Re: [Openstack] Instance can't reach dhcp or metadata
Hi Mandar, The whole "default" stuff is a common point of confusion with QuantumManager, and something we're planning on getting rid of in Folsom. For Essex, I've explained the details below. On Wed, Apr 11, 2012 at 8:47 PM, Vaze, Mandar wrote: > Dan, > > Bug you referred to specifically says “default” tenant – which is not the > case for me. > > I’m not using default tenant – I’m using uuid of “demo” tenant created by > stack.sh > I suspect you are spinning up VMs using the "demo" tenant, but unless you have modified your stack.sh, when you create a network using nova-manage, no "--project_id" flag is specified. Hence, from a Quantum perspective that network does not have a tenant, making this a "global" network, which is owned by a tenant named "default" (this string can be changed using the quantum_default_tenant_id flag in nova). For more details on creating quantum networks with nova-manage, see: http://docs.openstack.org/incubation/openstack-network/admin/content/Net-Create-dle455.html > > > BTW, stack.sh creates incorrect entries in quantum/mélange DB, it has > tenant ID = “default” – This should be uuid of some valid tenant, not > string “default” > This is not actually incorrect, given the logic above, though I admit it is confusing. This was really just to provide backward compatibility with old nova-manage calls that did not specify a project. In Folsom, we're planning on requiring a tenant for all Quantum network creation. > > > ** ** > > FWIW, this setup (although seems incorrect) works with quantum+nova_ipam – > not with quantum+mélange > > i.e. With quantum+nova_ipam I can just create a VM and ssh to it. > Yes, quantum + nova-ipam is a much more tested path. I suspect you are just running into the bug I mention below and that your quantum + melange setup would work once you have that patch in, but I don't know of anyone using Melange + DHCP in a real deployment, so I wouldn't be surprised if there were more issues lurking. Dan > > > ** ** > > -Mandar > > ** ** > > *From:* Dan Wendlandt [mailto:d...@nicira.com] > *Sent:* Wednesday, April 11, 2012 11:07 PM > *To:* Vaze, Mandar > *Cc:* Nicolas de BONFILS; openstack@lists.launchpad.net > > *Subject:* Re: [Openstack] Instance can't reach dhcp or metadata > > ** ** > > ** ** > > On Wed, Apr 11, 2012 at 5:24 AM, Vaze, Mandar > wrote: > > I have *never* been able to get to ssh to an instance with Quantum+mélange > > > https://lists.launchpad.net/openstack/msg09206.html > > ** ** > > Looking at that thread, I suspect you may be hitting the same issue I > mentioned above: https://bugs.launchpad.net/quantum/+bug/949234 > > ** ** > > If you try that fix and either do or don't have luck with it, please add a > comment to the issue. > > ** ** > > Dan > > ** ** > > > > > > -Mandar > > > > *From:* openstack-bounces+mandar.vaze=nttdata@lists.launchpad.net[mailto: > openstack-bounces+mandar.vaze=nttdata@lists.launchpad.net] *On Behalf > Of *Nicolas de BONFILS > *Sent:* Wednesday, April 11, 2012 1:45 AM > *To:* Dan Wendlandt > *Cc:* openstack@lists.launchpad.net > *Subject:* Re: [Openstack] Instance can't reach dhcp or metadata > > > > Hi, > > I can't get the ssh connection,VNC doesn't respond and 10.0.0.1 go to > localhost (ssh to server itself) > > > > > --- > > Nicolas > > ** ** > > On Tue, Apr 10, 2012 at 21:42, Dan Wendlandt wrote: > > > > On Tue, Apr 10, 2012 at 11:54 AM, Nicolas de BONFILS > wrote: > > Things starting to work ! > > When I boot boot cirros image with ip 10.0.0.1, I can ssh (even if it > doesn't accept my password), but when I lauch tty-uec image (or other) with > ip starting at 10.0.0.2, ssh tell "no route to host" > > > > If it works on one image, but not another, this is probably an issue with > the image itself. The "no route to host" error is likely because the > gateway interface 10.0.0.1 is arping for 10.0.0.2 and not getting a > response. Perhaps the image is not configured to use DCHP on the > interface? Or is not correctly launching? I'd use VNC to login and check > it out. > > > > Dan > > > > > I sum up the actual conf > >- br-int as bridge in OVS, no other conf (nothing in interfaces file), >ifconfig show a 10.0.0.1 ip >- nova conf got public_interface=br-int / flat_network_bridge=br-int /
Re: [Openstack] Instance can't reach dhcp or metadata
Dan, Bug you referred to specifically says "default" tenant - which is not the case for me. I'm not using default tenant - I'm using uuid of "demo" tenant created by stack.sh BTW, stack.sh creates incorrect entries in quantum/mélange DB, it has tenant ID = "default" - This should be uuid of some valid tenant, not string "default" FWIW, this setup (although seems incorrect) works with quantum+nova_ipam - not with quantum+mélange i.e. With quantum+nova_ipam I can just create a VM and ssh to it. -Mandar From: Dan Wendlandt [mailto:d...@nicira.com] Sent: Wednesday, April 11, 2012 11:07 PM To: Vaze, Mandar Cc: Nicolas de BONFILS; openstack@lists.launchpad.net Subject: Re: [Openstack] Instance can't reach dhcp or metadata On Wed, Apr 11, 2012 at 5:24 AM, Vaze, Mandar mailto:mandar.v...@nttdata.com>> wrote: I have never been able to get to ssh to an instance with Quantum+mélange https://lists.launchpad.net/openstack/msg09206.html Looking at that thread, I suspect you may be hitting the same issue I mentioned above: https://bugs.launchpad.net/quantum/+bug/949234 If you try that fix and either do or don't have luck with it, please add a comment to the issue. Dan -Mandar From: openstack-bounces+mandar.vaze=nttdata@lists.launchpad.net<mailto:nttdata@lists.launchpad.net> [mailto:openstack-bounces+mandar.vaze<mailto:openstack-bounces%2Bmandar.vaze>=nttdata@lists.launchpad.net<mailto:nttdata@lists.launchpad.net>] On Behalf Of Nicolas de BONFILS Sent: Wednesday, April 11, 2012 1:45 AM To: Dan Wendlandt Cc: openstack@lists.launchpad.net<mailto:openstack@lists.launchpad.net> Subject: Re: [Openstack] Instance can't reach dhcp or metadata Hi, I can't get the ssh connection,VNC doesn't respond and 10.0.0.1 go to localhost (ssh to server itself) --- Nicolas On Tue, Apr 10, 2012 at 21:42, Dan Wendlandt mailto:d...@nicira.com>> wrote: On Tue, Apr 10, 2012 at 11:54 AM, Nicolas de BONFILS mailto:ndebonf...