Lily Xing's profile - activity

It is caused by the mismatch mtu settings. After I changed them to support MTU 9000, it works.

Hi Alin, Yes you are right. I tried the latest 2.4 version and it works. Thanks.

Hi, is there a way to get the old version of openvswitch for windows? I need 2.3 for kilo, but seems there is only download link for 2.4.

The root cause seems I tried using MTU 9000. When I use the default MTU=1500, it works well. But whenever I use MTU 9000, it doesn't work.

What I have updated:

dnsmasq.conf:
dhcp-option-force=26,8950

l3_agent.ini:
[DEFAULT]
network_device_mtu = 8950

ml2_conf.ini:
[ml2]
path_mtu = 8950
segment_mtu = 8950`

nova.conf:
[DEFAULT]
network_device_mtu = 8950

Hi Alessandro, I have updated compute logs. Thanks.

I setup a controller+network node on Oracle Linux(Kilo), and install a hyper-v compute node following cloudbase doc. But after installation, I can find neutron hyper-v agent, but no related compute services. And I checked that table compute_nodes in nova database is empty. Any idea?

compute service is running correctly on windows 2012 R2. No error message reported on either compute node or controller node.

compute-node.log as below:

2015-12-30 09:46:21.717 2624 INFO oslo_messaging._drivers.impl_rabbit [req-e3650168-e7d4-4d2b-8a44-7c9f504f3248 - - - - -] Connecting to AMQP server on 192.168.12.20:5672
2015-12-30 09:46:21.733 2624 INFO oslo_messaging._drivers.impl_rabbit [req-e3650168-e7d4-4d2b-8a44-7c9f504f3248 - - - - -] Connected to AMQP server on 192.168.12.20:5672
2015-12-30 09:46:21.749 2624 INFO oslo_messaging._drivers.impl_rabbit [req-e3650168-e7d4-4d2b-8a44-7c9f504f3248 - - - - -] Connecting to AMQP server on 192.168.12.20:5672
2015-12-30 09:46:21.763 2624 INFO oslo_messaging._drivers.impl_rabbit [req-e3650168-e7d4-4d2b-8a44-7c9f504f3248 - - - - -] Connected to AMQP server on 192.168.12.20:5672
2015-12-30 09:46:21.826 2624 WARNING oslo_config.cfg [req-e3650168-e7d4-4d2b-8a44-7c9f504f3248 - - - - -] Option "lock_path" from group "DEFAULT" is deprecated. Use option "lock_path" from group "oslo_concurrency".
2015-12-30 09:46:26.059 2624 WARNING nova.compute.resource_tracker [req-0833d144-8cd2-48b4-a6f8-d23fe59fe102 - - - - -] No compute node record for COMPUTE-NODE1:COMPUTE-NODE1
2015-12-30 09:46:26.282 2624 INFO oslo_messaging._drivers.impl_rabbit [req-0833d144-8cd2-48b4-a6f8-d23fe59fe102 - - - - -] Connecting to AMQP server on 192.168.12.20:5672
2015-12-30 09:46:26.299 2624 INFO oslo_messaging._drivers.impl_rabbit [req-0833d144-8cd2-48b4-a6f8-d23fe59fe102 - - - - -] Connected to AMQP server on 192.168.12.20:5672

The root cause seems I tried using MTU 9000. When I use the default MTU=1500, it works well. But whenever I use MTU 9000, it doesn't work.

What I have updated:

dnsmasq.conf:
dhcp-option-force=26,8950

l3_agent.ini:
[DEFAULT]
network_device_mtu = 8950

ml2_conf.ini:
[ml2]
path_mtu = 8950
segment_mtu = 8950`

nova.conf:
[DEFAULT]
network_device_mtu = 8950

How could I enable mtu 9000?

I have two servers, one with Oracle Linux 7.1 as controller+network node, the other one with Windows 2012 R2 with Hyper-V as compute node. I have setup network node with VXLAN network type and installed openvswitch beta driver on hyper-v compute node. After launching an instance, everything on network node seems ok:

$ ovs-vsctl show
2c212252-1ea6-4b0d-9ead-c1f3513b9c16
    Bridge br-tun
        fail_mode: secure
        Port "vxlan-1"
            Interface "vxlan-1"
                type: vxlan
                options: {df_default="true", in_key=flow, local_ip="192.168.10.10", out_key=flow, remote_ip="192.168.10.11"}
        Port patch-int
            Interface patch-int
                type: patch
                options: {peer=patch-tun}
        Port br-tun
            Interface br-tun
                type: internal
    Bridge br-ex
        Port phy-br-ex
            Interface phy-br-ex
                type: patch
                options: {peer=int-br-ex}
        Port "eno1"
            Interface "eno1"
        Port br-ex
            Interface br-ex
                type: internal
    Bridge br-int
        fail_mode: secure
        Port "tapb40d133e-f1"
            tag: 1
            Interface "tapb40d133e-f1"
                type: internal
        Port int-br-ex
            Interface int-br-ex
                type: patch
                options: {peer=phy-br-ex}
        Port patch-tun
            Interface patch-tun
                type: patch
                options: {peer=patch-int}
        Port br-int
            Interface br-int
                type: internal

But there is no port created on compute node:

PS C:\Users\administrator> ovs-vsctl show
8927e0d4-fadb-48d2-8d23-f882478dcc1d
    Bridge br-tun
        Port br-tun
            Interface br-tun
                type: internal
        Port "vxlan-1"
            Interface "vxlan-1"
                type: vxlan
                options: {df_default="true", in_key=flow, local_ip="192.168.10.11", out_key=flow, remote_ip="192.168.10.10"}
        Port "external.1"
            Interface "external.1"

Is this a bug of ported openvswitch or there are any configurations?

