-----Original Message-----
From: openstack-requ...@lists.openstack.org
[mailto:openstack-requ...@lists.openstack.org]
Sent: Tuesday, January 28, 2014 12:00 PM
To: openstack@lists.openstack.org
Subject: Openstack Digest, Vol 7, Issue 32
Send Openstack mailing list submissions to
openstack@lists.openstack.org
To subscribe or unsubscribe via the World Wide Web, visit
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
or, via email, send a message with subject or body 'help' to
openstack-requ...@lists.openstack.org
You can reach the person managing the list at
openstack-ow...@lists.openstack.org
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Openstack digest..."
Today's Topics:
1. How to enable jumbo frames for instances (Ageeleshwar Kandavelu)
2. Glance suddenly unhappy with Keystone behind Apache
(Jonathan Proulx)
3. [Keystone] Difference in values returned after authentication
(Shrinand Javadekar)
4. [glance] problem with glance registry (gustavo panizzo <gfa>)
5. Re: How to enable jumbo frames for instances (Li, Chen)
6. Re: How to enable jumbo frames for instances (Steve Rieger)
7. Re: [glance] problem with glance registry (Remo Mattei)
8. Re: Need help on openstack mellanox quantum plugin setup
(nishant kumar)
----------------------------------------------------------------------
Message: 1
Date: Mon, 27 Jan 2014 13:56:46 +0000
From: Ageeleshwar Kandavelu <ageeleshwar.kandav...@csscorp.com>
To: "chen...@intel.com" <chen...@intel.com>
Cc: "openstack@lists.openstack.org" <openstack@lists.openstack.org>
Subject: [Openstack] How to enable jumbo frames for instances
Message-ID:
<7ce078ce0fb4a743b287bf573132b97708ae8...@inchespvw005.ad.csscorp.com>
Content-Type: text/plain; charset="iso-8859-1"
Hi,
Is it possible to have a higher MTU when your ethernet is limited to 1500?
Thank you,
Ageeleshwar K
http://www.csscorp.com/common/email-disclaimer.php
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.openstack.org/pipermail/openstack/attachments/20140127/42fa3de3/attachment-0001.html>
------------------------------
Message: 2
Date: Mon, 27 Jan 2014 11:58:52 -0500
From: Jonathan Proulx <j...@jonproulx.com>
To: "openstack@lists.openstack.org" <openstack@lists.openstack.org>
Subject: [Openstack] Glance suddenly unhappy with Keystone behind
Apache
Message-ID:
<CABZB-sjxemqv=p8xjc4hdzvmff-ywj0otp0nd-p6rkb3wbp...@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Hi All,
This was working fine until I rebooted to clear an issue with network
namespaces...
glance consistently gives authentication failures if I'm running
keystone in wsgi mode behind apache, all other services (cinder, nova,
neutron, and keytone) respond normally.
changing nothing else, if I run keystone in stand alone eventlet mode
everything works, well until it crawls to a stop due to overloading
the single keystone process.
tail /var/log/glance/api.log
2014-01-27 11:44:31.877 10634 ERROR
keystoneclient.middleware.auth_token [-] Unable to get version info
from keystone: 301
2014-01-27 11:44:31.878 10634 WARNING
keystoneclient.middleware.auth_token [-] Authorization failed for
token 5d49732f701d43ea831ff2c9bd4de5de
Help?
-Jon
------------------------------
Message: 3
Date: Mon, 27 Jan 2014 10:30:06 -0800
From: Shrinand Javadekar <shrin...@maginatics.com>
To: "openstack@lists.openstack.org" <openstack@lists.openstack.org>
Subject: [Openstack] [Keystone] Difference in values returned after
authentication
Message-ID:
<CABppvi74Zs9=jkc3uegvwfnrfsurndtvqejnab4vh2d5omd...@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Hi,
I am seeing a difference in the values returned by Keystone when a
user is authenticated. These differences are in the endpoints section
of the serviceCatalog.