@gmail.com>> wrote: Things starting to work ! When I boot boot cirros image with ip 10.0.0.1, I can ssh (even if it doesn't accept my password), but when I lauch tty-uec image (or other) with ip starting at 10.0.0.2, ssh tell "no route to host" If it works on one image, but not another, this is probably an issue with the image itself. The "no route to host" error is likely because the gateway interface 10.0.0.1 is arping for 10.0.0.2 and not getting a response. Perhaps the image is not configured to use DCHP on the interface? Or is not correctly launching? I'd use VNC to login and check it out. Dan I sum up the actual conf * br-int as bridge in OVS, no other conf (nothing in interfaces file), ifconfig show a 10.0.0.1 ip * nova conf got public_interface=br-int / flat_network_bridge=br-int / flat_interface=eth0 * quantum manage the assignation successfully Thanks --- Nicolas ___ Mailing list: https://launchpad.net/~openstack<https://launchpad.net/%7Eopenstack> Post to : openstack@lists.launchpad.net<mailto:openstack@lists.launchpad.net> Unsubscribe : https://launchpad.net/~openstack<https://launchpad.net/%7Eopenstack> More help : https://help.launchpad.net/ListHelp -- ~~~ Dan Wendlandt Nicira, Inc: www.nicira.com<http://www.nicira.com> twitter: danwendlandt ~~~ __ Disclaimer:This email and any attachments are sent in strictest confidence for the sole use of the addressee and may contain legally privileged, confidential, and proprietary data. If you are not the intended recipient, please advise the sender by replying promptly to this email and then delete and destroy this email and any attachments without any further use, copying or forwarding -- ~~~ Dan Wendlandt Nicira, Inc: www.nicira.com<http://www.nicira.com> twitter: danwendlandt ~~~ __ Disclaimer:This email and any attachments are sent in strictest confidence for the sole use of the addressee and may contain legally privileged, confidential, and proprietary data. If you are not the intended recipient, please advise the sender by replying promptly to this email and then delete and destroy this email and any attachments without any further use, copying or forwarding___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Instance can't reach dhcp or metadata
On Wed, Apr 11, 2012 at 5:24 AM, Vaze, Mandar wrote: > I have *never* been able to get to ssh to an instance with > Quantum+mélange > > https://lists.launchpad.net/openstack/msg09206.html > Looking at that thread, I suspect you may be hitting the same issue I mentioned above: https://bugs.launchpad.net/quantum/+bug/949234 If you try that fix and either do or don't have luck with it, please add a comment to the issue. Dan > > > ** ** > > -Mandar > > ** ** > > *From:* openstack-bounces+mandar.vaze=nttdata@lists.launchpad.net[mailto: > openstack-bounces+mandar.vaze=nttdata@lists.launchpad.net] *On Behalf > Of *Nicolas de BONFILS > *Sent:* Wednesday, April 11, 2012 1:45 AM > *To:* Dan Wendlandt > *Cc:* openstack@lists.launchpad.net > *Subject:* Re: [Openstack] Instance can't reach dhcp or metadata > > ** ** > > Hi, > > I can't get the ssh connection,VNC doesn't respond and 10.0.0.1 go to > localhost (ssh to server itself) > > > ** ** > > --- > > Nicolas > > > > > > On Tue, Apr 10, 2012 at 21:42, Dan Wendlandt wrote: > > ** ** > > On Tue, Apr 10, 2012 at 11:54 AM, Nicolas de BONFILS > wrote: > > Things starting to work ! > > When I boot boot cirros image with ip 10.0.0.1, I can ssh (even if it > doesn't accept my password), but when I lauch tty-uec image (or other) with > ip starting at 10.0.0.2, ssh tell "no route to host" > > ** ** > > If it works on one image, but not another, this is probably an issue with > the image itself. The "no route to host" error is likely because the > gateway interface 10.0.0.1 is arping for 10.0.0.2 and not getting a > response. Perhaps the image is not configured to use DCHP on the > interface? Or is not correctly launching? I'd use VNC to login and check > it out. > > > > Dan > > ** ** > > > I sum up the actual conf > >- br-int as bridge in OVS, no other conf (nothing in interfaces file), >ifconfig show a 10.0.0.1 ip >- nova conf got public_interface=br-int / flat_network_bridge=br-int / >flat_interface=eth0 >- quantum manage the assignation successfully > > Thanks > > --- > > Nicolas > > ** ** > > ___ > Mailing list: https://launchpad.net/~openstack > Post to : openstack@lists.launchpad.net > Unsubscribe : https://launchpad.net/~openstack > More help : https://help.launchpad.net/ListHelp > > > > > > ** ** > > -- > > ~~~ > Dan Wendlandt > > Nicira, Inc: www.nicira.com > > twitter: danwendlandt > ~~~ > > ** ** > > ** ** > > __ > Disclaimer:This email and any attachments are sent in strictest confidence > for the sole use of the addressee and may contain legally privileged, > confidential, and proprietary data. If you are not the intended recipient, > please advise the sender by replying promptly to this email and then delete > and destroy this email and any attachments without any further use, copying > or forwarding > -- ~~~ Dan Wendlandt Nicira, Inc: www.nicira.com twitter: danwendlandt ~~~ ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Instance can't reach dhcp or metadata
On Tue, Apr 10, 2012 at 1:15 PM, Nicolas de BONFILS wrote: > Hi, > > I can't get the ssh connection,VNC doesn't respond and 10.0.0.1 go to > localhost (ssh to server itself) > are you connecting to VNC over localhost and it doesn't respond? Do you see your VM running when you run "ps"? If so, are you using the vnc IP and port shown there? Dan > > --- > Nicolas > > > > On Tue, Apr 10, 2012 at 21:42, Dan Wendlandt wrote: > >> >> >> On Tue, Apr 10, 2012 at 11:54 AM, Nicolas de BONFILS < >> ndebonf...@gmail.com> wrote: >> >>> Things starting to work ! >>> >>> When I boot boot cirros image with ip 10.0.0.1, I can ssh (even if it >>> doesn't accept my password), but when I lauch tty-uec image (or other) with >>> ip starting at 10.0.0.2, ssh tell "no route to host" >>> >> >> If it works on one image, but not another, this is probably an issue with >> the image itself. The "no route to host" error is likely because the >> gateway interface 10.0.0.1 is arping for 10.0.0.2 and not getting a >> response. Perhaps the image is not configured to use DCHP on the >> interface? Or is not correctly launching? I'd use VNC to login and check >> it out. >> >> Dan >> >> >>> I sum up the actual conf >>> >>>- br-int as bridge in OVS, no other conf (nothing in interfaces >>>file), ifconfig show a 10.0.0.1 ip >>>- nova conf got public_interface=br-int / flat_network_bridge=br-int >>>/ flat_interface=eth0 >>>- quantum manage the assignation successfully >>> >>> Thanks >>> >>> --- >>> Nicolas >>> >>> ___ >>> Mailing list: https://launchpad.net/~openstack >>> Post to : openstack@lists.launchpad.net >>> Unsubscribe : https://launchpad.net/~openstack >>> More help : https://help.launchpad.net/ListHelp >>> >>> >> >> >> -- >> ~~~ >> Dan Wendlandt >> Nicira, Inc: www.nicira.com >> twitter: danwendlandt >> ~~~ >> >> > -- ~~~ Dan Wendlandt Nicira, Inc: www.nicira.com twitter: danwendlandt ~~~ ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Instance can't reach dhcp or metadata
