settingsLogin | Registersettings

[Openstack] No ping to Openstack instance

0 votes

Dear all,

I installed Openstack Juno using two KVM virtual machines, one
controller and one Compute, the neutron service is installed on the
controller. Everything is working without any problem, I activated the
masquerade on both nodes using UFW.

I have two nic on each vm, one for internal networking (10.0.17.0/24)
and one for external one.

I created an internal network and run an instance, the instance got an
IP from the correct range and the state is running, but there is no ping
to it from the controller or compute vm.

From compute when I am using: virsh list I am seeing the instance
running there.

When I am doing virsh dumpxml I am not seeing the IP which is shown on
HORIZON!!! which is strange.

Could you please help me to identify the problem and solve it.

Thank you very much in advance.

Best regards,
Wahi


Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
asked Sep 7, 2017 in openstack by wahi (220 points)  

5 Responses

0 votes

When I accessed cirros instance and checked the network file it was on
dhcp mode, but ip a doesn't show any given IP only the local IP. Does it
mean that the instance cannot get a DHCP IP ?

responded Sep 7, 2017 by wahi (220 points)  
0 votes

Whether you've switched on DHCP-server on the subnet which VM belongs to?
If yes - check logs of DHCP-server (dnsmasq) on the node where it
resides - whether it receives DHCP requests and whether it sends answers
or why it rejects requests.
If DHCP-server don't receives requests or if it sends answers and VM
don't receives them - check using tcpdump entire chain of ports between
VM and DHCP server and find where it drops DHCP exchange.

On 9/7/17 11:42 AM, wahi wrote:
When I accessed cirros instance and checked the network file it was on
dhcp mode, but ip a doesn't show any given IP only the local IP. Does
it mean that the instance cannot get a DHCP IP ?


Mailing list:
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to     : openstack@lists.openstack.org
Unsubscribe :
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

--
Volodymyr Litovka
"Vision without Execution is Hallucination." -- Thomas Edison

responded Sep 7, 2017 by Volodymyr_Litovka (1,100 points)   1 6 8
0 votes

Hi Wahi,
Why you want to ping the instance connected to internak net from controller
node or compute node?

The easiest way to access the instance is to use VNC console from horizon
to access the VM.
If you need IP reachability from your controller / compute hosts (ingress),
you need to assign floating IP to the instance.

On Sep 7, 2017 11:15 AM, "wahi" wahi@sci.am wrote:

Dear all,

I installed Openstack Juno using two KVM virtual machines, one controller
and one Compute, the neutron service is installed on the controller.
Everything is working without any problem, I activated the masquerade on
both nodes using UFW.

I have two nic on each vm, one for internal networking (10.0.17.0/24) and
one for external one.

I created an internal network and run an instance, the instance got an IP
from the correct range and the state is running, but there is no ping to it
from the controller or compute vm.

From compute when I am using: virsh list I am seeing the instance running
there.

When I am doing virsh dumpxml I am not seeing the IP which is shown on
HORIZON!!! which is strange.

Could you please help me to identify the problem and solve it.

Thank you very much in advance.

Best regards,
Wahi


Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
responded Sep 7, 2017 by Fawaz_Mohammed (660 points)  
0 votes

check below

On 9/7/17 12:02 AM, wahi wrote:
Dear all,

I installed Openstack Juno using two KVM virtual machines, one
controller and one Compute, the neutron service is installed on the
controller. Everything is working without any problem, I activated the
masquerade on both nodes using UFW.

I have two nic on each vm, one for internal networking (10.0.17.0/24)
and one for external one.

I created an internal network and run an instance, the instance got an
IP from the correct range and the state is running, but there is no
ping to it from the controller or compute vm.

From compute when I am using: virsh list I am seeing the instance
running there.

When I am doing virsh dumpxml I am not seeing the IP which is shown on
HORIZON!!! which is strange.
Not strange.. Horizon just assigns the ip address that the machine will
eventually get.. If DHCP etc is not working the machine will not get any
ip.. once fixed it will get the IP as shown on the dashboard.

Could you please help me to identify the problem and solve it.
Networking issues!
OVS check.

Thank you very much in advance.

Best regards,
 Wahi


Mailing list:
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to     : openstack@lists.openstack.org
Unsubscribe :
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack


Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
responded Sep 7, 2017 by Remo_Mattei (10,500 points)   1 2 4
0 votes

Thank you for your reply I will check and let you know.

On 09/07/2017 07:38 PM, Remo Mattei wrote:
check below

On 9/7/17 12:02 AM, wahi wrote:

Dear all,

I installed Openstack Juno using two KVM virtual machines, one
controller and one Compute, the neutron service is installed on the
controller. Everything is working without any problem, I activated the
masquerade on both nodes using UFW.

I have two nic on each vm, one for internal networking (10.0.17.0/24)
and one for external one.

I created an internal network and run an instance, the instance got an
IP from the correct range and the state is running, but there is no
ping to it from the controller or compute vm.

From compute when I am using: virsh list I am seeing the instance
running there.

When I am doing virsh dumpxml I am not seeing the IP which is shown on
HORIZON!!! which is strange.
Not strange.. Horizon just assigns the ip address that the machine will
eventually get.. If DHCP etc is not working the machine will not get any
ip.. once fixed it will get the IP as shown on the dashboard.
Could you please help me to identify the problem and solve it.
Networking issues!
OVS check.

Thank you very much in advance.

Best regards,
Wahi


Mailing list:
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe :
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

--
Wahi Narsisian
Vice-Head of Center for Scientific Computing,
Institute for Informatics and Automation Problems,
National Academy of Sciences of the Republic of Armenia
1, P. Sevak str., Yerevan 0014, Armenia
t: 374 91 557285
e: wahi@sci.am

responded Sep 7, 2017 by wahi (220 points)  
...