[Openstack] Quantum/Grizzy - Instance doesn't get IP
Nicolae Paladi
n.paladi at gmail.com
Fri Sep 6 15:23:47 UTC 2013
Hi Marcelo,
I might have a solution to the "not getting an IP address" problem, at
least it works for me;
so, briefly, I use network type VLAN (should not make a difference, not
sure) on 3 nodes atm
(one controller/network (A), + two compute hosts (B, C));
I went the "diagnostic" way, with trial and error and came to the
conclusion that my guests DO get an IP adress
when the dhcp-agent running on the respective compute host appears when in
the list for that net:
how to reproduce:
*quantum agent-list | grep DHCP*
(you get a table with DHCP agents on different hosts)
*quantum dhcp-agent-list-hosting-net <net1>* -- you see which
dhcp agents are "attached" (don't know how to express it otherwise) to
network
<net1>.
So, when a guest is scheduled on <net1> and on host B and the dhcp-agent
running on host B appears in
the list when calling, quantum dhcp-agent-list-hosting-net <net1>, then the
instance gets the IP address.
If if it's not listed, the instance will not get an Ip address.
calling
*quantum dhcp-agent-network-add <dhcp-agent-on-host-B> <net1> *
fixes the problem for me.
These are just observations, I don't know how the causes and what is the
"normal" behavior.
Also, even though they do get the IP, I can't reach these instances from
the outside (I suspect
this has to do with the l3-agent).
/Nicolae.
On 6 September 2013 16:29, Marcelo Dieder <marcelodieder at gmail.com> wrote:
> The python-quantumclient also has different versions at the two guides.
>
> Marcelo Dieder
>
>
> On 09/06/2013 11:26 AM, Marcelo Dieder wrote:
>
> Hi Nicolae,
>
> Apparently I have not located the problem, but I followed the tutorial and
> now the quantum agent-list command worked.
>
> I made a comparison with the packages and guide
> http://docs.openstack.org/grizzly/basic-install/apt/content/basic-install_controller.htmlThe difference between the two seems to be in openvswitch packages and
> kernel version. Guide in DOCS OpenStack seems the packages are incompatible
> and commands such as quantum-agent that doesn't exist.
>
> Guide:
> https://github.com/mseknibilel/OpenStack-Grizzly-Install-Guide/blob/OVS_MultiNode/OpenStack_Grizzly_Install_Guide.rst
>
> Packages:
>
> quantum-plugin-openvswitch:amd64 (2013.1.2-0ubuntu1~cloud0, automatic)
> quantum-l3-agent:amd64 (2013.1.2-0ubuntu1~cloud0)
> quantum-dhcp-agent:amd64 (2013.1.2-0ubuntu1~cloud0)
> quantum-plugin-openvswitch-agent:amd64 (2013.1.2-0ubuntu1~cloud0)
> python-quantumclient:amd64 (2.2.0-0ubuntu1~cloud0, automatic)
> python-quantum:amd64 (2013.1.2-0ubuntu1~cloud0, automatic)
> quantum-metadata-agent:amd64 (2013.1.2-0ubuntu1~cloud0)
> quantum-common:amd64 (2013.1.2-0ubuntu1~cloud0, automatic)
> openvswitch-switch:amd64 (1.4.0-1ubuntu1.5)
> openvswitch-common:amd64 (1.4.0-1ubuntu1.5, automatic)
> openvswitch-datapath-dkms:amd64 (1.4.0-1ubuntu1.5)
>
> root at network:~# lsb_release -a
> No LSB modules are available.