I have never been able to get to ssh to an instance with Quantum+mélange https://lists.launchpad.net/openstack/msg09206.html -Mandar From: openstack-bounces+mandar.vaze=nttdata@lists.launchpad.net [mailto:openstack-bounces+mandar.vaze=nttdata@lists.launchpad.net] On Behalf Of Nicolas de BONFILS Sent: Wednesday, April 11, 2012 1:45 AM To: Dan Wendlandt Cc: openstack@lists.launchpad.net Subject: Re: [Openstack] Instance can't reach dhcp or metadata Hi, I can't get the ssh connection,VNC doesn't respond and 10.0.0.1 go to localhost (ssh to server itself) --- Nicolas On Tue, Apr 10, 2012 at 21:42, Dan Wendlandt mailto:d...@nicira.com>> wrote: On Tue, Apr 10, 2012 at 11:54 AM, Nicolas de BONFILS mailto:ndebonf...@gmail.com>> wrote: Things starting to work ! When I boot boot cirros image with ip 10.0.0.1, I can ssh (even if it doesn't accept my password), but when I lauch tty-uec image (or other) with ip starting at 10.0.0.2, ssh tell "no route to host" If it works on one image, but not another, this is probably an issue with the image itself. The "no route to host" error is likely because the gateway interface 10.0.0.1 is arping for 10.0.0.2 and not getting a response. Perhaps the image is not configured to use DCHP on the interface? Or is not correctly launching? I'd use VNC to login and check it out. Dan I sum up the actual conf * br-int as bridge in OVS, no other conf (nothing in interfaces file), ifconfig show a 10.0.0.1 ip * nova conf got public_interface=br-int / flat_network_bridge=br-int / flat_interface=eth0 * quantum manage the assignation successfully Thanks --- Nicolas ___ Mailing list: https://launchpad.net/~openstack<https://launchpad.net/%7Eopenstack> Post to : openstack@lists.launchpad.net<mailto:openstack@lists.launchpad.net> Unsubscribe : https://launchpad.net/~openstack<https://launchpad.net/%7Eopenstack> More help : https://help.launchpad.net/ListHelp -- ~~~ Dan Wendlandt Nicira, Inc: www.nicira.com<http://www.nicira.com> twitter: danwendlandt ~~~ __ Disclaimer:This email and any attachments are sent in strictest confidence for the sole use of the addressee and may contain legally privileged, confidential, and proprietary data. If you are not the intended recipient, please advise the sender by replying promptly to this email and then delete and destroy this email and any attachments without any further use, copying or forwarding ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Instance can't reach dhcp or metadata
Hi, I can't get the ssh connection,VNC doesn't respond and 10.0.0.1 go to localhost (ssh to server itself) --- Nicolas On Tue, Apr 10, 2012 at 21:42, Dan Wendlandt wrote: > > > On Tue, Apr 10, 2012 at 11:54 AM, Nicolas de BONFILS > wrote: > >> Things starting to work ! >> >> When I boot boot cirros image with ip 10.0.0.1, I can ssh (even if it >> doesn't accept my password), but when I lauch tty-uec image (or other) with >> ip starting at 10.0.0.2, ssh tell "no route to host" >> > > If it works on one image, but not another, this is probably an issue with > the image itself. The "no route to host" error is likely because the > gateway interface 10.0.0.1 is arping for 10.0.0.2 and not getting a > response. Perhaps the image is not configured to use DCHP on the > interface? Or is not correctly launching? I'd use VNC to login and check > it out. > > Dan > > >> I sum up the actual conf >> >>- br-int as bridge in OVS, no other conf (nothing in interfaces >>file), ifconfig show a 10.0.0.1 ip >>- nova conf got public_interface=br-int / flat_network_bridge=br-int >>/ flat_interface=eth0 >>- quantum manage the assignation successfully >> >> Thanks >> >> --- >> Nicolas >> >> ___ >> Mailing list: https://launchpad.net/~openstack >> Post to : openstack@lists.launchpad.net >> Unsubscribe : https://launchpad.net/~openstack >> More help : https://help.launchpad.net/ListHelp >> >> > > > -- > ~~~ > Dan Wendlandt > Nicira, Inc: www.nicira.com > twitter: danwendlandt > ~~~ > > ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Instance can't reach dhcp or metadata
On Tue, Apr 10, 2012 at 11:54 AM, Nicolas de BONFILS wrote: > Things starting to work ! > > When I boot boot cirros image with ip 10.0.0.1, I can ssh (even if it > doesn't accept my password), but when I lauch tty-uec image (or other) with > ip starting at 10.0.0.2, ssh tell "no route to host" > If it works on one image, but not another, this is probably an issue with the image itself. The "no route to host" error is likely because the gateway interface 10.0.0.1 is arping for 10.0.0.2 and not getting a response. Perhaps the image is not configured to use DCHP on the interface? Or is not correctly launching? I'd use VNC to login and check it out. Dan > I sum up the actual conf > >- br-int as bridge in OVS, no other conf (nothing in interfaces file), >ifconfig show a 10.0.0.1 ip >- nova conf got public_interface=br-int / flat_network_bridge=br-int / >flat_interface=eth0 >- quantum manage the assignation successfully > > Thanks > > --- > Nicolas > > ___ > Mailing list: https://launchpad.net/~openstack > Post to : openstack@lists.launchpad.net > Unsubscribe : https://launchpad.net/~openstack > More help : https://help.launchpad.net/ListHelp > > -- ~~~ Dan Wendlandt Nicira, Inc: www.nicira.com twitter: danwendlandt ~~~ ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Instance can't reach dhcp or metadata
Things starting to work ! When I boot boot cirros image with ip 10.0.0.1, I can ssh (even if it doesn't accept my password), but when I lauch tty-uec image (or other) with ip starting at 10.0.0.2, ssh tell "no route to host" I sum up the actual conf - br-int as bridge in OVS, no other conf (nothing in interfaces file), ifconfig show a 10.0.0.1 ip - nova conf got public_interface=br-int / flat_network_bridge=br-int / flat_interface=eth0 - quantum manage the assignation successfully Thanks --- Nicolas ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Instance can't reach dhcp or metadata
Hi,
You have misconfigured ip addressing. 10.0.0.1 shoud be assigned to
ovs device br-int. Remove br100 from configuration and let
nova-network do all the stuff.
