Manuel,
Rather than with aggregate metadata we assign instances to NUMA nodes via
flavor extra_specs,
i.e. nova flavor-show reports something like
—>
| extra_specs| {"hw:numa_nodes": "1"} |
<—
for our NUMA-aware flavors.
This seems to work pretty well and gives the desired p
How your trying to add them is the main issue.
Source openrc with admin credentials and try to add the aggregate host.
Hope this helps.
/ Trinath
From: Manuel Sopena Ballesteros [mailto:manuel...@garvan.org.au]
Sent: Tuesday, February 07, 2017 5:49 AM
To: openstack@lists.openstack.org
Subject:
Hi,
I am trying to deploy cinder-backup service and c-vol uses multi-backend,
the configuration as follows:
...
enabled_backends = ceph1, ceph2
backup_driver = cinder.backup.drivers.ceph
backup_ceph_conf=/etc/ceph/ceph.conf
backup_ceph_user = admin
backup_ceph_chunk_size = 134217728
backup_ceph_p
Hi,
I am trying to isolate my instances by cpu socket in order to improve my NUMA
hardware performance.
[root@openstack-dev ~(keystone_admin)]# nova aggregate-set-metadata numa
pinned=true
Metadata has been successfully updated for aggregate 1.
+++++-
[replying to myself for posterity]
On 5 Feb 2017, at 20:19, Turbo Fredriksson wrote:
> I can’t see anything wrong here!
Turned out to be a time sync problem. The two machines was about three
minutes off.
I have a NTP server locally, but there seems to be something wrong with it.
Need to invest
