[Openstack] Network Connectivity issue

Nicolas Odermatt odermattn at gmail.com
Fri Mar 23 08:17:45 UTC 2012


Thank you very much for your answer!

There was indeed a missing route. I added the following routes:

Client: sudo route add -net 192.168.163.128 netmask 255.255.255.128 gw
192.168.163.20 dev eth2

Instanz:sudo route add -net 192.168.163.0 netmask 255.255.255.128 gw
10.0.0.1 dev eth0

And after that it worked like a charm ;)

2012/3/22 Diego Parrilla Santamaría <diego.parrilla.santamaria at gmail.com>

> Hi Nicolas, thanks for using our distro!
>
> Just to summarize:
> - Single node
> - Management network: 192.168.163.0/25 = 192.168.163.1 to
> 192.168.163.126. Let's assume gateway = 192.168.163.1
> - Service Network: 192.168.163.128/25 = 192.168.163.129 to
> 192.168.163.254. gateway should be = 192.168.163.129
>
> If you wan to ping from the service network to a host in the management
> network (and it's not the single node) you need a route to that network. Or
> you can configure your system to use the Management Network as the public
> network. Then you will reach your hosts.
>
> If you are using FlatDHCP, may be you can configure the smart installer to
> use as the service network a subset (not a subnet) of the management
> network. It's a dirty hack (I prefer to split networks always) we did in
> the past...
>
> For example if you want to setup a little lab in your lapto with the
> distro you can use this settings:
> - Single node
> - Management network: 192.168.10.0/24 (this is my network, use yours)
> - Service Network: 10.0.0.0/24
> - Public Network (floating range): 192.168.10.64/28 (and of course I
> don't have any host in this subnet)
>
> With this configuration floating IPs will reach hosts in the management
> network.
>
> Cheers!
> Diego
>
>
> --
> Diego Parrilla
> <http://www.stackops.com/>*CEO*
> *www.stackops.com | * diego.parrilla at stackops.com** | +34 649 94 43 29 |
> skype:diegoparrilla*
> * <http://www.stackops.com/>
> **
>
>
>
> On Thu, Mar 22, 2012 at 3:47 PM, Nicolas Odermatt <odermattn at gmail.com>wrote:
>
>> Hello guys, I am dealing with a strange phenomenon my StackOps
>> environment.
>>
>> The thing is like this:
>> I am running a single-node stackops deployment in a little /25 network.
>> The installation and the few configurations I had to do afterwards went
>> just fine, no problems so far. I was also able to upload an image to glance
>> and to start an instance using euca2ools. I authorized the necessary port
>> for ssh, allowed icmp and then connected to the instance with ssh.
>> Afterwards I checked the instance's connection to the internet, which was
>> up and running. Now to my so called "strange phenomenon": Although I am
>> able to ping my stackops-node, public ip adresses (e.g 8.8.8.8) and in
>> spite the fact that name resolving works as well, I can't ping a Client
>> machine in my local area network. My stackops-node has the ip-address
>> 192.168.163.20/25, the instance has the ip-address(192.168.163.130/25)
>> and the client has the ip-address 192.168.163.5/25.
>>
>> I can't think of an approach to this problem, as I have enabled ssh and
>> icmp and the connection between Client and stackops-node exists.
>> Is it possible that I might have forgott to add a certain rule?
>>
>> Thanks for your ideas!
>>
>> Best regards.
>> Nicolas
>> --
>> Freundliche Grüsse,
>> Nicolas Odermatt
>>
>>
>>
>> _______________________________________________
>> Mailing list: https://launchpad.net/~openstack
>> Post to     : openstack at lists.launchpad.net
>> Unsubscribe : https://launchpad.net/~openstack
>> More help   : https://help.launchpad.net/ListHelp
>>
>>
>


-- 
Freundliche Grüsse,
Nicolas Odermatt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20120323/c8b65c7e/attachment.html>


More information about the Openstack mailing list