On Tue, Apr 10, 2012 at 3:47 PM, Nicolas de BONFILS
wrote:
> Hi all,
>
> Output of list bridge (I reuse br-int as it was initially)
>>
>> ➜ ~nicolas ovs-vsctl list bridge br-int
>> _uuid : fccace0c-5518-4a8c-85d1-7d8638d57e6a
>> controller : []
>> datapath_id : "0ececafc8c4a"
>> datapath_type : ""
>> external_ids : {bridge-id=br-int}
>> fail_mode : []
>> flood_vlans : []
>> mirrors : []
>> name : br-int
>> netflow : []
>> other_config : {}
>> ports : [3315a03a-3ed0-48ef-b16c-5620ec468096,
>> f4499821-6061-4844-8297-9534b8be3128]
>> sflow : []
>> status : {}
>> stp_enable : false
>
>
> After advices on some modification, I launch a new instance and found some
> interesting log. It seems that the instance is not linked to eth0 or br100
> or br-int but tap661cb0d2-03
> See here :
>>
>> 2012-04-10 15:37:56 DEBUG nova.compute.manager
>> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
>> 1e00e717a3a3487381dfe5f8e7f63c92] [instance:
>> d0c21b65-2957-40cc-a3db-887cf7c959ee] Instance network_info:
>> |[VIF({'network': Network({'bridge': u'', 'subnets': [Subnet({'ips':
>> [FixedIP({'meta': {}, 'version': 4, 'type': u'fixed', 'floating_ips': [],
>> 'address': u'10.0.0.14'})], 'version': 4, 'meta': {u'dhcp_server': None},
>> 'dns': [IP({'meta': {}, 'version': 4, 'type': u'dns', 'address':
>> u'8.8.4.4'})], 'routes': [], 'cidr': u'10.0.0.0/24', 'gateway': IP({'meta':
>> {}, 'version': None, 'type': u'gateway', 'address': None})})], 'meta':
>> {u'tenant_id': u'default'}, 'id': u'1ff5adb5-2a37-44d4-84c0-d9b6ee50fc42',
>> 'label': u'private'}), 'meta': {}, 'id':
>> u'661cb0d2-03f6-4129--30f5be65cada', 'address': u'FE:EE:DD:00:00:0E'})]|
>> from (pid=28534) _allocate_network
>> /openstack/nova/nova/compute/manager.py:566
>> 2012-04-10 15:37:56 DEBUG nova.virt.libvirt.connection
>> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
>> 1e00e717a3a3487381dfe5f8e7f63c92] [instance:
>> d0c21b65-2957-40cc-a3db-887cf7c959ee] Starting toXML method from (pid=28534)
>> to_xml /openstack/nova/nova/virt/libvirt/connection.py:1640
>> 2012-04-10 15:37:56 DEBUG nova.virt.libvirt.connection
>> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
>> 1e00e717a3a3487381dfe5f8e7f63c92] block_device_list [] from (pid=28534)
>> _volume_in_mapping /openstack/nova/nova/virt/libvirt/connection.py:1363
>> 2012-04-10 15:37:56 DEBUG nova.virt.libvirt.connection
>> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
>> 1e00e717a3a3487381dfe5f8e7f63c92] Path '/openstack/nova/instances'
>> supports direct I/O from (pid=28534) _supports_direct_io
>> /openstack/nova/nova/virt/libvirt/connection.py:1003
>> 2012-04-10 15:37:56 DEBUG nova.virt.libvirt.connection
>> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
>> 1e00e717a3a3487381dfe5f8e7f63c92] block_device_list [] from (pid=28534)
>> _volume_in_mapping /openstack/nova/nova/virt/libvirt/connection.py:1363
>> 2012-04-10 15:37:56 DEBUG nova.utils
>> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
>> 1e00e717a3a3487381dfe5f8e7f63c92] Running cmd (subprocess): ip link show dev
>> tap661cb0d2-03 from (pid=28534) execute
>> /openstack/nova/nova/utils.py:220
>> 2012-04-10 15:37:56 DEBUG nova.utils
>> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
>> 1e00e717a3a3487381dfe5f8e7f63c92] Result was 1 from (pid=28534) execute
>> /openstack/nova/nova/utils.py:236
>> 2012-04-10 15:37:56 DEBUG nova.utils
>> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
>> 1e00e717a3a3487381dfe5f8e7f63c92] Running cmd (subprocess): sudo
>> /usr/local/bin/nova-rootwrap ip tuntap add tap661cb0d2-03 mode tap from
>> (pid=28534) execute /openstack/nova/nova/utils.py:220
>> 2012-04-10 15:37:56 DEBUG nova.utils
>> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
>> 1e00e717a3a3487381dfe5f8e7f63c92] Running cmd (subprocess): sudo
>> /usr/local/bin/nova-rootwrap ip link set tap661cb0d2-03 up from (pid=28534)
>> execute /openstack/nova/nova/utils.py:220
>> 2012-04-10 15:37:56 DEBUG nova.compute.manager [-] Updated the info_cache
>> for instance d0c21b65-2957-40cc-a3db-887cf7c959ee from (pid=28534)
>> _heal_instance_info_cache /openstack/nova/nova/compute/manager.py:2227
>> 2012-04-10 15:37:56 DEBUG nova.manager [-] Skipping
>> ComputeManager._run_image_cache_manager_pass, 30 ticks left until next run
>> from (pid=28534) periodic_tasks /openstack/nova/nova/manager.py:147
>> 2012-04-10 15:37:56 DEBUG nova.manager [-] Running periodic task
>> Com
Re: [Openstack] Instance can't reach dhcp or metadata
Hi all,
Output of list bridge (I reuse br-int as it was initially)
> ➜ ~nicolas ovs-vsctl list bridge br-int
> _uuid : fccace0c-5518-4a8c-85d1-7d8638d57e6a
> controller : []
> datapath_id : "0ececafc8c4a"
> datapath_type : ""
> external_ids: {bridge-id=br-int}
> fail_mode : []
> flood_vlans : []
> mirrors : []
> name: br-int
> netflow : []
> other_config: {}
> ports : [3315a03a-3ed0-48ef-b16c-5620ec468096,
> f4499821-6061-4844-8297-9534b8be3128]
> sflow : []
> status : {}
> stp_enable : false
>
After advices on some modification, I launch a new instance and found some
interesting log. It seems that the instance is not linked to *eth0* or *
br100* or *br-int* but *tap661cb0d2-03*
See here :
> 2012-04-10 15:37:56 DEBUG nova.compute.manager
> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
> 1e00e717a3a3487381dfe5f8e7f63c92] [instance:
> d0c21b65-2957-40cc-a3db-887cf7c959ee] Instance network_info:
> |[VIF({'network': Network({'bridge': u'', 'subnets': [Subnet({'ips':
> [FixedIP({'meta': {}, 'version': 4, 'type': u'fixed', 'floating_ips': [],
> 'address': u'10.0.0.14'})], 'version': 4, 'meta': {u'dhcp_server': None},
> 'dns': [IP({'meta': {}, 'version': 4, 'type': u'dns', 'address':