In one instances, I see the returned value has an "id":
"serviceCatalog": [
{
"endpoints": [
{
"adminURL": "http://server.domain.com/v1";,
"region": "egionOne",
"internalURL": "http://server.domain.com/v1/AUTH_foo";,
"id": "0123",
"publicURL": "http://server.domain.com/v1/AUTH_foo";
}
...
whereas in another instances, I see a "versionId":
"serviceCatalog": [
{
"endpoints": [
{
"adminURL": "https://server1.domain1.com/v1/blah";,
"region": "everywhere",
"tenantId": "blahTenant",
"versionId": "1",
"internalURL": "https://server1.domain1.com/v1";,
"publicURL": "https://server1.domain1.com//blah";
}
...
The official documentation of Keystone v2.0 [1], does not mention
anything about "id". It only shows a "versionId". What causes these
differences? And is there a way to configure keystone so that it
always returns one standard set of values?
Thanks in advance.
-Shri
[1]
http://docs.openstack.org/api/openstack-identity-service/2.0/content/POST_authenticate_v2.0_tokens_.html#POST_authenticate_v2.0_tokens_-Response
------------------------------
Message: 4
Date: Mon, 27 Jan 2014 16:28:58 -0300
From: "gustavo panizzo <gfa>" <g...@zumbi.com.ar>
To: "openstack@lists.openstack.org" <openstack@lists.openstack.org>
Subject: [Openstack] [glance] problem with glance registry
Message-ID: <52e6b37a.5090...@zumbi.com.ar>
Content-Type: text/plain; charset=ISO-8859-1
Hello
since today our glance install is not working, we are not able to boot
new machines
the only thing i see in logs is
2014-01-27 16:10:43.364 9237 INFO glance.registry.client.v1.client
[29e031a9-cb1c-4b20-8a7f-100ed10f3785 184136c2674746b5a4b27d3b13ca91f8
af41371701004068b2532695ed57f16f] Registry client request GET
/images/detail raised ServerError
on glance-registry log there is nothing like an error
2014-01-27 16:10:43.333 8771 INFO glance.registry.api.v1.images
[7c168813-dc04-4b65-ad00-0cc171afb10a 184136c2674746b5a4b27d3b13ca91f8
af41371701004068b2532695ed57f16f] Returning detailed image list
glance index fails. but it always failed for us with more than 10
images, is it expected?
# glance --debug index
[ snip snip ]
size": 3280732160}, {"status": "active", "name": "Win7x86", "deleted":
false, "container_format": "ovf", "created_at":
"2014-01-13T21:32:39.412830", "disk_format": "qcow2", "updated_at":
"2014-01-14T14:38:09.932192", "min_disk": 40, "protected": true, "id":
"d0d3b1f4-3e9e-430e-a60a-42de277291da", "min_ram": 2048, "checksum":
"f44b61df38f3d1de775a989eccc1798f", "owner":
"af41371701004068b2532695ed57f16f", "is_public": true, "deleted_at":
null, "properties": {"protected": null}, "size": 3374317568}]}
f31c2938-89b4-4fed-b78c-0d5a633313de debian-wheezy-amd64_2014-01-17
qcow2 ovf 996999168
16ec85ff-6451-4d36-8908-c5c3b57c862b wheezy-cristian-snap ami
ami 1067450368
7fbd62fd-72e6-4f18-ba26-3df9f8295f60 couchdb-snapshot ami
ami 1150812160
6051f420-c590-43f3-a3b1-2802a74d6dc4 vmlinuz-3.2.0-4-amd64 aki
aki 2827712
680680ef-0207-44fe-ad9f-6e01a4e0147a initrd-3.2.0-4-amd64 ari
ari 10486978
11339ff2-9240-4c44-ad8b-091192e0301b wheezy_postgresql
qcow2 bare 1071448064
44d691b8-57e3-44df-9d29-14a6c1f4a431 Win2012-sql
qcow2 ovf 13241942016
784e7546-d101-4cb0-92e0-26faf7f6dfaf Win2012
qcow2 ovf 4284743680
25e38d31-cf2b-4a13-b1bd-90936d90eab4 Win2008r2
qcow2 ovf 3280732160
d0d3b1f4-3e9e-430e-a60a-42de277291da Win7x86
qcow2 ovf 3374317568
curl -i -X GET -H 'X-Auth-Token: c88edf81cae841f4a5638f7267074160' -H
'Content-Type: application/json' -H 'User-Agent: python-glanceclient'
http://10.17.10.34:9292/v1/images/detail?marker=d0d3b1f4-3e9e-430e-a60a-42de277291da&limit=10
HTTP/1.1 500 Internal Server Error
date: Mon, 27 Jan 2014 19:14:28 GMT
content-length: 0
content-type: text/plain
connection: close
Request returned failure status.
