[rdo] [packstack] failing to setup Stein with Openvswitch/VXLAN networking

Bernd Bausch berndbausch at gmail.com
Thu Aug 1 02:39:05 UTC 2019


Hmmmmm. I just succeeded in setting up my Packstack after removing IPv6 
from the kernel and cleaning out ifgcfg-eth0. Since I don't need IPv6, I 
am good. My config:

$ *cat /etc/sysconfig/network-scripts/ifcfg-eth0*
TYPE=Ethernet
BOOTPROTO=none
DEFROUTE=yes
NAME=eth0
DEVICE=eth0
ONBOOT=yes
IPADDR=192.168.1.202
PREFIX=24
GATEWAY=192.168.1.1
DNS1=192.168.1.16
DNS2=1.1.1.1
DOMAIN=home
IPV6INIT=no
$ *cat /etc/sysctl.conf*
net.ipv6.conf.eth0.disable_ipv6=1

Thanks again, Alfredo.

On 8/1/2019 10:23 AM, Bernd Bausch wrote:
>
> Thanks Alfredo. Yes, a single eth0:
>
> $ ip a
> 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 
> group default qlen 1000
> ...
> 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast 
> state UP group default qlen 1000
>     link/ether 52:54:00:dd:2c:f3 brd ff:ff:ff:ff:ff:ff
>     inet 192.168.1.202/24 brd 192.168.1.255 scope global eth0
>        valid_lft forever preferred_lft forever
>     inet6 240d:1a:3c5:3d00:5054:ff:fedd:2cf3/64 scope global 
> mngtmpaddr dynamic
>        valid_lft 10790sec preferred_lft 10790sec
>     inet6 fe80::5054:ff:fedd:2cf3/64 scope link
>        valid_lft forever preferred_lft forever
>
> and ifcfg-eth0 is unchanged since installation (perhaps I should clean 
> it up a little; could IPv6 be causing problems?):
>
> $ cat /etc/sysconfig/network-scripts/ifcfg-eth0
> TYPE=Ethernet
> PROXY_METHOD=none
> BROWSER_ONLY=no
> BOOTPROTO=none
> DEFROUTE=yes
> IPV4_FAILURE_FATAL=no
> IPV6INIT=yes
> IPV6_AUTOCONF=yes
> IPV6_DEFROUTE=yes
> IPV6_FAILURE_FATAL=no
> IPV6_ADDR_GEN_MODE=stable-privacy
> NAME=eth0
> UUID=75c26b23-f3fb-4d3f-a473-5354075d5a25
> DEVICE=eth0
> ONBOOT=yes
> IPADDR=192.168.1.202
> PREFIX=24
> GATEWAY=192.168.1.1
> DNS1=192.168.1.16
> DNS2=1.1.1.1
> DOMAIN=home
> IPV6_PRIVACY=no
>
> More info: This is a VM running /CentOS 7.6.1810 /on a Fedora KVM 
> host. eth0 is bridged. I did disable and stop network manager and 
> firewalld on the guest.
>
> Bernd.
>
> On 7/31/2019 10:13 PM, Alfredo Moralejo Alonso wrote:
>> Hi,
>>
>> So, IIUC, your server has a single NIC, eth0, right?
>>
>> Could you provide the configuration file for eth0 before running 
>> packstack?, i guess you are using ifcfg files?
>>
>> Best regards,
>>
>> Alfredo
>>
>>
>> On Wed, Jul 31, 2019 at 2:20 AM Bernd Bausch <berndbausch at gmail.com 
>> <mailto:berndbausch at gmail.com>> wrote:
>>
>>     Trying to set up a Stein cloud with Packstack. I want the
>>     Openvswitch mech driver and VXLAN type driver. A few weeks ago,
>>     the following invocation was successful:
>>
>>     |sudo packstack --debug --allinone --default-password pw \
>>     --os-neutron-ovs-bridge-interfaces=br-ex:eth0 \
>>     --os-neutron-ml2-tenant-network-types=vxlan \
>>     --os-neutron-ml2-mechanism-drivers=openvswitch \
>>     --os-neutron-ml2-type-drivers=vxlan,flat \
>>     --os-neutron-l2-agent=openvswitch \
>>     --provision-demo-floatrange=10.1.1.0/24\
>>     --provision-demo-allocation-pools
>>     '["start=10.1.1.10,end=10.1.1.50"]'\ --os-heat-install=y
>>     --os-heat-cfn-install=y|||
>>
>>     Now, it fails during network setup. My network connection to the
>>     Packstack server is severed, and it turns out that its only
>>     network interface /eth0 /has no IP address and is down. No bridge
>>     exists.
>>
>>     In the /network.pp.finished /file, I find various /ovs-vsctl
>>     /commands including /add-br/, and a command /ifdown eth0 /which
>>     fails with exit code 1 (no error message from the /ifdown
>>     /command is logged).
>>
>>     *Can somebody recommend the options required to successfully
>>     deploy a Stein cloud based on the Openvswitch and VXLAN drivers?*
>>
>>     Thanks much,
>>
>>     Bernd
>>
>>
>>
>>     ||
>>
>>     ||
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-discuss/attachments/20190801/75a2dc7b/attachment-0001.html>


More information about the openstack-discuss mailing list