> u'8.8.4.4'})], 'routes': [], 'cidr': u'10.0.0.0/24', 'gateway':
> IP({'meta': {}, 'version': None, 'type': u'gateway', 'address': None})})],
> 'meta': {u'tenant_id': u'default'}, 'id':
> u'1ff5adb5-2a37-44d4-84c0-d9b6ee50fc42', 'label': u'private'}), 'meta': {},
> 'id': u'661cb0d2-03f6-4129--30f5be65cada', 'address':
> u'FE:EE:DD:00:00:0E'})]| from (pid=28534) _allocate_network
> /openstack/nova/nova/compute/manager.py:566
> 2012-04-10 15:37:56 DEBUG nova.virt.libvirt.connection
> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
> 1e00e717a3a3487381dfe5f8e7f63c92] [instance:
> d0c21b65-2957-40cc-a3db-887cf7c959ee] Starting toXML method from
> (pid=28534) to_xml /openstack/nova/nova/virt/libvirt/connection.py:1640
> 2012-04-10 15:37:56 DEBUG nova.virt.libvirt.connection
> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
> 1e00e717a3a3487381dfe5f8e7f63c92] block_device_list [] from (pid=28534)
> _volume_in_mapping /openstack/nova/nova/virt/libvirt/connection.py:1363
> 2012-04-10 15:37:56 DEBUG nova.virt.libvirt.connection
> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
> 1e00e717a3a3487381dfe5f8e7f63c92] Path '/openstack/nova/instances'
> supports direct I/O from (pid=28534) _supports_direct_io
> /openstack/nova/nova/virt/libvirt/connection.py:1003
> 2012-04-10 15:37:56 DEBUG nova.virt.libvirt.connection
> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
> 1e00e717a3a3487381dfe5f8e7f63c92] block_device_list [] from (pid=28534)
> _volume_in_mapping /openstack/nova/nova/virt/libvirt/connection.py:1363
> 2012-04-10 15:37:56 DEBUG nova.utils
> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
> 1e00e717a3a3487381dfe5f8e7f63c92] Running cmd (subprocess): ip link show
> dev tap661cb0d2-03 from (pid=28534) execute
> /openstack/nova/nova/utils.py:220
> 2012-04-10 15:37:56 DEBUG nova.utils
> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
> 1e00e717a3a3487381dfe5f8e7f63c92] Result was 1 from (pid=28534) execute
> /openstack/nova/nova/utils.py:236
> 2012-04-10 15:37:56 DEBUG nova.utils
> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
> 1e00e717a3a3487381dfe5f8e7f63c92] Running cmd (subprocess): sudo
> /usr/local/bin/nova-rootwrap ip tuntap add tap661cb0d2-03 mode tap from
> (pid=28534) execute /openstack/nova/nova/utils.py:220
> 2012-04-10 15:37:56 DEBUG nova.utils
> [req-bc5aac98-6c23-4bbe-8ec3-d8bc671ea4f1 94157c6bf9ad4753ab8e1f8c20617cbb
> 1e00e717a3a3487381dfe5f8e7f63c92] Running cmd (subprocess): sudo
> /usr/local/bin/nova-rootwrap ip link set tap661cb0d2-03 up from (pid=28534)
> execute /openstack/nova/nova/utils.py:220
> 2012-04-10 15:37:56 DEBUG nova.compute.manager [-] Updated the info_cache
> for instance d0c21b65-2957-40cc-a3db-887cf7c959ee from (pid=28534)
> _heal_instance_info_cache /openstack/nova/nova/compute/manager.py:2227
> 2012-04-10 15:37:56 DEBUG nova.manager [-] Skipping
> ComputeManager._run_image_cache_manager_pass, 30 ticks left until next run
> from (pid=28534) periodic_tasks /openstack/nova/nova/manager.py:147
> 2012-04-10 15:37:56 DEBUG nova.manager [-] Running periodic task
> ComputeManager._reclaim_queued_deletes from (pid=28534) periodic_tasks
> /openstack/nova/nova/manager.py:152
> 2012-04-10 15:37:56 DEBUG nova.compute.manager [-]
> FLAGS.reclaim_instance_interval <= 0, skipping... from (pid=28534)
> _reclaim_queued_deletes /openstack/nova/nova/compute/manager.py:2380
> 20
Re: [Openstack] Instance can't reach dhcp or metadata
Can you show us more configuration from openvswitch (ovs-vsctl list bridge br100) ? ovs-vsctl list bridge br100 On Tue, Apr 10, 2012 at 12:44 PM, Nicolas de BONFILS wrote: > Hi, > > I try it, and it doesn't change anything :( > Quantum still assign an ip to the VM, but the VM still doesn't ping the dhcp > server when it boot. I believe it's a bridge/interface problem. > > --- > Nicolas > > > > On Mon, Apr 9, 2012 at 18:41, Dan Wendlandt wrote: >> >> >> >> On Mon, Apr 9, 2012 at 4:57 AM, Nicolas de BONFILS >> wrote: >>> >>> Hi, >>> >>> Thanks for the advice but I instead of using FlatManager, I use >>> QuantumManager (network_manager=nova.network.quantum.manager.QuantumManager) >>> Also, to be more accurate, Quantum assign an ip to my VM (I see it in >>> horizon and in nova/quantum mysql db), but the VM doe not get it. >> >> >> Hi Nicolas, >> >> You may be hitting a known issue with QuantumManager + DHCP + Melange: >> - https://bugs.launchpad.net/quantum/+bug/949234 >> >> Looks like a partial fix was suggested for QuantumManager in Nova, but >> validation ran into a possible melange issue, that ultimately was rejected, >> without submitting the QuantumManager fix: >> - https://bugs.launchpad.net/melange/+bug/951499 >> >> Can you try just applying the one-liner patch described in 949234 and see >> if it solves your problem? If so, we can get that merged in. >> >> Dan >> >> >>> >>> >>> Regards >>> >>> --- >>> Nicolas >>> >>> >>> >>> On Mon, Apr 9, 2012 at 13:04, wrote: Hi Nicolas, You need another interface ( say eth1 UP & running ) . Configure br100 to use eth1 . typically your configuration will look like the below .. /etc/network/interfaces # This file describes the network interfaces available on your system # and how to activate them. For more information, see interfaces(5). # The loopback network interface auto lo iface lo inet loopback # The primary network interface allow-hotplug eth0 iface eth0 inet dhcp auto eth1 auto br100 iface br100 inet static bridge_ports eth1 address 10.0.0.1 netmask 255.255.255.0 bridge_stp off bridge_fd 0 nova.conf will have the following lines too.. --fixed_range=10.0.0.0/24 --network_size=256 --network_manager=nova.network.manager.FlatManager --flat_interface=eth1 --flat_injected=False --public_interface=eth0 Meena Raja Consultant __ WIPRO TECHNOLOGIES No 53/1 Ganapa Towers ,Near Madivala Police Station , Hosur Main Road ,Bangalore-560068 Hand Phone : +91-9880549725 | Desk : +91-80-39912554 |Fax No: +91-80-25502160 Email : raja.me...