[ovs-discuss] n00b Requesting Help - I'm Missing Something Basic

duluxoz duluxoz at gmail.com
Mon Mar 29 07:00:43 UTC 2021


Hi All,

Sorry if this is the wrong place to post this and/or I missed the answer 
in the doco.

I'm new to OVS and have been struggling for days with this issue - I 
hope someone can help.

I have a Centos 8 box loaded with kvm/qemu/libvert, NetworkManager-ovs, 
and ovs (v12.2). The box has three nics in it (en01, en02, en03). I am 
using eno3 as a management-nic (with an ip address of 192.168.0.16) for 
the box itself, and the other two nics set up as a lacp team (team1, no 
ip address, with team1-slave1 (eno1) and team1-slave2 (eno2)).

Everything is working correctly, and my physical switch has the two team 
nics both active and load-balanced.

I created an ovs (ovs1) and set it up with a static ipv4 address 
(192.168.0.17). I also modified the ifcfg-team1 file to include the 
references to ovs1.

But I can't ping ovs1 and nmcli says that ovs1 is down - and yes, I have 
tried bringing it up (nmcli c up ovs1) - which gives an error.

I have include the relevant ifcfg-* files below.

Can someone please tell/show me what I'm doing wrong - thanks.

ifcfg-ovs1:

~~~
DEVICE=ovs1
ONBOOT=yes
DEVICETYPE=ovs
TYPE=OVSBridge
BOOTPROTO=none
IPADDR=192.168.0.17
PREFIX=24
DEFROUTE=yes
GATEWAY=192.168.0.1
DNS1=192.168.0.2
DNS2=192.168.0.3
IPV6INIT=no
HOTPLUG=no
~~~

ifcfg-team1:

~~~
TEAM_CONFIG=$'{\"runner\": {\"name\": \"lacp\", \"active\": true, 
\"fast_rate\": false, \"agg_select_policy\": \"lacp_prio\"}, 
\"link_watch\": {\"name\": \"ethtool\"}}'
PROXY_METHOD=none
BROWSER_ONLY=no
BOOTPROTO=none
DEFROUTE=yes
IPV4_FAILURE_FATAL=no
IPV6INIT=no
NAME=team1
DEVICE=team1
ONBOOT=yes
DEVICETYPE=ovs
TYPE=OVSPort
OVS_BRIDGE=ovs1
~~~

ifcfg-team1-slave1:

~~~
ETHTOOL_OPTS="autoneg off speed 1000 duplex full"
NAME=team1-slave1
DEVICE=eno1
ONBOOT=yes
TEAM_MASTER=team1
DEVICETYPE=TeamPort
~~~

ifcfg-team1-slave2:

~~~
ETHTOOL_OPTS="autoneg off speed 1000 duplex full"
NAME=team1-slave2
DEVICE=eno2
ONBOOT=yes
TEAM_MASTER=team1
DEVICETYPE=TeamPort
~~~

Finally, nmcli d status give:

~~~

DEVICE TYPE           STATE         CONNECTION
eno1         ethernet          connected       team1-slave1
eno2         ethernet          connected       team1-slave2
eno3 ethernet          connected       eno3
team1 team               connected       team1
ovs1         ovs-interface  disconnected  --
lo             loopback         unmanaged    --
ovs1        ovs-bridge       unmanaged    --
team1      ovs-port           unmanaged    --
ovs1        ovs-port           unmanaged    --
~~~

Cheers

Dulux-Oz

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openvswitch.org/pipermail/ovs-discuss/attachments/20210329/e402ff8b/attachment-0001.html>


More information about the discuss mailing list