<div dir="ltr"><div class="gmail_default" style="font-family:'times new roman',serif"><p>I figured out how to give the instance a virtual NIC. It is getting
an IP address from our DHCP, I can ssh in, and I can surf the web.<br>
</p>
<pre>root@compute1-prod.compute1-prod:/var/log/neutron# ifconfig VLAN_15<br>VLAN_15 Link encap:Ethernet HWaddr 00:25:90:5B:AA:A0 <br> inet addr:10.50.15.48 Bcast:10.50.15.255 Mask:255.255.255.0<br> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1<br>
RX packets:8355781 errors:0 dropped:0 overruns:0 frame:0<br> TX packets:1644788 errors:0 dropped:0 overruns:0 carrier:0<br> collisions:0 txqueuelen:0 <br> RX bytes:5040893113 (4.6 GiB) TX bytes:211825561 (202.0 MiB)<br>
root@compute1-prod.compute1-prod:/var/log/neutron# /etc/init.d/openvswitch start<br><br>ovs-vsctl list-br # you should see nothing<br>ovs-vsctl add-br br-int<br>ovs-vsctl list-br<br>ovs-vsctl add-br VLAN_15<br>ovs-vsctl add-br VLAN_20<br>
ovs-vsctl list-br<br><br>Connect the open virtual switch to the instance<br><br>root@compute1-prod.compute1-prod:/var/log/neutron# virsh attach-interface instance-00000006 bridge VLAN_15<br>Interface attached successfully<br>
<br><br>root@compute1-prod.compute1-prod:/var/log/neutron# <br><br>I rebooted the guest and voila!<br><br>$ ifconfig eth0<br>eth0 Link encap:Ethernet HWaddr 52:54:00:8E:E5:25 <br> inet addr:10.50.15.239 Bcast:10.50.15.255 Mask:255.255.255.0<br>
inet6 addr: fe80::5054:ff:fe8e:e525/64 Scope:Link<br> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1<br> RX packets:1404 errors:0 dropped:0 overruns:0 frame:0<br> TX packets:256 errors:0 dropped:0 overruns:0 carrier:0<br>
collisions:0 txqueuelen:1000 <br> RX bytes:82015 (80.0 KiB) TX bytes:27940 (27.2 KiB)<br> Interrupt:10 Base address:0x6000 <br><br>$ <br>The IP address came from our dhcpd and works fine. dhcpd also provided a default router and populated /etc/resolv.conf<br>
<br>My lead asked to add ports to the VLAN since there are two NICs which are bonded together using LACP:<br><br>root@compute1-prod.compute1-prod:/var/log/neutron# ovs-vsctl list-ports VLAN_15<br>root@compute1-prod.compute1-prod:/var/log/neutron# ovs-vsctl list-ports VLAN_20<br>
root@compute1-prod.compute1-prod:/var/log/neutron# ovs-vsctl list-ports br-int<br>root@compute1-prod.compute1-prod:/var/log/neutron# ovs-vsctl add-port VLAN_15 bond0.15<br>root@compute1-prod.compute1-prod:/var/log/neutron# ovs-vsctl add-port VLAN_20 bond0.20<br>
root@compute1-prod.compute1-prod:/var/log/neutron# ovs-vsctl list-ports VLAN_15<br>bond0.15<br>root@compute1-prod.compute1-prod:/var/log/neutron# ovs-vsctl list-ports VLAN_20<br>bond0.20<br>root@compute1-prod.compute1-prod:/var/log/neutron# <br>
<br>root@compute1-prod.compute1-prod:/var/log/neutron# virsh dumpxml 3 <br><domain>
...<br> <interface type='bridge'><br> <mac address='52:54:00:8e:e5:25'/><br> <source bridge='VLAN_15'/><br> <target dev='vnet1'/><br> <alias name='net0'/><br>
<address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/><br> </interface><br>...<br></domain><br><br>root@compute1-prod.compute1-prod:/var/log/neutron# <br>
</pre><pre><br></pre><pre>So I know that the hypervisor is working okay. This is exactly what I want: I want the compute node to connect direct to the physical switch/router. I don't want to go through neutron, which I think will be a bottleneck. I think I have good enough security upstream to protect the openstack. But how do I do this configuration using neutron and open virtual switch? Or is this the way it is supposed to be done?</pre>