@wipro.com | Website : www.wipro.com >>> >>> >>> ___ >>> Mailing list: https://launchpad.net/~openstack >>> Post to : openstack@lists.launchpad.net >>> Unsubscribe : https://launchpad.net/~openstack >>> More help : https://help.launchpad.net/ListHelp >>> >> >> >> >> -- >> ~~~ >> Dan Wendlandt >> Nicira, Inc: www.nicira.com >> twitter: danwendlandt >> ~~~ >> > > > ___ > Mailing list: https://launchpad.net/~openstack > Post to : openstack@lists.launchpad.net > Unsubscribe : https://launchpad.net/~openstack > More help : https://help.launchpad.net/ListHelp > -- Tomasz Paszkowski SS7, Asterisk, SAN, Datacenter, Cloud Computing +48500166299 ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Instance can't reach dhcp or metadata
Hi, I try it, and it doesn't change anything :( Quantum still assign an ip to the VM, but the VM still doesn't ping the dhcp server when it boot. I believe it's a bridge/interface problem. --- Nicolas On Mon, Apr 9, 2012 at 18:41, Dan Wendlandt wrote: > > > On Mon, Apr 9, 2012 at 4:57 AM, Nicolas de BONFILS > wrote: > >> Hi, >> >> Thanks for the advice but I instead of using FlatManager, I use >> QuantumManager * >> (network_manager=nova.network.quantum.manager.QuantumManager) >> * >> Also, to be more accurate, Quantum assign an ip to my VM (I see it in >> horizon and in nova/quantum mysql db), but the VM doe not get it. >> > > Hi Nicolas, > > You may be hitting a known issue with QuantumManager + DHCP + Melange: > - https://bugs.launchpad.net/quantum/+bug/949234 > > Looks like a partial fix was suggested for QuantumManager in Nova, but > validation ran into a possible melange issue, that ultimately was rejected, > without submitting the QuantumManager fix: > - https://bugs.launchpad.net/melange/+bug/951499 > > Can you try just applying the one-liner patch described in 949234 and see > if it solves your problem? If so, we can get that merged in. > > Dan > > > >> >> Regards >> >> --- >> Nicolas >> >> >> >> On Mon, Apr 9, 2012 at 13:04, wrote: >> >>> >>> Hi Nicolas, >>> >>> You need another interface ( say eth1 UP & running ) . Configure >>> br100 to use eth1 . >>> >>> >>> >>> typically your configuration will look like the below .. >>> >>> >>> >>> */etc/network/interfaces* >>> # This file describes the network interfaces available on your system >>> # and how to activate them. For more information, see interfaces(5). >>> >>> # The loopback network interface >>> auto lo >>> iface lo inet loopback >>> >>> # The primary network interface >>> allow-hotplug eth0 >>> iface eth0 inet dhcp >>> >>> auto eth1 >>> >>> auto br100 >>> iface br100 inet static >>> bridge_ports eth1 >>> address 10.0.0.1 >>> netmask 255.255.255.0 >>> bridge_stp off >>> bridge_fd 0 >>> >>> >>> >>> nova.conf will have the following lines too.. >>> >>> --fixed_range=10.0.0.0/24 >>> --network_size=256 >>> --network_manager=nova.network.manager.FlatManager >>> --flat_interface=eth1 >>> --flat_injected=False >>> --public_interface=eth0 >>> >>> >>> Meena Raja >>> Consultant >>> __ >>> *WIPRO TECHNOLOGIES >>> *No 53/1 Ganapa Towers ,Near Madivala Police Station , Hosur Main Road >>> ,Bangalore-560068 >>> Hand Phone : +91-9880549725 | Desk : +91-80-39912554 |Fax No: >>> +91-80-25502160 >>> Email : raja.me...@wipro.com | Website : www.wipro.com >>> >> >> ___ >> Mailing list: https://launchpad.net/~openstack >> Post to : openstack@lists.launchpad.net >> Unsubscribe : https://launchpad.net/~openstack >> More help : https://help.launchpad.net/ListHelp >> >> > > > -- > ~~~ > Dan Wendlandt > Nicira, Inc: www.nicira.com > twitter: danwendlandt > ~~~ > > ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Instance can't reach dhcp or metadata
On Mon, Apr 9, 2012 at 4:57 AM, Nicolas de BONFILS wrote: > Hi, > > Thanks for the advice but I instead of using FlatManager, I use > QuantumManager * > (network_manager=nova.network.quantum.manager.QuantumManager) > * > Also, to be more accurate, Quantum assign an ip to my VM (I see it in > horizon and in nova/quantum mysql db), but the VM doe not get it. > Hi Nicolas, You may be hitting a known issue with QuantumManager + DHCP + Melange: - https://bugs.launchpad.net/quantum/+bug/949234 Looks like a partial fix was suggested for QuantumManager in Nova, but validation ran into a possible melange issue, that ultimately was rejected, without submitting the QuantumManager fix: - https://bugs.launchpad.net/melange/+bug/951499 Can you try just applying the one-liner patch described in 949234 and see if it solves your problem? If so, we can get that merged in. Dan > > Regards > > --- > Nicolas > > > > On Mon, Apr 9, 2012 at 13:04, wrote: > >> >> Hi Nicolas, >> >> You need another interface ( say eth1 UP & running ) . Configure >> br100 to use eth1 . >> >> >> >> typically your configuration will look like the below .. >> >> >> >> */etc/network/interfaces* >> # This file describes the network interfaces available on your system >> # and how to activate them. For more information, see interfaces(5). >> >> # The loopback network interface >> auto lo >> iface lo inet loopback >> >> # The primary network interface >> allow-hotplug eth0 >> iface eth0 inet dhcp >> >> auto eth1 >> >> auto br100 >> iface br100 inet static >> bridge_ports eth1 >> address 10.0.0.1 >> netmask 255.255.255.0 >> bridge_stp off >> bridge_fd 0 >> >> >> >> nova.conf will have the following lines too.. >> >> --fixed_range=10.0.0.0/24 >> --network_size=256 >> --network_manager=nova.network.manager.FlatManager >> --flat_interface=eth1 >> --flat_injected=False >> --public_interface=eth0 >> >> >> Meena Raja >> Consultant >> __ >> *WIPRO TECHNOLOGIES >> *No 53/1 Ganapa Towers ,Near Madivala Police Station , Hosur Main Road >> ,Bangalore-560068 >> Hand Phone : +91-9880549725 | Desk : +91-80-39912554 |Fax No: >> +91-80-25502160 >> Email : raja.me...@wipro.com | Website : www.wipro.com >> > > ___ > Mailing list: https://launchpad.net/~openstack > Post to : openstack@lists.launchpad.net > Unsubscribe : https://launchpad.net/~openstack > More help : https://help.launchpad.net/ListHelp > > -- ~~~ Dan Wendlandt Nicira, Inc: www.nicira.com twitter: danwendlandt ~~~ ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Instance can't reach dhcp or metadata
