[Openstack-operators] Deploying openstack with current vlans
Juan José Pavlik Salles
jjpavlik at gmail.com
Thu Apr 4 19:43:10 UTC 2013
I made a mistak when copied the subnets to the mail, my subnets are:
root at cocinero:~# quantum port-list
+--------------------------------------+------+-------------------+---------------------------------------------------------------------------------------+
| id | name | mac_address |
fixed_ips
|
+--------------------------------------+------+-------------------+---------------------------------------------------------------------------------------+
| 1d024fd1-1a9b-425f-aaed-4475f988d0cb | | fa:16:3e:ae:34:fe |
{"subnet_id": "aca36e3b-ae10-4367-8843-3cae7034d114", "ip_address":
"172.16.254.2"} |
| a642220c-e315-4aee-bc54-2fca2ba0103b | | fa:16:3e:61:fc:d3 |
{"subnet_id": "21955b15-453d-46de-8e15-c285b68c0767", "ip_address":
"172.19.136.100"} |
+--------------------------------------+------+-------------------+---------------------------------------------------------------------------------------+
Sorry,
2013/4/4 Juan José Pavlik Salles <jjpavlik at gmail.com>
> Hi robert, this last 2 days i've been trying to fix the problem i
> mentioned before, but didn't succeed.
>
> This is what i've done:
>
> -In the quantum plugin file:
> ...
> [OVS]
> tenant_network_type = vlan
> network_vlan_ranges = physnet1:1:4094
> integration_bridge = br-int
> bridge_mappings = physnet1:br-eth0
> ...
>
> -Configured my switch to trunk all vlans on eth0 of the compute node
> server (named asado).
>
> -Created a provider network:
> quantum net-create --tenant-id ad60fae6c2b3496684b96b013a0e64a3 vlan5
> --provider:network_type vlan --provider:physical_network physnet1
> --provider:segmentation_id 5
>
> -Created the subnet for this network:
> quantum subnet-create --tenant-id ad60fae6c2b3496684b96b013a0e64a3 vlan5
> 172.16.254.0/24 --enable_dhcp=False
>
> -Started a vm with horizon
>
> The VM starts fine, but there's not network access to it.
>
> I've checked security groups, and allowed all type of traffic to the
> machine in the default policy.
>
> When i checked ovs in the compute node, this is what i see:
>
> root at asado:~# ovs-vsctl show
> 60d97036-7718-4dbe-b744-48933a3041f5
> Bridge br-int
> Port "qvo1d024fd1-1a"
> tag: 4
> Interface "qvo1d024fd1-1a"
> Port br-int
> Interface br-int
> type: internal
> Port "int-br-eth0"
> Interface "int-br-eth0"
> Bridge "br-eth0"
> Port "br-eth0"
> Interface "br-eth0"
> type: internal
> Port "phy-br-eth0"
> Interface "phy-br-eth0"
> Port "eth0"
> Interface "eth0"
> ovs_version: "1.4.0+build0"
> root at asado:~#
>
> The tag attribute doesn't make sense, i think it should be 5 instead of 4,
> so i changed it:
>
> root at asado:~# ovs-vsctl set Port qvo1d024fd1-1a tag=5
> root at asado:~# ovs-vsctl show
> 60d97036-7718-4dbe-b744-48933a3041f5
> Bridge br-int
> Port "qvo1d024fd1-1a"
> tag: 5
> Interface "qvo1d024fd1-1a"
> Port br-int
> Interface br-int
> type: internal
> Port "int-br-eth0"
> Interface "int-br-eth0"
> Bridge "br-eth0"
> Port "br-eth0"
> Interface "br-eth0"
> type: internal
> Port "phy-br-eth0"
> Interface "phy-br-eth0"
> Port "eth0"
> Interface "eth0"
> ovs_version: "1.4.0+build0"
> root at asado:~#
>
> But still no ping to the machine, i'm really confused. i'm pretty sure i
> followed the steps correctly what am i missing???
>
> I assume that this new port on my br-int is the one created by quantum for
> the new vm, but why does it have a wrong tag? One more thing, this tag
> seems to be auto incremental, because the first VM i create had tag=1 the
> next one tag=2 and so on.
>
> I still can't fin the connection between br-int and br-eth0 (the bridge on
> eth0 interface, the one with the trunk).
>
> Here i'll paste some more info that may help :D. The controller node is
> called cocinero.