HTTPInternalServerError (HTTP 500)
we use rbd as backend storage and pgsql as db
our install runs havana 2013.2.1-2
keystone uses memcached backend and uuid, we have tried restarting
glance-api and registry, keystone and memcached, but nothing.
db looks exactly the same as friday
no changes were made during the weekend. friday we were running smoothly
PS: i can send logs, config files, whatever, but they are pretty default
thanks!
--
1AE0 322E B8F7 4717 BDEA BF1D 44BB 1BA7 9F6C 6333
------------------------------
Message: 5
Date: Tue, 28 Jan 2014 01:32:26 +0000
From: "Li, Chen" <chen...@intel.com>
To: Ageeleshwar Kandavelu <ageeleshwar.kandav...@csscorp.com>
Cc: "openstack@lists.openstack.org" <openstack@lists.openstack.org>
Subject: Re: [Openstack] How to enable jumbo frames for instances
Message-ID:
<988e98d31b01e44893af6e48ed9defd4019b9...@shsmsx101.ccr.corp.intel.com>
Content-Type: text/plain; charset="us-ascii"
Hi Ageeleshwar,
I guess I just asked similar question yesterday.
But, I'm not understand what you mean "when your ethernet is limited to 1500".
Not sure if we're talking at the right direction.
Anyway, here is what I know:
This is based on if you're using neutron + ovs.
1. Configure /etc/neutron/plugins/openvswitch/ovs_neutron_plugin.ini
veth_mtu = ${your_MTU}
If you're using gre/vxlan, this configure can be ignored.
2. Configure /etc/nova/nova.conf
network_device_mtu= ${your_MTU}
But this may not work due to bug https://bugs.launchpad.net/nova/+bug/1260697.
Unless you apply the fix patch.
3. Configure /etc/neutron/ dhcp_agent.ini
dnsmasq_config_file = /etc/neutron/dnsmasq-neutron.conf
And add to /etc/neutron/dnsmasq-neutron.conf:
dhcp-option-force=26,${your_MTU}
I'm working under CentOS 6.4+ Havana + OVS + GRE.
And I just tested it, It works fine to me.
I have configured MTU=8500 in all configure files, while the physical NIC used
for OVS (local IP) is using MTU=9000.
Thanks.
-chen
From: Ageeleshwar Kandavelu [mailto:ageeleshwar.kandav...@csscorp.com]
Sent: Monday, January 27, 2014 9:57 PM
To: Li, Chen
Cc: openstack@lists.openstack.org
Subject: [Openstack] How to enable jumbo frames for instances
Hi,
Is it possible to have a higher MTU when your ethernet is limited to 1500?
Thank you,
Ageeleshwar K
http://www.csscorp.com/common/email-disclaimer.php
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.openstack.org/pipermail/openstack/attachments/20140128/fa54a739/attachment-0001.html>
------------------------------
Message: 6
Date: Mon, 27 Jan 2014 17:48:03 -0800
From: Steve Rieger <ste...@up-south.com>
To: openstack@lists.openstack.org
Subject: Re: [Openstack] How to enable jumbo frames for instances
Message-ID: <52e70c53.2010...@up-south.com>
Content-Type: text/plain; charset="iso-8859-1"
I am assuming that the OP meant that his upstream is set to 1500
Jumbo frames has to be set end to end, all devices.