<pre><br></pre><pre>Thank you</pre><pre><br></pre><pre>Jeff</pre><pre><br></pre><pre><br></pre></div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Mon, Aug 18, 2014 at 5:00 AM, <span dir="ltr"><<a href="mailto:openstack-operators-request@lists.openstack.org" target="_blank">openstack-operators-request@lists.openstack.org</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Send OpenStack-operators mailing list submissions to<br>
<a href="mailto:openstack-operators@lists.openstack.org">openstack-operators@lists.openstack.org</a><br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators</a><br>
<br>
or, via email, send a message with subject or body 'help' to<br>
<a href="mailto:openstack-operators-request@lists.openstack.org">openstack-operators-request@lists.openstack.org</a><br>
<br>
You can reach the person managing the list at<br>
<a href="mailto:openstack-operators-owner@lists.openstack.org">openstack-operators-owner@lists.openstack.org</a><br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than "Re: Contents of OpenStack-operators digest..."<br>
<br>
<br>
Today's Topics:<br>
<br>
1. How can I select host node to create VM? (Taeheum Na)<br>
2. Re: I have finally created an instance, and it works!<br>
However, there is no ethernet card (Andreas Scheuring)<br>
3. Operations summit (Sean Roberts)<br>
<br>
<br>
----------------------------------------------------------------------<br>
<br>
Message: 1<br>
Date: Mon, 18 Aug 2014 12:55:58 +0900<br>
From: Taeheum Na <<a href="mailto:thna@nm.gist.ac.kr">thna@nm.gist.ac.kr</a>><br>
To: "<a href="mailto:openstack-operators@lists.openstack.org">openstack-operators@lists.openstack.org</a>"<br>
<<a href="mailto:openstack-operators@lists.openstack.org">openstack-operators@lists.openstack.org</a>><br>
Subject: [Openstack-operators] How can I select host node to create<br>
VM?<br>
Message-ID:<br>
<<a href="mailto:0145D5DF1C56174FABB6EB834BC4909C03FEBD1915AC@NML2007.netmedia.kjist.ac.kr">0145D5DF1C56174FABB6EB834BC4909C03FEBD1915AC@NML2007.netmedia.kjist.ac.kr</a>><br>
<br>
Content-Type: text/plain; charset="us-ascii"<br>
<br>
Hello,<br>
When I create instance, I want to choice host node.<br>
I saw that by default, VMs are scheduled by filter scheduler which only consider computing resource.<br>
I want to apply substrate network considered scheduling algorithm on OpenStack.<br>
To do this one, I have to monitor utilization of computing/network resources.<br>
Now, I'm planning to use OpenvSwitch commend to know network situation.<br>
<br>
Do you have any comment for me? (monitor/scheduling)<br>
<br>
Regards<br>
Taeheum Na<br>
****************************************************<br>
M.S. candidate of Networked Computing Systems Lab.<br>
School of Information and Communications<br>
GIST (Gwangju Inst. of Sci. and Tech.)<br>
E-mail: <a href="mailto:thna@nm.gist.ac.kr">thna@nm.gist.ac.kr</a><mailto:<a href="mailto:thna@nm.gist.ac.kr">thna@nm.gist.ac.kr</a>><br>
Phone: <a href="tel:%2B82-10-2238-9424" value="+821022389424">+82-10-2238-9424</a><br>
Office: <a href="tel:%2B82-62-715-2273" value="+82627152273">+82-62-715-2273</a><br>
<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <<a href="http://lists.openstack.org/pipermail/openstack-operators/attachments/20140818/470f4ef4/attachment-0001.html" target="_blank">http://lists.openstack.org/pipermail/openstack-operators/attachments/20140818/470f4ef4/attachment-0001.html</a>><br>
<br>
------------------------------<br>
<br>
Message: 2<br>
