openstack team mailing list archive

["Chathura M. Sarathchandra Magurawalage" <77.chathura@xxxxxxxxx>]

Hello Jean,

Thanks for your reply.

I followed the instructions in
http://docs.openstack.org/folsom/basic-install/content/basic-install_network.html.
And my Controller and the Network-node is installed in the same physical
node.

I am using Folsom but without Network namespaces.

But in the website you have provided it states that "If you run both L3 +
DHCP services on the same node, you should enable namespaces to avoid
conflicts with routes :"

But currently quantum-dhcp-agent and quantum-l3-agent are running in the
same node?

Additionally the control node serves as a DHCP server for the local network
( Don't know if that would make and difference)

Any idea what the problem could be?


On 16 February 2013 16:21, Jean-Baptiste RANSY <
jean-baptiste.ransy@xxxxxxxxxx> wrote:

>  Hello Chathura,
>
> Are you using Folsom with Network Namespaces ?
>
> If yes, have a look here :
> http://docs.openstack.org/folsom/openstack-network/admin/content/ch_limitations.html
>
>
> Regards,
>
> Jean-Baptsite RANSY
>
>
>
> On 02/16/2013 05:01 PM, Chathura M. Sarathchandra Magurawalage wrote:
>
> Hello guys,
>
>  The problem still exists. Any ideas?
>
>  Thanks
>
>   On 15 February 2013 14:37, Sylvain Bauza <sylvain.bauza@xxxxxxxxxxxx>wrote:
>
>> Metadata API allows to fetch SSH credentials when booting (pubkey I mean).
>> If a VM is unable to reach metadata service, then it won't be able to get
>> its public key, so you won't be able to connect, unless you specifically go
>> thru a Password authentication (provided password auth is enabled in
>> /etc/ssh/sshd_config, which is not the case with Ubuntu cloud archive).
>> There is also a side effect, the boot process is longer as the instance
>> is waiting for the curl timeout (60sec.) to finish booting up.
>>
>> Re: Quantum, the metadata API is actually DNAT'd from Network node to the
>> Nova-api node (here 172.16.0.1 as internal management IP) :
>> Chain quantum-l3-agent-PREROUTING (1 references)
>>
>> target     prot opt source               destination
>>  DNAT       tcp  --  0.0.0.0/0            169.254.169.254      tcp
>> dpt:80 to:172.16.0.1:8775
>>
>>
>> Anyway, the first step is to :
>> 1. grab the console.log
>> 2. access thru VNC to the desired instance
>>
>> Troubleshooting will be easier once that done.
>>
>> -Sylvain
>>
>>
>>
>> Le 15/02/2013 14:24, Chathura M. Sarathchandra Magurawalage a écrit :
>>
>>>  Hello Guys,
>>>
>>> Not sure if this is the right port but these are the results:
>>>
>>>  *Compute node:*
>>>
>>>
>>> root@computenode:~# netstat -an | grep 8775
>>>  tcp        0      0 0.0.0.0:8775 <http://0.0.0.0:8775>  0.0.0.0:*
>>>           LISTEN
>>>
>>> *Controller: *
>>>
>>>
>>> root@controller:~# netstat -an | grep 8775
>>>  tcp        0      0 0.0.0.0:8775 <http://0.0.0.0:8775>  0.0.0.0:*
>>>           LISTEN
>>>
>>> *Additionally I cant curl 169.254.169.254 from the compute node. I am
>>> not sure if this is related to not being able to PING the VM.*
>>>
>>>
>>> curl -v http://169.254.169.254
>>> * About to connect() to 169.254.169.254 port 80 (#0)
>>> *   Trying 169.254.169.254...
>>>
>>> Thanks for your help
>>>
>>>
>>>
>>> -----------------------------------------------------------------------------------------------------------------------------
>>> Chathura Madhusanka Sarathchandra Magurawalage.
>>> 1NW.2.1, Desk 2
>>> School of Computer Science and Electronic Engineering
>>> University Of Essex
>>> United Kingdom.
>>>
>>>  Email: csarata@xxxxxxxxxxx <mailto:csarata@xxxxxxxxxxx>
>>>           chathura.sarathchandra@xxxxxxxxx <mailto:77.chathura@xxxxxxxxx
>>> >
>>> 77.chathura@xxxxxxxxx <mailto:77.chathura@xxxxxxxxx>
>>>
>>>
>>>
>>> On 15 February 2013 11:03, Anil Vishnoi <vishnoianil@xxxxxxxxx <mailto:
>>> vishnoianil@xxxxxxxxx>> wrote:
>>>
>>>     If you are using ubuntu cloud image then the only way to log-in is
>>>     to do ssh with the public key. For that you have to create ssh key
>>>     pair and download the ssh key. You can create this ssh pair using
>>>     horizon/cli.
>>>
>>>
>>>     On Fri, Feb 15, 2013 at 4:27 PM, Sylvain Bauza
>>>      <sylvain.bauza@xxxxxxxxxxxx <mailto:sylvain.bauza@xxxxxxxxxxxx>>
>>>
>>>     wrote:
>>>
>>>
>>>         Le 15/02/2013 11:42, Chathura M. Sarathchandra Magurawalage a
>>>         écrit :
>>>
>>>
>>>             How can I log into the VM from VNC? What are the credentials?
>>>
>>>
>>>         You have multiple ways to get VNC access. The easiest one is
>>>         thru Horizon. Other can be looking at the KVM command-line for
>>>         the desired instance (on the compute node) and check the vnc
>>>         port in use (assuming KVM as hypervisor).
>>>         This is basic knowledge of Nova.
>>>
>>>
>>>
>>>             nova-api-metadata is running fine in the compute node.
>>>
>>>
>>>         Make sure the metadata port is avaible thanks to telnet or
>>>         netstat, nova-api can be running without listening on metadata
>>>         port.
>>>
>>>
>>>
>>>
>>>         _______________________________________________
>>>         Mailing list: https://launchpad.net/~openstack
>>>          <https://launchpad.net/%7Eopenstack>
>>>         Post to     : openstack@xxxxxxxxxxxxxxxxxxx
>>>         <mailto:openstack@xxxxxxxxxxxxxxxxxxx>
>>>         Unsubscribe : https://launchpad.net/~openstack
>>>         <https://launchpad.net/%7Eopenstack>
>>>
>>>         More help   : https://help.launchpad.net/ListHelp
>>>
>>>
>>>
>>>
>>>     --     Thanks & Regards
>>>     --Anil Kumar Vishnoi
>>>
>>>
>>>
>>
>
>
> _______________________________________________
> Mailing list: https://launchpad.net/~openstack
>
> Post to     : openstack@xxxxxxxxxxxxxxxxxxx
> Unsubscribe : https://launchpad.net/~openstack
> More help   : https://help.launchpad.net/ListHelp
>
>
>