Otherwise you will be introducing more errors...
On 01/27/2014 05:32 PM, Li, Chen wrote:
>
> Hi Ageeleshwar,
>
>
>
> I guess I just asked similar question yesterday.
>
>
>
> But, I'm not understand what you mean "when your ethernet is limited
> to 1500".
>
>
>
> Not sure if we're talking at the right direction.
>
> Anyway, here is what I know:
>
>
>
> This is based on if you're using neutron + ovs.
>
>
>
> 1. Configure /etc/neutron/plugins/openvswitch/ovs_neutron_plugin.ini
>
> veth_mtu = ${your_MTU}
>
>
>
> If you're using gre/vxlan, this configure can be ignored.
>
>
>
> 2. Configure /etc/nova/nova.conf
>
> network_device_mtu= ${your_MTU}
>
>
>
> But this may not work due to bug
> https://bugs.launchpad.net/nova/+bug/1260697.
>
> Unless you apply the fix patch.
>
>
>
> 3. Configure /etc/neutron/ dhcp_agent.ini
>
> dnsmasq_config_file = /etc/neutron/dnsmasq-neutron.conf
>
>
>
> And add to /etc/neutron/dnsmasq-neutron.conf:
>
> dhcp-option-force=26,${your_MTU}
>
>
>
>
>
> I'm working under CentOS 6.4+ Havana + OVS + GRE.
>
> And I just tested it, It works fine to me.
>
> I have configured MTU=8500 in all configure files, while the physical
> NIC used for OVS (local IP) is using MTU=9000.
>
>
>
> Thanks.
>
> -chen
>
>
>
> *From:*Ageeleshwar Kandavelu [mailto:ageeleshwar.kandav...@csscorp.com]
> *Sent:* Monday, January 27, 2014 9:57 PM
> *To:* Li, Chen
> *Cc:* openstack@lists.openstack.org
> *Subject:* [Openstack] How to enable jumbo frames for instances
>
>
>
> Hi,
> Is it possible to have a higher MTU when your ethernet is limited to 1500?
>
> Thank you,
> Ageeleshwar K
>
> http://www.csscorp.com/common/email-disclaimer.php
>
>
>
> _______________________________________________
> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> Post to : openstack@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/20140127/f8e8ff3e/attachment-0001.html>
------------------------------
Message: 7
Date: Mon, 27 Jan 2014 13:48:16 -0600
From: Remo Mattei <r...@italy1.com>
To: "gustavo panizzo <gfa>" <g...@zumbi.com.ar>
Cc: "openstack@lists.openstack.org" <openstack@lists.openstack.org>
Subject: Re: [Openstack] [glance] problem with glance registry
Message-ID: <03e7b78b-b5c0-486c-bbcc-97ad4c212...@italy1.com>
Content-Type: text/plain; charset=utf-8
What os are you running? Maybe they cannot reach each other i would check that
Inviato da IPad (?)
> Il giorno Jan 27, 2014, alle ore 13:28, "gustavo panizzo <gfa>"
> <g...@zumbi.com.ar> ha scritto:
>
> Hello
> since today our glance install is not working, we are not able to boot
> new machines
>
>
> the only thing i see in logs is
>
> 2014-01-27 16:10:43.364 9237 INFO glance.registry.client.v1.client
> [29e031a9-cb1c-4b20-8a7f-100ed10f3785 184136c2674746b5a4b27d3b13ca91f8
> af41371701004068b2532695ed57f16f] Registry client request GET
> /images/detail raised ServerError
>
> on glance-registry log there is nothing like an error
>
> 2014-01-27 16:10:43.333 8771 INFO glance.registry.api.v1.images
> [7c168813-dc04-4b65-ad00-0cc171afb10a 184136c2674746b5a4b27d3b13ca91f8
> af41371701004068b2532695ed57f16f] Returning detailed image list
>
>
> glance index fails. but it always failed for us with more than 10
> images, is it expected?
