Lawrence Lo's profile - activity

2017-07-30 16:47:54 -0500 received badge  Nice Answer (source)
2017-07-28 09:16:40 -0500 answered a question Why does a compute node require at least 2 network interfaces (minimal architecture ) ??

Usually in data centers, there are 4 types of networks in an OpenStack deployment. I think the minimum two networks you mentioned are External network and Management Network.

1 External network: this is to access to OpenStack API from the Internet in the case of a public cloud or intranet in the case of a private cloud.

2 Management Network: a private network which is used for communication between the OpenStack control plane and the compute infrastructure. The message bus and database are on this network.

3 Tenant network

4 Storage Network

2017-07-28 08:22:55 -0500 received badge  Editor (source)
2017-07-28 08:03:15 -0500 answered a question What is the equivalent command for user-role-list in openstack client for V3 api ?

In Ocata release, you will encounter this error:

# openstack role list --user swift
Listing assignments using role list is deprecated. Use role assignment list --user <user-name> --domain default --names instead.

Then you use this instead

# openstack role assignment list --names
+---------------+--------------------+-------+------------------+--------+-----------+
| Role          | User               | Group | Project          | Domain | Inherited |
+---------------+--------------------+-------+------------------+--------+-----------+
| admin         | neutron@Default    |       | services@Default |        | False     |
| admin         | placement@Default  |       | services@Default |        | False     |
| admin         | glance@Default     |       | services@Default |        | False     |
| admin         | cinder@Default     |       | services@Default |        | False     |
| admin         | aodh@Default       |       | services@Default |        | False     |
| admin         | admin@Default      |       | admin@Default    |        | False     |
| admin         | nova@Default       |       | services@Default |        | False     |
| admin         | swift@Default      |       | services@Default |        | False     |
| admin         | gnocchi@Default    |       | services@Default |        | False     |
| ResellerAdmin | ceilometer@Default |       | services@Default |        | False     |
| admin         | ceilometer@Default |       | services@Default |        | False     |
+---------------+--------------------+-------+------------------+--------+-----------+
2015-11-25 12:37:54 -0500 answered a question No valid host was found. Unable to launch instance in admin[KILO][UBUNTU 14.04]

Please check firewall on the controller node

2015-10-02 10:06:52 -0500 commented answer Unable to install keystone packageI

If your controller does not have a valid yum repo, it will end up in a package dependency hell. If for some reason you cannot register Red Hat, I would recommend you use Fedora or Centos.

2015-10-01 17:32:38 -0500 answered a question Unable to install keystone packageI

When you run "yum list python-webob", does it show as "Available Packages"? I suspect yum cannot find a valid repo that contains the required "python-webob" package.

If your controller does not have a valid yum repo, it will end up in a package dependency hell. If for some reason you cannot register Red Hat, I would recommend you use Fedora or Centos.

2014-10-14 10:10:36 -0500 received badge  Nice Answer (source)
2014-08-30 22:58:38 -0500 answered a question Openstack Single node worked until 1st reboot

I encountered this error before. My case was resolved after neutron bounce. Please try:

service neutron-server restart

2014-08-29 13:03:56 -0500 commented answer why pxe bootloader after os deployed

Thank you. I did not realize IRONIC would do that.

2014-08-27 19:41:08 -0500 answered a question why pxe bootloader after os deployed

If I understand your question correctly : Boot order from PXE/Disk can be configured in your BIOS setting. I don't think OpenStack deployment will change BIOS boot order setting. The scenario could be that in order to setup OpenStack, the MAC address of this bare metal server was registered into the Kickstart server. If the BIOS boot order in the bare metal server happens to have "boot from network" as the first one in boot options and "boot from disk" as the second, then it will boot from PXE for every reboot. You can fix it by changing the order in the BIOS setting.

2014-08-21 08:29:22 -0500 received badge  Supporter (source)
2014-06-19 10:43:30 -0500 commented question Connection to neutron failed:[Errno 11] Connection refused

please show debug information

neutron --debug net-create ext-net --shared --router:external=True
2014-06-19 09:52:56 -0500 answered a question Display not specified?
2014-06-17 09:03:17 -0500 answered a question how to convert ami/aki/ari image to qcow2?

Hi bishoy,

I did not find a direct way to convert Amazon EC2 ami to qcow2.

Please try follow the steps in the following link to obtain the raw image. And then replace the very last step to covert to qcow2. In the following example, it is trying to convert to vmdk format. Hope it helps.

http://www.smashingboxes.com/convert-...

2014-06-17 00:01:09 -0500 answered a question swift keystone configuration no handlers 401 failure

Please also check out http://docs.openstack.org/admin-guide... To troubleshoot the Identity service, review the logs in the /var/log/keystone/keystone.log file. The logs show the components that have come in to the WSGI request, and ideally show an error that explains why an authorization request failed. If you do not see the request in the logs, run keystone with --debug parameter. Pass the --debug parameter before the command parameters.

2014-06-13 07:41:28 -0500 commented question InvalidBDM: Block Device Mapping is Invalid.

Could you please paste details from compute.log ? There are several exceptions related to InvlidBDM in nova/block_device.py file

2014-06-11 15:22:42 -0500 commented question launching instance fails "add bridge failed: No such process" if network is assigned

Is bridge module loaded ?

lsmod | grep bridge

bridge 83273 0 stp 2218 2 bridge,garp llc 5546 3 bridge,garp,stp

2014-06-11 12:39:41 -0500 commented question launching instance fails "add bridge failed: No such process" if network is assigned

What happens if you run the following command with nova ID on host asds63 ?

sudo nova-rootwrap /etc/nova/rootwrap.conf brctl addbr qbr1d1a5a28-10
2014-06-11 11:03:42 -0500 received badge  Enthusiast
2014-05-28 08:02:02 -0500 commented answer nova compute --Falling back to VFSLocalFS

You are very welcome!

2014-05-27 09:47:31 -0500 answered a question nova compute --Falling back to VFSLocalFS

@Sonia,

Please check if you have the following on your compute node and controller node :

$ rpm -qa | grep guestfs
libguestfs-1.20.11-2.el6.x86_64
libguestfs-tools-c-1.20.11-2.el6.x86_64
python-libguestfs-1.20.11-2.el6.x86_64
2014-05-26 22:52:47 -0500 answered a question Binding floating IP can not access

I don't think the warning message "WARNING nova.virt.disk.vfs.guestf" will block you from accessing the instance. This looks more like a neutron area problem instead of nova. Please post the output from "neutron net-list" and "neutron port-list"

2014-05-26 07:02:18 -0500 received badge  Teacher (source)
2014-05-24 17:52:44 -0500 answered a question RDO Icehouse: Can't boot from volume

@netracer

The warning was from nova/virt/libvirt/driver.py

elif inject_files and CONF.libvirt.inject_partition != -2:
        if booted_from_volume:
            LOG.warn(_('File injection into a boot from volume '
                       'instance is not supported'), instance=instance)
        self._inject_data(
            instance, network_info, admin_pass, files, suffix)

It will still move on to call self._inject_data though.

Does it generate extra logs in /var/log/nova/compute.log ?