danardelean's profile - activity

Hi, If the original host is still up you could try to migrate the VM with 'nova migrate <vm>.

Hi, If the original host is still up you could try to migrate the VM with 'nova migrate <vm> <destination host

Hi, If the original host is still up you could try to migrate the VM with 'nova migrate <vm> <destination host

Hi, Can you please tell me what OpenStack release you have? Also, for VMs to migrate from one host to another, a cluste

Hi, Please paste the nova commands outputs in something like https://paste.ubuntu.com/ Thank you.

Hello awestin1. Seem like this is the nova config file from the controller side. Can you also please paste nova.conf from the Hyper-V side? It should be in 'C:\Program Files\Cloudbase Solutions\OpenStack\Nova\etc'. Thanks, Dan.

Depending on how you installed the MSI, the logs should be in 'C:\OpenStack\Log'. Please open nova-compute log file and paste its contents in http://paste.ubuntu.com.

Please paste the nova logs from hyperv. Please use something like http://paste.ubuntu.com Thanks.

Hi,

Please make sure that "Enable MAC address spoofing" is enabled on the openstack-controller VM network adapter in which br-eth4 is connected. You find this in Advanced Features tab.

Thank you.

Hi,

If you don't want to use the openstack-controller VM but still keep it, you can just shut it down.

For the service down issue, please open powershell as Administrator and run the following commands "Restart-Service nova-compute" and "Restart-Service neutron-hyperv-agent".

We do not have a mobile phone version for v-magine. You can still use RDP to connect to the Windows host.

Thank you.

Please edit the cinder.conf file and set debug=true and verbose=true. Who is IP 192.168.1.80?

Hi, after the error occurs, what is the state of the VM? Does is it boot eventually? What about if you reboot it? Thanks.

Hi. Could you please repaste your cinder configuration file and the errors in something like http://paste.openstack.org/? It would do my work a lot easier. Thank you.

This seems to be a Windows Server 2016 host issue. Does it have the latest updates?

Right after the instance fails, if you do a "echo 'list disk' | diskpart" in powershell, what does it say? At VM creation, chose a distinct volume size so you can easily see the disk in diskpart.

Seems like hyper-v can connect to target but cannot attach it to the VM. In powershell please try 'echo "san" | diskpart'. My suspicion is that san policy is online, which should be offline. If this is the case, try to put it offline 'san POLICY=OfflineShared'.

The LAN gw IP must be set on the public subnet. If you do a 'neutron subnet-show [ID]' there is a field gateway_ip, here is where you need to set LAN gw.

So, outside OpenStack env, you have a LAN 192.168.1.0/24 with 192.168.1.1 as default gw, right? If this is the case, the OpenStack router must NOT have 192.168.1.1 IP on its own interface since that is the LAN gw. However, the router must have an interface with an IP in that range.

You are using the stantard LVM cinder backed or using an added FC backend?

Please check if you have an external bridge connected to the wlan0 interface (you can do 'ovs-vsctl show' to check if the bridge exists and has a port in wlan0). In /etc/neutron/plugins/ml2/ml2_config.ini check if a network is mapped to that bridge.

Hi. Did you use Hyper-V for compute? If yes, did you use OVS or hyperv-networking? Make sure the the bridge mappings are correctly configured. Information about host network configuration would be useful also.

Please try to restart rabbitmq and neutron. Check their status after. If this is not working check rabbitmq logs and paste them please.

Hi, thanks for the logs. Looks like the neutron DHCP service could not connect to the AMQP server. Took a look at the ports opened paste and could not find port 5672 listening, that is the AMQP port. Please do a 'netstat -tanpu | grep -i listen | grep 5672' and 'systemctl status rabbitmq-server'.

Please provide us the neturon logs. Also please check the state of the neutron services, like 'systemctl status neutron-server.service'. Thank you.

Looks like no one was listening on port 9696 before reboot. There should have been neutron. Good to hear it works now. If you have any issues in the future please let us know.

Can you please do a nova boot and a 'neutron net-list' from CLI? I want to see if this is an isolated issues with Horizon, something wrong with keystone, or the DB. Thanks.

A fix has been released, with updated version 2016.1.0.4. As a workaround, try to used only alphanumeric chars for the password, no special chars.

Hi, sorry for the delay but we were not able to reproduce this for a while. It appears to be an RDO bug, the issue was related to escaping issues when calling Bash on the Linux controller from Python with some special chars. The characters affected are the following: \ ! "

Hi, the timeout issue can occur during the RDO VM reboot. This can happen if the VM is slow on reboot, as the v-magine installer tries to reconnect the the VM, it times out. We had this in the past and we increased the time out. Is this the case, the VM is slow on reboot?

The second log is unavailable. It would be great if you could re-paste it. Also, please check if there is a pyboot log file and paste it aswell. Thanks.

You can copy the contents of the two files and paste them, separately, to http://paste.openstack.org/ for example. Please be careful to mask any passwords in the log files.

In the install folder, do you have v-magine.log and v-magine-console.log populated? If yes, it would be great if you could share that with me. Thanks.

Hi Markus, sorry to hear you have issues with v-magine. In the install folder there are 2 .log files, v-magine.log and v-magine-console.log. It would be great if you could share those 2 files with me so I can look for issues there. Also, what version of Windows do you use? Thanks.

During the puppet run everything was OK? Can you please freeze the VM (pause in Hyper-V) right after the RDO puppet is done and look for the /etc/nova? That directory shoud be there right after the RDO finishes. Thank you.

Seems like config variables were not passed on. Please connect to the VM and see if there is a ssh key in ~/.ssh/ and if there is a in /etc/nova/nova.conf config file. This happens even after several retries?

Please make sure you are using Windows Server 2012r2 and have Hyper-V enabled on it.

I will ask you to add this code snippet http://paste.openstack.org/show/585463/ in v-magine installer folder -> resources -> install-rdo.sh. Add it on line 202, right after '#disable_network_manager' line.

Hi. What version of Windows do you have? Do you have a v-magine process running after running the installer?

Thanks for the details. I will try to reproduce this and come up with solutions. Will let you know the status.

What if you try ro run this command 'ntpdate pool.ntp.org' ?

Hi. Thanks for the feedback. We will discuss these features.

Hi. On what Windows are your trying this? In the installer folder are 2 log files, can you please provide me the output? Also, on GUI right after the error there should be a more verbose explanation on why it failed. Thanks.

Perfect, glad to hear it worked!

So there are two possible networks. The one that on which the Windows host is, and the one you specify during the install. The one important for us is the first one. Make sure you have an active DHCP on that network.

/etc/nova and the ssh key should exist at that step of the install process. Please make sure to not delete the deployment when the error occurs. Better, when the GUI throws the error, do not touch it but go straight to command line. This way we can better isolate the problem. Thanks.

Hi, the installer should retry the install process after this error so just let it run. I've seen this before and the solution was to just let it retry.

The IP must the external IP of the VM, 'mgmt-ext' interface in 'ip a' e.g. "python script.sh 10.7.1.97 root admin"