>
> # glance --debug index
>
> [ snip snip ]
>
> size": 3280732160}, {"status": "active", "name": "Win7x86", "deleted":
> false, "container_format": "ovf", "created_at":
> "2014-01-13T21:32:39.412830", "disk_format": "qcow2", "updated_at":
> "2014-01-14T14:38:09.932192", "min_disk": 40, "protected": true, "id":
> "d0d3b1f4-3e9e-430e-a60a-42de277291da", "min_ram": 2048, "checksum":
> "f44b61df38f3d1de775a989eccc1798f", "owner":
> "af41371701004068b2532695ed57f16f", "is_public": true, "deleted_at":
> null, "properties": {"protected": null}, "size": 3374317568}]}
>
> f31c2938-89b4-4fed-b78c-0d5a633313de debian-wheezy-amd64_2014-01-17
> qcow2 ovf 996999168
> 16ec85ff-6451-4d36-8908-c5c3b57c862b wheezy-cristian-snap ami
> ami 1067450368
> 7fbd62fd-72e6-4f18-ba26-3df9f8295f60 couchdb-snapshot ami
> ami 1150812160
> 6051f420-c590-43f3-a3b1-2802a74d6dc4 vmlinuz-3.2.0-4-amd64 aki
> aki 2827712
> 680680ef-0207-44fe-ad9f-6e01a4e0147a initrd-3.2.0-4-amd64 ari
> ari 10486978
> 11339ff2-9240-4c44-ad8b-091192e0301b wheezy_postgresql
> qcow2 bare 1071448064
> 44d691b8-57e3-44df-9d29-14a6c1f4a431 Win2012-sql
> qcow2 ovf 13241942016
> 784e7546-d101-4cb0-92e0-26faf7f6dfaf Win2012
> qcow2 ovf 4284743680
> 25e38d31-cf2b-4a13-b1bd-90936d90eab4 Win2008r2
> qcow2 ovf 3280732160
> d0d3b1f4-3e9e-430e-a60a-42de277291da Win7x86
> qcow2 ovf 3374317568
> curl -i -X GET -H 'X-Auth-Token: c88edf81cae841f4a5638f7267074160' -H
> 'Content-Type: application/json' -H 'User-Agent: python-glanceclient'
> http://10.17.10.34:9292/v1/images/detail?marker=d0d3b1f4-3e9e-430e-a60a-42de277291da&limit=10
>
> HTTP/1.1 500 Internal Server Error
> date: Mon, 27 Jan 2014 19:14:28 GMT
> content-length: 0
> content-type: text/plain
> connection: close
>
> Request returned failure status.
> HTTPInternalServerError (HTTP 500)
>
>
> we use rbd as backend storage and pgsql as db
> our install runs havana 2013.2.1-2
> keystone uses memcached backend and uuid, we have tried restarting
> glance-api and registry, keystone and memcached, but nothing.
> db looks exactly the same as friday
>
> no changes were made during the weekend. friday we were running smoothly
>
>
> PS: i can send logs, config files, whatever, but they are pretty default
>
> thanks!
>
>
> --
> 1AE0 322E B8F7 4717 BDEA BF1D 44BB 1BA7 9F6C 6333
>
>
> _______________________________________________
> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> Post to : openstack@lists.openstack.org
> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>
> !DSPAM:1,52e6b60e31611208556276!
>
------------------------------
Message: 8
Date: Mon, 27 Jan 2014 19:56:56 -0800 (PST)
From: nishant kumar <nishantkumar...@yahoo.com>
To: Irena Berezovsky <ire...@mellanox.com>
Cc: Openstack <openst...@mellanox.com>,
"openstack@lists.openstack.org" <openstack@lists.openstack.org>
Subject: Re: [Openstack] Need help on openstack mellanox quantum
plugin setup
Message-ID:
<1390881416.33636.yahoomail...@web142305.mail.bf1.yahoo.com>
Content-Type: text/plain; charset="utf-8"
Hey Irena,
Thanks for your reply. I will try what you suggested.