> Distributor ID: Ubuntu
> Description: Ubuntu 12.04.3 LTS
> Release: 12.04
> Codename: precise
> root at network:~# uname -r
> 3.5.0-39-generic
>
> APT Sources
>
> deb http://ubuntu-cloud.archive.canonical.com/ubuntuprecise-updates/grizzly main
>
>
> Guide:
> http://docs.openstack.org/grizzly/basic-install/apt/content/basic-install_controller.html
>
> quantum-plugin-openvswitch:amd64 (2013.1.2-0ubuntu1~cloud0, automatic)
> quantum-l3-agent:amd64 (2013.1.2-0ubuntu1~cloud0)
> quantum-dhcp-agent:amd64 (2013.1.2-0ubuntu1~cloud0)
> quantum-plugin-openvswitch-agent:amd64 (2013.1.2-0ubuntu1~cloud0)
> python-quantumclient:amd64 (3.0.a1-1, automatic)
> python-quantum:amd64 (2013.1.2-0ubuntu1~cloud0, automatic)
> quantum-metadata-agent:amd64 (2013.1.2-0ubuntu1~cloud0, automatic)
> quantum-common:amd64 (2013.1.2-0ubuntu1~cloud0, automatic)
> openvswitch-switch:amd64 (1.4.2+git20120612-9.1~deb7u1)
> openvswitch-common:amd64 (1.4.2+git20120612-9.1~deb7u1, automatic)
> openvswitch-datapath-dkms:amd64 (1.4.2+git20120612-9.1~deb7u1)
>
> root at network:~# lsb_release -a
> No LSB modules are available.
> Distributor ID: Ubuntu
> Description: Ubuntu 12.04.3 LTS
> Release: 12.04
> Codename: precise
> root at network:~# uname -r
> 3.2.0-52-generic
>
> APT Sources
>
> deb http://ubuntu-cloud.archive.canonical.com/ubuntuprecise-updates/grizzly main
> deb http://archive.gplhost.com/debian grizzly main
> deb http://archive.gplhost.com/debian grizzly-backports main
>
> After I set everything up again with guide mseknibilel, the quantum
> agent-list works, and the instance can finally get the IP. An important
> issue to be checked is whether the network interfaces of the bridges are in
> promiscuous mode on the hypervisor. In my case I use XenServer 6.2, I had
> to enable promisc mode interfaces VIF and PIF.
>
> Do you have any logs with errors?
>
> Thanks for all.
>
> Marcelo Dieder
>
> On 09/06/2013 07:20 AM, Nicolae Paladi wrote:
>
> Hi Marcelo,
>
> I have the same issue (I'm on CentOS 6.4 though); have you found a
> solution?
>
> There was a similar thread earler:
> http://openstack.redhat.com/forum/discussion/230/warning-quantum-db-agentschedulers_db-fail-scheduling-network/p1
> Make sure that all agents are up in 'quantum agent-list'
>
> Also, in your quantum/server.log, do you get something like:
> WARNING [quantum.api.extensions] Extension routed-service-insertion not
> supported by any of loaded plugins
>
> I am trying to understand if this is a related problem or something that
> can be ignored atm;
>
> I can say that after some fiddling with the quantum dhcp agents my
> instances were getting an IP address and I could reach them yesterday, but
> apparently that wasn't very stable and today I'm back with the same issue.
>
> cheers,
> /Nicolae
>
>
> On 5 September 2013 03:42, happy idea <guolongcang.work at gmail.com> wrote:
>
>> Are you sure you had follow this page'guide ?
>> http://docs.openstack.org/grizzly/basic-install/apt/content/basic-install_network.html
>>
>>
>> 2013/9/5 Marcelo Dieder <marcelodieder at gmail.com>
>>
>>> Hi, yes, I have dnsmasq installed on Network Node.
>>>
>>> root at network:~# apt-get install dnsmasq
>>> Reading package lists... Done
>>> Building dependency tree
>>> Reading state information... Done
>>> dnsmasq is already the newest version.
