Ask Your Question

armando-migliaccio's profile - activity

2017-07-18 05:07:41 -0500 received badge  Famous Question (source)
2017-06-05 10:19:44 -0500 received badge  Notable Question (source)
2017-06-05 10:19:44 -0500 received badge  Famous Question (source)
2017-06-05 10:19:44 -0500 received badge  Popular Question (source)
2014-10-30 07:40:05 -0500 received badge  Notable Question (source)
2014-09-17 13:22:14 -0500 received badge  Student (source)
2014-08-28 15:57:56 -0500 received badge  Popular Question (source)
2013-08-02 11:01:08 -0500 answered a question How do you configure glance to allow snapshots but not image creation?

Glance has no concept of differential images as far as I can tell (or at least not yet), so creating/uploading snapshots is equivalent to creating/uploading images; hence you cannot grant access to one and deny the other. That said, if you are talking about snapshotting of running VM's, your low-privs users can still snapshot instances by using the Compute API directly. Denying access to the Glance API altogether will allow you to prevent your users from creating images from scratch.

2013-08-02 10:51:59 -0500 answered a question nova-compute and nova-scheduler keep going down

are the scheduler and compute services running on the same host? Ensure their clocks are in sync. It's unlikely that the compute log has no traces of error, but the scheduler's might.

2013-07-24 02:54:36 -0500 received badge  Great Answer (source)
2013-07-16 17:26:43 -0500 answered a question Unable to load the virtualization driver: No module named vmwareapi

You might be missing a package dependency for vmwareapi. Pasting more information would be useful to understand what's going on.

2013-07-16 11:24:33 -0500 answered a question What's the minimal stable architecture?

If the crash is not catastrophic, VM's can boot up after the server has been recovered. They can even boot up automatically. With XCP you can rely on both live and block storage migration. The former requires shared storage, the latter does not (only from XCP 1.6 and OpenStack Folsom).

2013-07-11 20:07:38 -0500 answered a question why my controller node in hypervisor-list

if your controller runs a service called nova-compute you can disable it by using nova-manage.

You can do so:

nova-manage service disable [-h] [--host <host>] [--service <service>]

2013-07-10 10:45:46 -0500 answered a question which language for openstack

For the sake of OpenStack, it does not matter. You should use the language that matches your keyboard layout, really so that you'd be comfortable typing.

2013-07-10 10:43:42 -0500 answered a question auto assignment of floating ip

That works only when using nova-network. It won't work with Neutron I'm afraid.

2013-07-10 10:42:28 -0500 answered a question How to setup VPN As A Service with Neutron

You need a fresh node and follow the instructions under the Installation paragraph.

2013-07-10 10:40:32 -0500 answered a question How can i inject a file into windows instance in openstack?

The short answer is: you may need a newer version of OpenStack. Long answer is: you could try to use the metadata server and use an equivalent of cloud-init for Windows. Have a look at this great post:

cloud-init-for-windows

2013-07-10 10:33:38 -0500 answered a question devstack failing: conflicts with requirement python-keystoneclient

I think you may have one of the projects that uses python-keystoneclient out of sync with this particular library. Try to update your python-keystoneclient local repo and you should be good.

2013-07-04 02:03:02 -0500 received badge  Nice Answer (source)
2013-07-02 10:21:22 -0500 answered a question How to create images

Have a look at here:

Image management

and in particular:

Creating Custom Images

See if they help!

2013-06-27 22:07:39 -0500 answered a question I think the port should not be deleted when the interface is detched,if the port is created independently.

They just help provide a little more context.

2013-06-27 22:07:01 -0500 answered a question I think the port should not be deleted when the interface is detched,if the port is created independently.

I believe the bug reports above are somewhat related to the issue we are seeing here, i.e. the port gets deleted when the VM is destroyed.

2013-06-25 12:33:19 -0500 received badge  Good Answer (source)
2013-06-25 10:05:23 -0500 received badge  Nice Answer (source)
2013-06-25 09:38:45 -0500 commented answer Where can I get more details on setting up OpenStack for HA?