I couldn't find instructions directly under the link you provided. I think you
meant this:
http://community.mellanox.com/docs/DOC-1317
I got it from the mellanox community.
?
Thanks,
Nishant
On Monday, January 27, 2014 2:26 PM, Irena Berezovsky <ire...@mellanox.com>
wrote:
Hi Nishant,
Mellanox plugin supports two types of VF provisioning , as pci passthrough
(hostdev) and as macvtap (mlnx_direct) vNIC.
According to the log, you want to use the first flavor (hostdev).
?
Please follow the following instructions in:
http://Community.mellanox.com/community/develop/ - After the installation
perform the following changes for ?setting the ?probe_vf? ?option in mlx4_core.
In case you want all VFs to be used in pci passthrough mode, set the probe_vf =
0.
?
In case you need further assistance, please contact us at openst...@mellanox.com
?
Best Regards,
Irena
?
?
From:nishant kumar [mailto:nishantkumar...@yahoo.com]
Sent: Monday, January 27, 2014 10:19 AM
To: openstack@lists.openstack.org
Subject: [Openstack] Need help on openstack mellanox quantum plugin setup
?
Hi there!
I have setup a compute node with openstack networking enabled by the mellanox
neutron plugin on RH6.4. I have enabled SR-IOV and created 4 VFs. They all show
up in ifconfig. I tried creating an instance using 'nova boot' command. However
it is in error state.
I created network, subnet and ports successfully by following this doc:
http://www.mellanox.com/related-docs/applications/Mellanox-OpenStack-Solution-for-Red-Hat.pdf
I created an instance using following command:
nova boot --flavor 1 --image 7f8460fd-6928-49ae-bef8-055a1de3e39a cirrOS --nic
port-id=98f537b2-27dc-4583-ad1a-85edbd74987a
The instance was in error state and eswitchd.log shows this:
2014-01-26 09:41:36,549 DEBUG PF eth4, eths=['eth6', 'eth7', 'eth8', 'eth9'],
vfs=[]
- - - - - - - - - - - - truncated - - - - - - - - - - - -
2014-01-26 10:43:13,320 DEBUG vnics are {}
2014-01-26 10:43:13,746 DEBUG Handling message - {u'action': u'get_vnics',
u'fabric': u'*'}
2014-01-26 10:43:13,746 DEBUG fabrics =['default']
2014-01-26 10:43:13,747 DEBUG vnics are {}
2014-01-26 10:43:14,565 DEBUG Handling message - {u'fabric': u'default',
u'dev_name': None, u'vnic_type': u'hostdev', u'action': u'create_port',
u'vnic_mac': u'fa:16:3e:c9:c2:23', u'device_id':
u'e106da24-9b10-4c4c-87a0-66276f025c6e'}
2014-01-26 10:43:14,565 ERROR exception on device allocation on dev? None - No
available resources.
2014-01-26 10:43:14,566 ERROR Create port operation failed
My queries are:
??? - Does my setup look ok?
??? - I guess since I am using SR-IOV, the adapter is presenting VFs as PFs to
OS. Is this understanding correct?
??? - vfs list is empty (Log says vfs=[]). Is this OK?
Please comment/advise on what could be the issue.
Thanks,
Nishant
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.openstack.org/pipermail/openstack/attachments/20140127/c661b11c/attachment-0001.html>
------------------------------
_______________________________________________
Openstack mailing list
openstack@lists.openstack.org
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
End of Openstack Digest, Vol 7, Issue 32
****************************************
--------------------------------------------------------------
Intel Shannon Limited
Registered in Ireland
Registered Office: Collinstown Industrial Park, Leixlip, County Kildare
Registered Number: 308263
Business address: Dromore House, East Park, Shannon, Co. Clare
This e-mail and any attachments may contain confidential material for the sole
use of the intended recipient(s). Any review or distribution by others is
strictly prohibited. If you are not the intended recipient, please contact the
sender and delete all copies.
_______________________________________________
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