Hi, Thanks for the advice but I instead of using FlatManager, I use QuantumManager * (network_manager=nova.network.quantum.manager.QuantumManager) * Also, to be more accurate, Quantum assign an ip to my VM (I see it in horizon and in nova/quantum mysql db), but the VM doe not get it. Regards --- Nicolas On Mon, Apr 9, 2012 at 13:04, wrote: > > Hi Nicolas, > > You need another interface ( say eth1 UP & running ) . Configure > br100 to use eth1 . > > > > typically your configuration will look like the below .. > > > > */etc/network/interfaces* > # This file describes the network interfaces available on your system > # and how to activate them. For more information, see interfaces(5). > > # The loopback network interface > auto lo > iface lo inet loopback > > # The primary network interface > allow-hotplug eth0 > iface eth0 inet dhcp > > auto eth1 > > auto br100 > iface br100 inet static > bridge_ports eth1 > address 10.0.0.1 > netmask 255.255.255.0 > bridge_stp off > bridge_fd 0 > > > > nova.conf will have the following lines too.. > > --fixed_range=10.0.0.0/24 > --network_size=256 > --network_manager=nova.network.manager.FlatManager > --flat_interface=eth1 > --flat_injected=False > --public_interface=eth0 > > > Meena Raja > Consultant > __ > *WIPRO TECHNOLOGIES > *No 53/1 Ganapa Towers ,Near Madivala Police Station , Hosur Main Road > ,Bangalore-560068 > Hand Phone : +91-9880549725 | Desk : +91-80-39912554 |Fax No: > +91-80-25502160 > Email : raja.me...@wipro.com | Website : www.wipro.com > ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
Re: [Openstack] Instance can't reach dhcp or metadata
Hi Nicolas, You need another interface ( say eth1 UP & running ) . Configure br100 to use eth1 . typically your configuration will look like the below .. /etc/network/interfaces # This file describes the network interfaces available on your system # and how to activate them. For more information, see interfaces(5). # The loopback network interface auto lo iface lo inet loopback # The primary network interface allow-hotplug eth0 iface eth0 inet dhcp auto eth1 auto br100 iface br100 inet static bridge_ports eth1 address 10.0.0.1 netmask 255.255.255.0 bridge_stp off bridge_fd 0 nova.conf will have the following lines too.. --fixed_range=10.0.0.0/24 --network_size=256 --network_manager=nova.network.manager.FlatManager --flat_interface=eth1 --flat_injected=False --public_interface=eth0 Meena Raja Consultant __ WIPRO TECHNOLOGIES No 53/1 Ganapa Towers ,Near Madivala Police Station , Hosur Main Road ,Bangalore-560068 Hand Phone : +91-9880549725 | Desk : +91-80-39912554 |Fax No: +91-80-25502160 Email : raja.me...@wipro.com | Website : www.wipro.com From: openstack-bounces+raja.meena=wipro@lists.launchpad.net [openstack-bounces+raja.meena=wipro@lists.launchpad.net] on behalf of Nicolas de BONFILS [ndebonf...@gmail.com] Sent: Monday, April 09, 2012 3:57 PM To: openstack@lists.launchpad.net Subject: [Openstack] Instance can't reach dhcp or metadata Hi stackers, I got, using devstack, openstack running. I can launch instance correctly (as seen in horizon state) but the instance can't reach the dhcp or the metadata. Here the topology (as it should be, but I maybe need to configured otherwise in the conf) of my server (one box only) : * router : 192.168.23.1 * server * ip : 192.168.23.5 (with dhcp but always the same) * interface with external (network and internet) : eth0 * interface for openvswitch bridge : br100 * interface for nova/quantum/melange bridge : br100 * openstack * glance with file backend * quantum and melange for networking * openvswitch as plugin You can find all the config files and log here : https://gist.github.com/2342747. And networking info below. Any tips, or pointing misconfiguration will be very appreciated. /etc/network/interfaces # This file describes the network interfaces available on your system # and how to activate them. For more information, see interfaces(5). # The loopback network interface auto lo iface lo inet loopback # The primary network interface allow-hotplug eth0 iface eth0 inet dhcp auto br100 iface br100 inet static address 10.0.0.1 netmask 255.255.255.0 bridge_stp off bridge_fd 0 ip addr show 1: lo: mtu 16436 qdisc noqueue state UNKNOWN link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8<http://127.0.0.1/8> scope host lo inet 169.254.169.254/32<http://169.254.169.254/32> scope link lo inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: ipddp0: mtu 585 qdisc noop state DOWN qlen 1000 link/ip/ddp 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff 3: eth0: mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether f4:XX:04:10:XX:XX brd ff:ff:ff:ff:ff:ff inet 192.168.23.5/24<http://192.168.23.5/24> brd 192.168.23.255 scope global eth0 inet6 XX:e35:XX:dce0:XX:4ff:fe10:675b/64 scope global dynamic valid_lft 