Date: Mon, 18 Aug 2014 09:04:26 +0200<br>
From: Andreas Scheuring <<a href="mailto:scheuran@linux.vnet.ibm.com">scheuran@linux.vnet.ibm.com</a>><br>
To: "<a href="mailto:openstack-operators@lists.openstack.org">openstack-operators@lists.openstack.org</a>"<br>
<<a href="mailto:openstack-operators@lists.openstack.org">openstack-operators@lists.openstack.org</a>><br>
Subject: Re: [Openstack-operators] I have finally created an instance,<br>
and it works! However, there is no ethernet card<br>
Message-ID: <<a href="mailto:1408345466.4188.6.camel@oc5515017671.ibm.com">1408345466.4188.6.camel@oc5515017671.ibm.com</a>><br>
Content-Type: text/plain; charset="UTF-8"<br>
<br>
Assuming you're running a libvirt based hypervisor (e.g. kvm). Could you<br>
please dump the libvirt xml of your instance?<br>
<br>
You can get it this way<br>
<br>
virsh list --all<br>
--> shows a list of all virtual servers running on your hypervisor<br>
<br>
virsh dumpxml <vm-id><br>
--> dumps the xml of your vm, addressed by the id (first column). Id<br>
does not correlate with UUID!!!!! So if you're not sure which list entry<br>
belongs to your instance just stop all other via openstack to have only<br>
one running.<br>
<br>
PS: on some systems you have to sudo when using virsh.<br>
<br>
<br>
There should be a subtag of devices called <interface type='bridge'> or<br>
'network' or somehting like that representing your eth interface<br>
<br>
<br>
Andreas<br>
<br>
<br>
On Fri, 2014-08-15 at 11:43 -0700, Jeff Silverman wrote:<br>
> I have been surfing the internet, and one of the ideas that comes to<br>
> mind is modifying the /etc/neutron/agent.ini file on the compute<br>
> nodes. In the agent.ini file, there is a comment near the top that is<br>
> almost helpful:<br>
><br>
><br>
> # L3 requires that an interface driver be set. Choose the one that<br>
> best<br>
> # matches your plugin.<br>
><br>
><br>
> The only plug I know about is ml2. I have no idea if that is right<br>
> for me or not. And I have no idea to choose the interface drive that<br>
> best matches my plugin.<br>
><br>
><br>
> Thank you<br>
><br>
><br>
><br>
><br>
> Jeff<br>
><br>
><br>
><br>
><br>
><br>
><br>
><br>
><br>
><br>
><br>
> On Fri, Aug 15, 2014 at 10:26 AM, Jeff Silverman <<a href="mailto:jeff@sweetlabs.com">jeff@sweetlabs.com</a>><br>
> wrote:<br>
> By "defined a network space for your instances", does that<br>
> mean going through the process as described<br>
> in <a href="http://docs.openstack.org/icehouse/install-guide/install/yum/content/neutron-ml2-compute-node.html" target="_blank">http://docs.openstack.org/icehouse/install-guide/install/yum/content/neutron-ml2-compute-node.html</a>?<br>
><br>
><br>
> I got part way through that when I realized that the procedure<br>
> was going to bridge packets through neutron. That's not what<br>
> I want. I want the packets to go directly to the physical<br>
> router. For example, I have two tenants, with IP addresses<br>
> <a href="http://10.50.15.80/24" target="_blank">10.50.15.80/24</a> and 10.50.18.15.90/24.and the router is at<br>
> 10.50.15.1. There is a nice picture of what I am trying to do<br>
> at <a href="http://docs.openstack.org/openstack-ops/content/network_troubleshooting.html#nova_network_traffic_in_cloud" target="_blank">http://docs.openstack.org/openstack-ops/content/network_troubleshooting.html#nova_network_traffic_in_cloud</a> . But if the hypervisor doesn't present a virtual device to the guests, then nothing else is going to happen. The network troubleshooting guide <a href="http://docs.openstack.org/openstack-ops/content/network_troubleshooting.html#nova_network_traffic_in_cloud" target="_blank">http://docs.openstack.org/openstack-ops/content/network_troubleshooting.html#nova_network_traffic_in_cloud</a> does not explain what to do if the virtual NIC is missing.<br>