I ever had a similar problem: http://ask.cloudbase.it/question/587/.... And also there is a good video for neutron troubleshooting here: https://www.openstack.org/summit/open.... I resolved my problem by following the steps.

Hi I installed Openstack Hyper-v compute driver with FreeRDP selected. After installation, the compute node works well, but I can't find related config files under 'c:\program files (x86)\cloudbase solutions'. No response from 8000 port too. Does this mean FreeRDP is not installed? How to fix it? Thanks for any help.

I finally figure it out. My problem here is my physical switch connected to the interface used for VM sets up a VLAN range which mismatches the vlan range I used when installing openstack:

ML2_VLAN_RANGES=default:671:690

Another possibility is iptable rules that prohibit icmp request and you need delete them:

# iptables-save | grep icmp
......
-A INPUT -j REJECT --reject-with icmp-host-prohibited
-A FORWARD -j REJECT --reject-with icmp-host-prohibited

I setup a controller+network node by DevStack with VLAN network type using Kilo version(networking_hyperv is installed), and a hyper-v compute node by cloudbase hyper-v compute driver(Kilo version too). Everything works well except that instances cannot get fixed ip.

>Get-VMNetworkAdapterVlan -VMNetworkAdapterName ec49017e-6d5b-489e-80ff-a311a67dee07 -ErrorAction Ignore

VMName            VMNetworkAdapterName                 Mode   VlanList
------            --------------------                 ----   --------
instance-00000007 ec49017e-6d5b-489e-80ff-a311a67dee07 Access 27

And there are below error messages in neutron-hyperv-agent.log:

ERROR hyperv.neutron.hyperv_neutron_agent [req-f284ee93-0de6-47ad-842d-cfafe7dda8cb ] Error in agent event loop
TRACE hyperv.neutron.hyperv_neutron_agent Traceback (most recent call last):
TRACE hyperv.neutron.hyperv_neutron_agent   File "C:\Program Files (x86)\Cloudbase Solutions\OpenStack\Nova\Python27\lib\site-packages\hyperv\neutron\hyperv_neutron_agent.py", line 356, in daemon_loop
TRACE hyperv.neutron.hyperv_neutron_agent     sync = self._process_network_ports(port_info)
TRACE hyperv.neutron.hyperv_neutron_agent   File "C:\Program Files (x86)\Cloudbase Solutions\OpenStack\Nova\Python27\lib\site-packages\hyperv\neutron\hyperv_neutron_agent.py", line 332, in _process_network_ports
TRACE hyperv.neutron.hyperv_neutron_agent     resync_a = self._treat_devices_added(port_info['added'])
TRACE hyperv.neutron.hyperv_neutron_agent   File "C:\Program Files (x86)\Cloudbase Solutions\OpenStack\Nova\Python27\lib\site-packages\hyperv\neutron\hyperv_neutron_agent.py", line 296, in _treat_devices_added
TRACE hyperv.neutron.hyperv_neutron_agent     device_details['admin_state_up'])
TRACE hyperv.neutron.hyperv_neutron_agent   File "C:\Program Files (x86)\Cloudbase Solutions\OpenStack\Nova\Python27\lib\site-packages\hyperv\neutron\hyperv_neutron_agent.py", line 264, in _treat_vif_port
TRACE hyperv.neutron.hyperv_neutron_agent     physical_network, segmentation_id)
TRACE hyperv.neutron.hyperv_neutron_agent   File "C:\Program Files (x86)\Cloudbase Solutions\OpenStack\Nova\Python27\lib\site-packages\hyperv\neutron\hyperv_neutron_agent.py", line 195, in _port_bound
TRACE hyperv.neutron.hyperv_neutron_agent     self._utils.connect_vnic_to_vswitch(map['vswitch_name'], port_id)
TRACE hyperv.neutron.hyperv_neutron_agent   File "C:\Program Files (x86)\Cloudbase Solutions\OpenStack\Nova\Python27\lib\site-packages\hyperv\neutron\utilsv2.py", line 86, in connect_vnic_to_vswitch
TRACE hyperv.neutron.hyperv_neutron_agent     self._add_virt_resource(vm, port)
TRACE hyperv.neutron.hyperv_neutron_agent   File "C:\Program Files (x86)\Cloudbase Solutions\OpenStack\Nova\Python27\lib\site-packages\hyperv\neutron\utilsv2.py", line 100, in _add_virt_resource
TRACE hyperv.neutron.hyperv_neutron_agent     self._check_job_status(ret_val, job_path)
TRACE hyperv.neutron.hyperv_neutron_agent   File "C:\Program Files (x86)\Cloudbase Solutions\OpenStack\Nova\Python27\lib\site-packages\hyperv\neutron\utils.py", line 137, in _check_job_status
TRACE hyperv.neutron.hyperv_neutron_agent     raise HyperVException(msg=_('Job failed with error %d') % ret_val)
TRACE hyperv.neutron.hyperv_neutron_agent HyperVException: Hyper-V Exception: Job failed with error 32775

Could anyone help on this?