86091sec preferred_lft 86091sec inet6 XX::f66d:XX:fe10:XX/64 scope link valid_lft forever preferred_lft forever 18: tap9cec6d69-f0: mtu 1500 qdisc pfifo_fast state DOWN qlen 500 link/ether 12:23:4b:e3:3e:e7 brd ff:ff:ff:ff:ff:ff 19: tapba4b75d8-10: mtu 1500 qdisc pfifo_fast state DOWN qlen 500 link/ether ce:40:f0:7c:48:b1 brd ff:ff:ff:ff:ff:ff 20: tap828bc57f-57: mtu 1500 qdisc pfifo_fast state DOWN qlen 500 link/ether 52:69:4a:f4:45:d7 brd ff:ff:ff:ff:ff:ff 23: br100: mtu 1500 qdisc noqueue state UNKNOWN link/ether f2:a9:d2:11:8b:49 brd ff:ff:ff:ff:ff:ff inet 10.0.0.1/24<http://10.0.0.1/24> brd 10.0.0.255 scope global br100 inet6 fe80::f0a9:d2ff:fe11:8b49/64 scope link valid_lft forever preferred_lft forever 24: tapcf8c4c18-f3: mtu 1500 qdisc pfifo_fast state DOWN qlen 500 link/ether d2:9f:4e:cf:ce:fa brd ff:ff:ff:ff:ff:ff 25: tap60c76595-d6: mtu 1500 qdisc pfifo_fast state DOWN qlen 500 link/ether 76:90:ce:a4:25:6c brd ff:ff:ff:ff:ff:ff 26: tap34019134-5f: mtu 1500 qdisc pfifo_fast state DOWN qlen 500 link/ether 2e:4f:bc:16:07:71 brd ff:ff:ff:ff:ff:ff 31: tap90748d8d-5d: mtu 1500 qdisc pfifo_fast state UP qlen 500 link/ether 26:45:91:52:12:bd brd ff:ff:ff:ff:ff:ff inet6 fe80::2445:91ff:fe52:12bd/64 scope link valid_lft forever preferred_lft forever 32: tapf937eb70-ff: mt
[Openstack] Instance can't reach dhcp or metadata
Hi stackers, I got, using devstack, openstack running. I can launch instance correctly (as seen in horizon state) but the instance can't reach the dhcp or the metadata. Here the topology (as it *should* be, but I maybe need to configured otherwise in the conf) of my server (one box only) : - router : 192.168.23.1 - server - ip : 192.168.23.5 (with dhcp but always the same) - interface with external (network and internet) : eth0 - interface for openvswitch bridge : br100 - interface for nova/quantum/melange bridge : br100 - openstack - glance with file backend - quantum and melange for networking - openvswitch as plugin You can find all the config files and log here : https://gist.github.com/2342747. And networking info below. Any tips, or pointing misconfiguration will be very appreciated. > */etc/network/interfaces* > > # This file describes the network interfaces available on your system > # and how to activate them. For more information, see interfaces(5). > > # The loopback network interface > auto lo > iface lo inet loopback > > # The primary network interface > allow-hotplug eth0 > iface eth0 inet dhcp > > auto br100 > iface br100 inet static > address 10.0.0.1 > netmask 255.255.255.0 > bridge_stp off > bridge_fd 0 > *ip addr show* > > 1: lo: mtu 16436 qdisc noqueue state UNKNOWN > link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 > inet 127.0.0.1/8 scope host lo > inet 169.254.169.254/32 scope link lo > inet6 ::1/128 scope host >valid_lft forever preferred_lft forever > 2: ipddp0: mtu 585 qdisc noop state DOWN qlen > 1000 > link/ip/ddp 00:00:00:00:00:00 brd ff:ff:ff:ff:ff:ff > 3: eth0: mtu 1500 qdisc pfifo_fast state > UP qlen 1000 > link/ether f4:XX:04:10:XX:XX brd ff:ff:ff:ff:ff:ff > inet 192.168.23.5/24 brd 192.168.23.255 scope global eth0 > inet6 XX:e35:XX:dce0:XX:4ff:fe10:675b/64 scope global dynamic >valid_lft 86091sec preferred_lft 86091sec > inet6 XX::f66d:XX:fe10:XX/64 scope link >valid_lft forever preferred_lft forever > 18: tap9cec6d69-f0: mtu 1500 qdisc > pfifo_fast state DOWN qlen 500 > link/ether 12:23:4b:e3:3e:e7 brd ff:ff:ff:ff:ff:ff > 19: tapba4b75d8-10: mtu 1500 qdisc > pfifo_fast state DOWN qlen 500 > link/ether ce:40:f0:7c:48:b1 brd ff:ff:ff:ff:ff:ff > 20: tap828bc57f-57: mtu 1500 qdisc > pfifo_fast state DOWN qlen 500 > link/ether 52:69:4a:f4:45:d7 brd ff:ff:ff:ff:ff:ff > 23: br100: mtu 1500 qdisc > noqueue state UNKNOWN > link/ether f2:a9:d2:11:8b:49 brd ff:ff:ff:ff:ff:ff > inet 10.0.0.1/24 brd 10.0.0.255 scope global br100 > inet6 fe80::f0a9:d2ff:fe11:8b49/64 scope link >valid_lft forever preferred_lft forever > 24: tapcf8c4c18-f3: mtu 1500 qdisc > pfifo_fast state DOWN qlen 500 > link/ether d2:9f:4e:cf:ce:fa brd ff:ff:ff:ff:ff:ff > 25: tap60c76595-d6: mtu 1500 qdisc > pfifo_fast state DOWN qlen 500 > link/ether 76:90:ce:a4:25:6c brd ff:ff:ff:ff:ff:ff > 26: tap34019134-5f: mtu 1500 qdisc > pfifo_fast state DOWN qlen 500 > link/ether 2e:4f:bc:16:07:71 brd ff:ff:ff:ff:ff:ff > 31: tap90748d8d-5d: mtu 1500 qdisc > pfifo_fast state UP qlen 500 > link/ether 26:45:91:52:12:bd brd ff:ff:ff:ff:ff:ff > inet6 fe80::2445:91ff:fe52:12bd/64 scope link >valid_lft forever preferred_lft forever > 32: tapf937eb70-ff: mtu 1500 qdisc > pfifo_fast state UP qlen 500 > link/ether 42:4e:31:c3:77:ba brd ff:ff:ff:ff:ff:ff > inet6 fe80::404e:31ff:fec3:77ba/64 scope link >valid_lft forever preferred_lft forever > 33: tap010cac48-f4: mtu 1500 qdisc > pfifo_fast state UP qlen 500 > link/ether b6:3e:1c:ee:ca:45 brd ff:ff:ff:ff:ff:ff > inet6 fe80::b43e:1cff:feee:ca45/64 scope link >valid_lft forever preferred_lft forever > *nova-manage service list* > > Binary Host Zone > Status State Updated_At > nova-compute 192.168.23.5nova > enabled:-) 2012-04-09 11:12:33 > nova-network 192.168.23.5nova > enabled:-) 2012-04-09 11:12:32 > nova-volume 192.168.23.5nova > enabled:-) 2012-04-09 11:12:32 > nova-scheduler 192.168.23.5nova > enabled:-) 2012-04-09 11:12:32 > nova-cert192.168.23.5 nova > enabled:-) 2012-04-09 11:12:32 > nova-consoleauth 192.168.23.5nova > enabled:-) 2012-04-09 11:12:31 > *brctl show* > > bridge name bridge id STP enabled interfaces > Thanks, --- Nicolas ___ Mailing list: https://launchpad.net/~openstack Post to : openstack@lists.launchpad.net Unsubscribe : https://launchpad.net/~openstack More help : https://help.launchpad.net/ListHelp