><br>
><br>
><br>
><br>
> Thank you<br>
><br>
><br>
> Jeff<br>
><br>
><br>
><br>
><br>
> On Fri, Aug 15, 2014 at 9:38 AM, Abel Lopez<br>
> <<a href="mailto:alopgeek@gmail.com">alopgeek@gmail.com</a>> wrote:<br>
> Curious if you?ve defined a network space for your<br>
> instances. If you?re using the traditional<br>
> flat_network, this is known as the ?fixed_address?<br>
> space.<br>
> If you?re using neutron, you would need to create a<br>
> network and a subnet (and router with gateway, etc).<br>
> You?d then assign the instance to a network at launch<br>
> time.<br>
><br>
><br>
><br>
> On Aug 15, 2014, at 9:17 AM, Jeff Silverman<br>
> <<a href="mailto:jeff@sweetlabs.com">jeff@sweetlabs.com</a>> wrote:<br>
><br>
> > <ip_a.png><br>
> ><br>
> > ?<br>
> ><br>
> > For those of you that can't see pictures:<br>
> > $ sudo ip a<br>
> > 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 qdisc<br>
> > noqueue<br>
> > link/loopback 00:00:00:00:00:00 brd<br>
> > 00:00:00:00:00:00<br>
> > inet <a href="http://127.0.0.1/8" target="_blank">127.0.0.1/8</a> scope host lo<br>
> > inet6 ::1/128 scope host<br>
> > valid_lft forever preferred_1ft forever<br>
> ><br>
> ><br>
> > I suspect that the issue is that the hypervisor is<br>
> > not presenting a virtual ethernet card.<br>
> ><br>
> ><br>
> > Thank you<br>
> ><br>
> ><br>
> ><br>
> ><br>
> > Jeff<br>
> ><br>
> ><br>
> ><br>
> ><br>
> > On Thu, Aug 14, 2014 at 6:57 PM, Nhan Cao<br>
> > <<a href="mailto:nhanct92@gmail.com">nhanct92@gmail.com</a>> wrote:<br>
> > can you show output of command:<br>
> > ip a<br>
> ><br>
> ><br>
> ><br>
> ><br>
> ><br>
> ><br>
> > 2014-08-15 7:41 GMT+07:00 Jeff Silverman<br>
> > <<a href="mailto:jeff@sweetlabs.com">jeff@sweetlabs.com</a>>:<br>
> > People,<br>
> ><br>
> ><br>
> > I have brought up an instance, and I<br>
> > can connect to it using my browser!<br>
> > I am so pleased.<br>
> ><br>
> ><br>
> > However, my instance doesn't have an<br>
> > ethernet device, only a loopback<br>
> > device. My management wants me to<br>
> > use a provider network, which I<br>
> > understand to mean that my instances<br>
> > will have IP addresses in the same<br>
> > space as the controller, block<br>
> > storage, and compute node<br>
> > administrative addresses. However,<br>
> > I think that discussing addressing<br>
> > is premature until I have a working<br>
> > virtual ethernet card.<br>
> ><br>
> ><br>
> > I am reading<br>
> > through <a href="http://docs.openstack.org/icehouse/install-guide/install/yum/content/neutron-ml2-compute-node.html" target="_blank">http://docs.openstack.org/icehouse/install-guide/install/yum/content/neutron-ml2-compute-node.html</a> and I think that the ML2 plugin is what I need. However, I think I do not want a network type of GRE, because that encapsulates the packets and I don't have anything to un-encapsulate them.<br>
> ><br>
> ><br>
> > Thank you<br>
> ><br>
> ><br>
> ><br>
> ><br>
> > Jeff<br>
> ><br>
> ><br>
> ><br>
> ><br>
> > --<br>
> > Jeff Silverman<br>
> > Systems Engineer<br>
> > <a href="tel:%28253%29%20459-2318" value="+12534592318">(253) 459-2318</a> (c)<br>
> ><br>
> ><br>
> ><br>
> ><br>
> > _______________________________________________<br>