As for Keystone, what auth backend are you using? If using a simple datastore like SQL, you can run multiple keystone servers and memcache for their token backend.

2013-06-25 09:36:55 -0500 commented answer Where can I get more details on setting up OpenStack for HA?

For your queue clustering options please have a look at: http://www.rabbitmq.com/clustering.html and http://www.rabbitmq.com/ha.html, as it pretty much depends on your needs.

2013-06-24 11:00:29 -0500 answered a question Where can I get more details on setting up OpenStack for HA?

An OpenStack controller node may run multiple OpenStack services: for instance, API and Scheduler are some of these services. For a definition of controller node and its running services please have a look at here:

Controller Node

This is a configuration that is typically adopted for smaller or PoC deployments. In this particular configuration, if your controller were to die, your OpenStack cloud would die. To achieve true scale and avoid SPoF (Single Point of Failure), you need to distribute these services. For instance:

  • Run your DB and Queue in dedicated clustered nodes
  • Scale out Glance servers, use a distributed file storage solution like swift as backend
  • Scale out Keystone servers
  • Run multiple Nova API and Schedulers
  • ...

And so on. For an architectural overview of the project, have a look at:

Architecture

For High Availability considerations have a look at OpenStack-HA

For your queue clustering options please have a look at: http://www.rabbitmq.com/clustering.html and http://www.rabbitmq.com/ha.html, as it pretty much depends on your needs.

As for Keystone, what auth backend are you using? If using a simple datastore like SQL, you can run multiple keystone servers and memcache for their token backend.

2013-06-21 15:46:53 -0500 answered a question quantum agent list

quantum agent-list is an admin-only feature that shows what agents, like l2 and l3 (for plugins that support this extension) are available. To see what compute nodes are available in your deployment use the nova CLI.

2013-06-20 10:45:43 -0500 answered a question How to get data from forms Horizon Dashboard

I would suggest you to redirect your question to the openstack-dev Mailing List, as it will draw attention from a more appropriate audience.

2013-06-20 10:40:10 -0500 answered a question Quantum DHCP Issue

There's very little information to give you any sort of useful input. However it seems like your VM's and DHCP agent cannot see each other, and this may be for a number of reasons. In order to troubleshoot this further we would need to understand your network topology and what quantum plugins you are using. Please provide things like:

  • config files
  • log files
  • host networking configs (e.g. if you are using ovs, a dump of ovs-vsctl show would be really useful, especially on the dhcp side).
2013-06-17 15:43:01 -0500 answered a question I think the port should not be deleted when the interface is detched,if the port is created independently.

Currently there is no distinction between a port created explicitly by a user or admin and a port created by the infrastructure (in this case Nova), so all ports are equal and disposed of when the resource that is using it is destroyed (in this case a VM).

I think it is fair to create a bug report and mark the bug as wishlist or opinion to see what people think about this particular issue.

2013-06-09 03:07:52 -0500 received badge  Nice Answer (source)
2013-06-06 12:33:51 -0500 commented answer Glance unable to download images, cannot create instances.

Typically Glance endpoints use port 9292. Why are you using port 80? As for the cinder rootwrapper, check also permissions to the file

2013-06-05 12:15:39 -0500 answered a question Glance unable to download images, cannot create instances.

As for the Cinder issue, check file:

/etc/cinder/rootwrap.d/volume.filters

it should contain something like this:

# cinder-rootwrap command filters for volume nodes
# This file should be owned by (and only-writeable by) the root user

[Filters]
# cinder/volume/iscsi.py: iscsi_helper '--op' ...
ietadm: CommandFilter, ietadm, root
tgtadm: CommandFilter, tgtadm, root
tgt-admin: CommandFilter, tgt-admin, root
rtstool: CommandFilter, rtstool, root