>>> 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
>>>
>>>
>>> root at network:~# ps aux | grep -i dnsmasq
>>>
>>> dnsmasq 3807 0.0 0.1 28820 980 ? S 15:29 0:00
>>> /usr/sbin/dnsmasq -x /var/run/dnsmasq/dnsmasq.pid -u dnsmasq -r
>>> /var/run/dnsmasq/resolv.conf -7
>>> /etc/dnsmasq.d,.dpkg-dist,.dpkg-old,.dpkg-new
>>>
>>> nobody 26040 0.0 0.2 28820 1004 ? S 15:45 0:00 dnsmasq
>>> --no-hosts --no-resolv --strict-order --bind-interfaces
>>> --interface=tap91e05e25-7f --except-interface=lo
>>> --pid-file=/var/lib/quantum/dhcp/a8f7c937-e8d0-4952-bff6-7d364335df22/pid
>>> --dhcp-hostsfile=/var/lib/quantum/dhcp/a8f7c937-e8d0-4952-bff6-7d364335df22/host
>>> --dhcp-optsfile=/var/lib/quantum/dhcp/a8f7c937-e8d0-4952-bff6-7d364335df22/opts
>>> --dhcp-script=/usr/bin/quantum-dhcp-agent-dnsmasq-lease-update
>>> --leasefile-ro --dhcp-range=set:tag0,10.5.5.0,static,120s --conf-file=
>>> --domain=openstacklocal
>>>
>>> root 26041 0.0 0.0 28792 244 ? S 15:45 0:00 dnsmasq
>>> --no-hosts --no-resolv --strict-order --bind-interfaces
>>> --interface=tap91e05e25-7f --except-interface=lo
>>> --pid-file=/var/lib/quantum/dhcp/a8f7c937-e8d0-4952-bff6-7d364335df22/pid
>>> --dhcp-hostsfile=/var/lib/quantum/dhcp/a8f7c937-e8d0-4952-bff6-7d364335df22/host
>>> --dhcp-optsfile=/var/lib/quantum/dhcp/a8f7c937-e8d0-4952-bff6-7d364335df22/opts
>>> --dhcp-script=/usr/bin/quantum-dhcp-agent-dnsmasq-lease-update
>>> --leasefile-ro --dhcp-range=set:tag0,10.5.5.0,static,120s --conf-file=
>>> --domain=openstacklocal
>>>
>>> I restarted the service dnsmasq, but occured the same problem when I
>>> started a new instance.
>>>
>>> 2013-09-04 15:39:44 WARNING [quantum.db.agentschedulers_db] Fail
>>> scheduling network {'status': u'ACTIVE', 'subnets':
>>> [u'80b21701-4b05-4585-985a-60905ff42531'], 'name': u'public',
>>> 'provider:physical_network': None, 'admin_state_up': True, 'tenant_id':
>>> u'27d2b93f11ac4e91a3edb26edb28fb6b', 'provider:network_type': u'gre',
>>> 'router:external': True, 'shared': False, 'id':
>>> u'b3e465b7-b5a2-45d5-8b24-aa8bea0ab0a0', 'provider:segmentation_id': 2L}
>>>
>>> 2013-09-04 15:47:00 WARNING [quantum.db.agentschedulers_db] Fail
>>> scheduling network {'status': u'ACTIVE', 'subnets':
>>> [u'80b21701-4b05-4585-985a-60905ff42531'], 'name': u'public',
>>> 'provider:physical_network': None, 'admin_state_up': True, 'tenant_id':
>>> u'27d2b93f11ac4e91a3edb26edb28fb6b', 'provider:network_type': u'gre',
>>> 'router:external': True, 'shared': False, 'id':
>>> u'b3e465b7-b5a2-45d5-8b24-aa8bea0ab0a0', 'provider:segmentation_id': 2L}
>>>
>>> Thanks.
>>> Marcelo Dieder
>>>
>>>
>>> On 09/04/2013 11:15 AM, Hathaway.Jon wrote:
>>>
>>> Do you have dnsmasq installed? I found that it isnt installed as a
>>> dependency. Without it i never received dhcp either.
>>>
>>> Sent from my iPhone
>>>
>>> On Sep 3, 2013, at 10:31 PM, "happy idea" <guolongcang.work at gmail.com>
>>> wrote:
>>>
>>> you didn't install the dhcp agent, please refer to this guide
>>> https://github.com/mseknibilel/OpenStack-Grizzly-Install-Guide/blob/OVS_MultiNode/OpenStack_Grizzly_Install_Guide.rst
>>>
>>>
>>> 2013/9/4 Marcelo Dieder <marcelodieder at gmail.com>
>>>
>>>> Hi All!
>>>>
>>>> I have a ambient with 3 hosts (Network, Controller and Node1 (Qemu)). I
>>>> created an ambient based this tutorial (
>>>> http://docs.openstack.org/grizzly/basic-install/apt/content/basic-install_controller.html).
>>>> My problem is when I create a instance. The instance Instance doesn't get
>>>> IP address.