> > OpenStack-operators mailing list<br>
> > <a href="mailto:OpenStack-operators@lists.openstack.org">OpenStack-operators@lists.openstack.org</a><br>
> > <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators</a><br>
> ><br>
> ><br>
> ><br>
> ><br>
> ><br>
> ><br>
> ><br>
> > --<br>
> > Jeff Silverman<br>
> > Systems Engineer<br>
> > <a href="tel:%28253%29%20459-2318" value="+12534592318">(253) 459-2318</a> (c)<br>
> ><br>
> ><br>
> > _______________________________________________<br>
> > OpenStack-operators mailing list<br>
> > <a href="mailto:OpenStack-operators@lists.openstack.org">OpenStack-operators@lists.openstack.org</a><br>
> > <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators</a><br>
> ><br>
><br>
><br>
><br>
><br>
><br>
><br>
> --<br>
> Jeff Silverman<br>
> Systems Engineer<br>
> <a href="tel:%28253%29%20459-2318" value="+12534592318">(253) 459-2318</a> (c)<br>
><br>
><br>
><br>
><br>
><br>
><br>
> --<br>
> Jeff Silverman<br>
> Systems Engineer<br>
> <a href="tel:%28253%29%20459-2318" value="+12534592318">(253) 459-2318</a> (c)<br>
><br>
><br>
> _______________________________________________<br>
> OpenStack-operators mailing list<br>
> <a href="mailto:OpenStack-operators@lists.openstack.org">OpenStack-operators@lists.openstack.org</a><br>
> <a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators</a><br>
<br>
<br>
<br>
<br>
<br>
------------------------------<br>
<br>
Message: 3<br>
Date: Mon, 18 Aug 2014 00:34:22 -0700<br>
From: Sean Roberts <<a href="mailto:seanroberts66@gmail.com">seanroberts66@gmail.com</a>><br>
To: <a href="mailto:openstack-operators@lists.openstack.org">openstack-operators@lists.openstack.org</a><br>
Subject: [Openstack-operators] Operations summit<br>
Message-ID: <<a href="mailto:FA2E9C27-6CB8-45D3-A21B-C2431F54ADBC@gmail.com">FA2E9C27-6CB8-45D3-A21B-C2431F54ADBC@gmail.com</a>><br>
Content-Type: text/plain; charset="us-ascii"<br>
<br>
I wanted to post a few items on the upcoming operator summit<br>
- I am going to moderate the Tuesday 10:30am Deploy/Config/Upgrade session. Any ideas on content is welcome.<br>
- I would like to add Congress / Policy to the Tuesday 1:30pm session along side Puppet, Chef, Salt, and Ansible. I think we are still missing someone to represent Salt.<br>
- I believe the Monday 10:30am Network session will be on the Nova.network to Neutron migration path. Any ideas on content is welcome.<br>
<br>
I am going to think some on the Deploy/Config/Upgrade session agenda and post it here for early discussion.<br>
<br>
~ sean<br>
<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <<a href="http://lists.openstack.org/pipermail/openstack-operators/attachments/20140818/3ca3c098/attachment-0001.html" target="_blank">http://lists.openstack.org/pipermail/openstack-operators/attachments/20140818/3ca3c098/attachment-0001.html</a>><br>
<br>
------------------------------<br>
<br>
_______________________________________________<br>
OpenStack-operators mailing list<br>
<a href="mailto:OpenStack-operators@lists.openstack.org">OpenStack-operators@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators</a><br>
<br>
<br>
End of OpenStack-operators Digest, Vol 46, Issue 20<br>
***************************************************<br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><div dir="ltr"><b>Jeff Silverman</b><div>Systems Engineer</div><div>(253) 459-2318 (c)</div><div><img src="https://dl.dropboxusercontent.com/u/16943296/SweetLabs-Signatures/New_2014/signature-logo.png"><br>
</div></div>
</div>