
> Wed Mar 20 10:25:40 2019 [Z0][VMM][I]: Successfully execute virtualization driver operation: migrate.

VM has stucked in the MIGRATE status.  
To solve this issue I’ve restarted network service:

> [sadm@one-kvm-node-02 ~]$ sudo systemctl restart network

And migration process has successfully finished:  
VM log:

> Wed Mar 20 10:33:10 2019 [Z0][VMM][I]: clean: Executed “sudo ip link delete br1”.  
> Wed Mar 20 10:33:10 2019 [Z0][VMM][I]: ExitCode: 0  
> Wed Mar 20 10:33:10 2019 [Z0][VMM][I]: Successfully execute network driver operation: clean.  
> Wed Mar 20 10:33:10 2019 [Z0][VMM][I]: ExitCode: 0  
> Wed Mar 20 10:33:10 2019 [Z0][VMM][I]: Successfully execute network driver operation: post.  
> Wed Mar 20 10:33:10 2019 [Z0][VMM][I]: Successfully execute transfer manager driver operation: tm\_postmigrate.  
> Wed Mar 20 10:33:10 2019 [Z0][VM][I]: New LCM state is RUNNING

Node’s outputs:

> [sadm@one-kvm-node-02 ~]$ brctl show  
> bridge name bridge id STP enabled interfaces  
> br0 8000.0025902dd6fc no eno1  
> br1 8000.0025902dd6fd no eno2
> 
> [sadm@one-kvm-node-01 ~]$ brctl show  
> bridge name bridge id STP enabled interfaces  
> br0 8000.ac162db72720 no eno1  
> br1 8000.ac162db72721 no eno2  
> one-20-0

Could you assist is there a VM migration method without restarting the network service?

brg,  
Serhiy.

---

<div class="post-metadata">

**Author:** ![gray380](https://avatars.discourse-cdn.com/v4/letter/g/7993a0/32.png) [@gray380](https://forum.opennebula.io/u/gray380)\
**Post date:** [March 20, 2019, 1:08pm UTC](https://forum.opennebula.io/t/kvm-vm-live-migration-with-attached-nic-issue/7049/2 "2019-03-20T13:08:10Z")

</div>

Solution or workaround:  
Edit /var/lib/one/remotes/etc/vnm/OpenNebulaNetwork.conf to

> ```
> # Don't delete bridge with no virtual interfaces left
> #:keep_empty_bridge: false
> :keep_empty_bridge: true
> 
> ```