# cinder/volume/driver.py: 'vgs', '--noheadings', '-o', 'name'
vgs: CommandFilter, vgs, root

# cinder/volume/driver.py: 'lvcreate', '-L', sizestr, '-n', volume_name,..
# cinder/volume/driver.py: 'lvcreate', '-L', ...
lvcreate: CommandFilter, lvcreate, root

# cinder/volume/driver.py: 'dd', 'if=%s' % srcstr, 'of=%s' % deststr,...
dd: CommandFilter, dd, root

# cinder/volume/driver.py: 'lvremove', '-f', %s/%s % ...
lvremove: CommandFilter, lvremove, root

# cinder/volume/driver.py: 'lvdisplay', '--noheading', '-C', '-o', 'Attr',..
lvdisplay: CommandFilter, lvdisplay, root

# cinder/volume/driver.py: 'iscsiadm', '-m', 'discovery', '-t',...
# cinder/volume/driver.py: 'iscsiadm', '-m', 'node', '-T', ...
iscsiadm: CommandFilter, iscsiadm, root

.....

The error complaint is about a missing 'Filters' section; that makes me think there is something wrong with this file.

As for the Nova issue, it looks like the image location is None. Try:

glance show 960aa9a4-78cf-417b-8171-3a2bdab103f3

That should tell you the URI for the image, and troubleshoot this further.

2013-06-05 11:44:28 -0500 received badge  Commentator
2013-06-05 11:44:28 -0500 commented question How to attach Windows volumes to OpenStack nova?

Please provide more information about your environment, e.g. hypervisor, running services, list of commands executed, and possibly even logs. That helps us to give you a more detailed answer.

2013-06-04 15:06:55 -0500 answered a question Xen - Compute node is trying to run pygrub in DomU - bug?

There is nothing wrong with Grizzly and Xen per se. From your log it looks like you are trying to boot a VM in para-virtualized mode, and you haven't specified an external ramdisk and kernel. To this aim, your compute service is trying to find them inside the image using pygrub. However pygrub does not look to be installed correctly. See message:

Error while executing pygrub! Please, ensure the binary is installed correctly, and available in your PATH; on some Linux distros, pygrub may be installed in /usr/lib/xen-X.Y/bin/pygrub. Attempting to boot in HVM mode.

With XenServer and nova-compute in domU, pygrub must be available in your CentOS guest otherwise you won't be able to boot raw images, but only vhd's based or three-part images.

2013-05-29 12:08:25 -0500 answered a question Migrating from nova-network to quantum

As far as I can tell, I am not aware of any migration path that preserves network connectivity for your VM's; if such path does exists, this may only work in specific settings (i.e. with specific hypervisors, storage, etc).

A more viable approach is snapshotting your VM's into Glance prior to the migration from nova-network to Quantum and redeploying them after the upgrade. This causes some downtime, but this can be minimized if you run the two environments (nova-network based and quantum-based) along-side each other.

2013-05-15 06:04:38 -0500 commented answer nova-compute rabbitmq fail to connect

Your line:

dhcpbridge_flagfile=/etc/nova/nova.conf

Does not look right.

2013-05-15 05:59:16 -0500 commented answer What causes error "service n-net is not running" when running stack.sh?

File in question is /etc/nova/nova.conf; not the one under /opt/stack

2013-05-15 05:58:11 -0500 commented answer "Could not find Service or Region in Service Catalog" trying quantum net-list

how about the admin account that quantum server user use to access keystone?

2013-05-15 05:56:31 -0500 commented answer how do I disable a nova compute node?

I don't think that's the right way to go about it. Using nova/nova-manage is the best option.

2013-05-15 05:55:21 -0500 commented answer how do I disable a nova compute node?

there's no need to restart these services, the update happens on the fly.

2013-05-14 14:02:35 -0500 received badge  Editor (source)
2013-05-14 14:02:35 -0500 edited answer What causes error "service n-net is not running" when running stack.sh?

It looks like there might be something wrong with your sql connection string (it's either malformed or empty).

Check the content of /etc/nova.conf to investigate.

A similar error was reported here:

bug #1142782

But it looks a lot like the problem you are having, but it is still an open issue.