Open Stack

Yes, I am able to ping the gateway address from within the snat namespace:

--
$ sudo ip netns exec snat-9e849e49-ed36-4280-a53c-47d6f5afbea2 ping
10.36.7.253
PING 10.36.7.253 (10.36.7.253) 56(84) bytes of data.
64 bytes from 10.36.7.253: icmp_seq=1 ttl=255 time=1.42 ms
64 bytes from 10.36.7.253: icmp_seq=2 ttl=255 time=0.685 ms
64 bytes from 10.36.7.253: icmp_seq=3 ttl=255 time=0.439 ms
^C
--- 10.36.7.253 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2000ms
rtt min/avg/max/mdev = 0.439/0.850/1.426/0.419 ms
--

On Mon, May 2, 2016 at 10:46 PM, Dileep Varma Bairraju <varma123 at gmail.com>
wrote:

> It seems like you have 5 tenants, correlating to 5 snat namespaces. Your
> 'qg-' interfaces have proper ip configured, within the snat namespaces,
> verify if you are able to resolve arp for '10.36.7.253'. From within the
> namespace try pinging gw.
>
> -Dileep
>
> On Mon, May 2, 2016 at 10:30 PM, Jagga Soorma <jagga13 at gmail.com> wrote:
>
>> We us a external vm network of 10.36.6.0/23.  Looks like I do have some
>> snat rules but no idea what I should be specifically looking for in here:
>>
>> $ ip netns | grep -i snat
>> snat-9e849e49-ed36-4280-a53c-47d6f5afbea2
>> snat-716dc7bd-9d6b-41da-aa6a-a484398785b1
>> snat-bece0591-c55b-4a48-bc2b-77873a3ebce1
>> snat-803e06a4-4499-4ce0-bda6-fb158e717b9e
>> snat-6e4669f9-0b63-4b60-bdf6-94037b4c1e23
>>
>>
>> $ sudo ip netns exec snat-9e849e49-ed36-4280-a53c-47d6f5afbea2 ip a |
>> grep "inet"
>>     inet 127.0.0.1/8 scope host lo
>>     inet6 ::1/128 scope host
>>     inet 192.168.5.4/24 brd 192.168.5.255 scope global sg-86abc456-8d
>>     inet6 fe80::f816:3eff:fe23:7166/64 scope link
>>     inet 10.36.6.240/23 brd 10.36.7.255 scope global qg-09e400d1-28
>>     inet6 fe80::f816:3eff:fe52:dc9a/64 scope link
>>
>>
>> $ sudo ip netns exec snat-bece0591-c55b-4a48-bc2b-77873a3ebce1 ip a |
>> grep "inet"
>>     inet 127.0.0.1/8 scope host lo
>>     inet6 ::1/128 scope host
>>     inet 192.168.8.4/24 brd 192.168.8.255 scope global sg-ec9b41fe-3b
>>     inet6 fe80::f816:3eff:feb5:a225/64 scope link
>>     inet 10.36.6.79/23 brd 10.36.7.255 scope global qg-b1f38a3f-0b
>>     inet6 fe80::f816:3eff:fe4b:4a1e/64 scope link
>>
>> On Mon, May 2, 2016 at 10:09 PM, Remo Mattei <remo at italy1.com> wrote:
>>
>>> not sure how you build your public network.. but usually it does not do
>>> dhcp. So those are details that are needed in order for us to give you
>>> solutions / options / checking etc based on what you are running, how it
>>> was configured etc..
>>>
>>> CentOS, Ubuntu, scripting just as an example..
>>>
>>> Remo
>>>
>>> On May 2, 2016, at 22:02, Jagga <jagga13 at gmail.com> wrote:
>>>
>>> That is what I thought but it does not seem to be working this way.  How
>>> would I check our snat namespace and what specifically should I be looking
>>> for?  My apologies but am very new to openstack.
>>>
>>> Thanks.
>>>
>>>
>>> On May 2, 2016, at 9:51 PM, Dileep Varma Bairraju <varma123 at gmail.com>
>>> wrote:
>>>
>>> Hi Jagga,
>>>
>>> I don't think that's the right approach.Floating ip will effectively do
>>> a 1:1 NAT for a given a vm to reach external resources. But, there should
>>> be a ip from the external network that gets assigned to SNAT namespace on
>>> network node, this effectively will let all vm's (without floating ip)
>>> access external resources.
>>>
>>> I'd suggest you check at your snat namespace for possible issues, as you
>>> seem to have patched the problem for that vm with floating ip's.
>>>
>>> > Is that by design or is there something wrong with our configuration?
>>> As per design, you don't need to assign floating ip's for your vm's to
>>> get out, this should be done by SNAT by default as mentioned earlier, where
>>> all the vm's internal ip space maps one external ip.
>>>
>>> Regards,
>>> Dileep
>>>
>>> On Mon, May 2, 2016 at 8:32 PM, Jagga Soorma <jagga13 at gmail.com> wrote:
>>>
>>>> Hi Guys,
>>>>
>>>> Need some clarification regarding routing for instances without a
>>>> floating ip address. Basically we have instances connected to a priv
>>>> network that is also connected to our external network and our security
>>>> group allows all egress traffic. However, we can't seem to get to any
>>>> resource on our external network till a floating ip address is assigned.
>>>> Once we assign a floating ip address we can get out.  Is that by design or
>>>> is there something wrong with our configuration?
>>>>
>>>> Thanks.
>>>>
>>>> _______________________________________________
>>>> Mailing list:
>>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>>> Post to     : openstack at lists.openstack.org
>>>> Unsubscribe :
>>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>>>
>>>>
>>>
>>>
>>> --
>>> Regards,
>>> Dileep V Bairraju
>>>
>>> !DSPAM:1,572831b2317776163816806!
>>> _______________________________________________
>>> Mailing list:
>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>> Post to     : openstack at lists.openstack.org
>>> Unsubscribe :
>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>>
>>>
>>> !DSPAM:1,572831b2317776163816806!
>>>
>>>
>>>
>>
>
>
> --
> Regards,
> Dileep V Bairraju
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20160502/2e99ebe4/attachment.html>