>>>>
>>>> checking http://169.254.169.254/20090404/instanceid
>>>> failed 1/20: up 187.68. request failed
>>>> failed 2/20: up 190.06. request failed
>>>> failed 3/20: up 192.24. request failed
>>>> failed 4/20: up 194.43. request failed
>>>> failed 5/20: up 196.61. request failed
>>>> failed 6/20: up 198.82. request failed
>>>> failed 7/20: up 201.03. request failed
>>>> failed 8/20: up 203.22. request failed
>>>> failed 9/20: up 205.42. request failed
>>>> failed 10/20: up 207.64. request failed
>>>> failed 11/20: up 209.87. request failed
>>>> failed 12/20: up 212.08. request failed
>>>> failed 13/20: up 214.29. request failed
>>>> failed 14/20: up 216.49. request failed
>>>> failed 15/20: up 218.70. request failed
>>>> failed 16/20: up 220.91. request failed
>>>> failed 17/20: up 223.13. request failed
>>>> failed 18/20: up 225.38. request failed
>>>> failed 19/20: up 227.62. request failed
>>>> failed 20/20: up 229.87. request failed
>>>> failed to read iid from metadata. tried 20
>>>> no results found for mode=net. up 232.10. searched: nocloud configdrive ec2
>>>> failed to get instanceid of datasource
>>>> Starting dropbear sshd: generating rsa key... generating dsa key... OK
>>>> === network info ===
>>>> ifinfo: lo,up,127.0.0.1,8,::1
>>>> ifinfo: eth0,up,,8,fe80::f816:3eff:fef3:2a6d
>>>> === datasource: None None ===
>>>>
>>>>
>>>> At the controller I received the warning:
>>>>
>>>>
>>>> 2013-09-04 00:40:44 WARNING [quantum.scheduler.dhcp_agent_scheduler]
>>>> No active DHCP agents
>>>> 2013-09-04 00:40:44 WARNING [quantum.db.agentschedulers_db] Fail
>>>> scheduling network {'status': u'ACTIVE', 'subnets':
>>>> [u'80b21701-4b05-4585-985a-60905ff42531'], 'name': u'public',
>>>> 'provider:physical_network': None, 'admin_state_up': True, 'tenant_id':
>>>> u'27d2b93f11ac4e91a3edb26edb28fb6b', 'provider:network_type': u'gre',
>>>> 'router:external': True, 'shared': False, 'id':
>>>> u'b3e465b7-b5a2-45d5-8b24-aa8bea0ab0a0', 'provider:segmentation_id': 2L}
>>>>
>>>> And when I executed:
>>>>
>>>>
>>>> root at cloud:~# quantum agent-list
>>>> Unknown command ['agent-list']
>>>>
>>>> Other commands:
>>>>
>>>> root at cloud:~# nova-manage service list
>>>> Binary Host Zone
>>>> Status State Updated_At
>>>> nova-cert cloud internal
>>>> enabled :-) 2013-09-04 03:59:12
>>>> nova-consoleauth cloud internal
>>>> enabled :-) 2013-09-04 03:59:12
>>>> nova-scheduler cloud internal
>>>> enabled :-) 2013-09-04 03:59:12
>>>> nova-conductor cloud internal
>>>> enabled :-) 2013-09-04 03:59:12
>>>> nova-compute c01 nova
>>>> enabled :-) 2013-09-04 03:59:04
>>>>
>>>> root at c01:~# nova list
>>>>
>>>> +--------------------------------------+---------+--------+------------------------+
>>>> | ID | Name | Status |
>>>> Networks |
>>>>
>>>> +--------------------------------------+---------+--------+------------------------+
>>>> | 2c704622-1b5f-4651-9553-51aabee9090c | test29 | ACTIVE |
>>>> public=xxx.xxx.xxx.xxx |
>>>>
>>>> I searched but I couldn't find any resolution. Anybody has this problem?
>>>>
>>>> Cheers.
>>>>
>>>> Marcelo Dieder
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> 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
>>>>
>>>>
>>> _______________________________________________
>>> 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
>>>
>>>
>>>
>>
>> _______________________________________________
>> 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
>>
>>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20130906/443bcaa4/attachment.html>
More information about the Openstack
mailing list