>
> root at cocinero:~# quantum net-list
>
> +--------------------------------------+---------+--------------------------------------+
> | id | name | subnets
> |
>
> +--------------------------------------+---------+--------------------------------------+
> | fde5024e-8457-4a4d-af2a-bd8afbdea67e | vlan5 |
> aca36e3b-ae10-4367-8843-3cae7034d114 |
>
> +--------------------------------------+---------+--------------------------------------+
>
> root at cocinero:~# quantum net-show vlan5
> +---------------------------+--------------------------------------+
> | Field | Value |
> +---------------------------+--------------------------------------+
> | admin_state_up | True |
> | id | fde5024e-8457-4a4d-af2a-bd8afbdea67e |
> | name | vlan5 |
> | provider:network_type | vlan |
> | provider:physical_network | physnet1 |
> | provider:segmentation_id | 5 |
> | router:external | False |
> | shared | False |
> | status | ACTIVE |
> | subnets | aca36e3b-ae10-4367-8843-3cae7034d114 |
> | tenant_id | ad60fae6c2b3496684b96b013a0e64a3 |
> +---------------------------+--------------------------------------+
>
> root at cocinero:~# quantum subnet-list
>
> +--------------------------------------+------+-----------------+------------------------------------------------------+
> | id | name | cidr |
> allocation_pools |
>
> +--------------------------------------+------+-----------------+------------------------------------------------------+
> | 21955b15-453d-46de-8e15-c285b68c0767 | | 172.19.136.0/24 |
> {"start": "172.19.136.100", "end": "172.19.136.200"} |
>
> +--------------------------------------+------+-----------------+------------------------------------------------------+
> root at cocinero:~# quantum port-list
>
> +--------------------------------------+------+-------------------+---------------------------------------------------------------------------------------+
> | id | name | mac_address |
> fixed_ips
> |
>
> +--------------------------------------+------+-------------------+---------------------------------------------------------------------------------------+
> | 1d024fd1-1a9b-425f-aaed-4475f988d0cb | | fa:16:3e:ae:34:fe |
> {"subnet_id": "aca36e3b-ae10-4367-8843-3cae7034d114", "ip_address":
> "172.16.254.2"} |
>
> +--------------------------------------+------+-------------------+---------------------------------------------------------------------------------------+
>
> root at cocinero:~# quantum port-show 1d024fd1-1a9b-425f-aaed-4475f988d0cb
>
> +----------------+-------------------------------------------------------------------------------------+
> | Field | Value
> |
>
> +----------------+-------------------------------------------------------------------------------------+
> | admin_state_up | True
> |
> | device_id | ee2b23cf-1c6d-4234-a972-d64d6065990f
> |
> | device_owner | compute:nova
> |
> | fixed_ips | {"subnet_id": "aca36e3b-ae10-4367-8843-3cae7034d114",
> "ip_address": "172.16.254.2"} |
> | id | 1d024fd1-1a9b-425f-aaed-4475f988d0cb
> |
> | mac_address | fa:16:3e:ae:34:fe
> |
> | name |
> |
> | network_id | fde5024e-8457-4a4d-af2a-bd8afbdea67e
> |
> | status | ACTIVE
> |
> | tenant_id | ad60fae6c2b3496684b96b013a0e64a3
> |
>
> +----------------+-------------------------------------------------------------------------------------+
>
>
>
>
> 2013/4/2 Robert van Leeuwen <Robert.vanLeeuwen at spilgames.com>
>
>> > We already have many different vlans, one for each department. Some of
>> them have public ip networks and some private networks.
>> > For instance when a department ask for a virtual server we run it in
>> our IP network,
>> > when if we were using vlans in xen we could run it on the departments
>> network lets say vlan 3.
>> > what we want is to use this vlans we already have assigned to each
>> department in our openstack deploy
>>
>> Juan,
>>
>> You can use the bridge-mapped network in Quantum to do this.
>> This will create a network in openstack which is mapped to a physical
>> network/vlan.
>> So if basically create a network for each tenant with its own vlan
>> mapping to the appropriate network.
>>
>> This link might help you setting things up:
>> https://wiki.openstack.org/wiki/ConfigureOpenvswitch
>>
>> I think you can also do this with the "old-style" networking without
>> quantum/openvswitch.
>> If you have the time I would invest it in Quantum though, it adds a lot
>> of flexibility and as long as you are still in the setup phase you have
>> some time to play around with it :)
>>
>> Cheers,
>> Robert van Leeuwen
>> _______________________________________________
>> OpenStack-operators mailing list
>> OpenStack-operators at lists.openstack.org
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators
>>
>
>
>
> --
> Pavlik Juan José
>
--
Pavlik Juan José
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-operators/attachments/20130404/3ef7347e/attachment.html>
More information about the OpenStack-operators
mailing list