<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Hi, <div><br></div><div>Thanks for the reply.</div><div><br></div><div>Some comments are inlined. <br><div><div>On Oct 12, 2012, at 5:55 AM, Matt Joyce wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word"><div>- Do I need to have nova-api on each compute node? I guess yes. <br>
<br></div></div></blockquote><div> </div><div>You might want it on another node for HA but it's not at all necessary. <br></div></div></blockquote><div><br></div>Ok clear. <br><blockquote type="cite"><div class="gmail_quote"><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div style="word-wrap:break-word"><div></div><div>- What is good practice to run compute node? I mean do I need nova-volume, nova-cert and nova-scheduler on each compute node? </div></div></blockquote><div><br>You only need nova-compute for ec2 usage on each node. nova-volume is needed if you intend to create volumes on those nodes. <br>
<br><a href="https://github.com/cloudscaling/tarkin">https://github.com/cloudscaling/tarkin</a> has some simple tests.<br><br>Tempest I hear can do more?<br></div></div></blockquote><div><br></div>Thanks I will use those.<br><blockquote type="cite"><div class="gmail_quote"><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div style="word-wrap:break-word"><div><br></div><div>- Do I need to create a bridge on the new compute node, when I don't want to run nova-network on that node? I think bridge is only required by the nova-network? But than I get confused with how the network will work? I mean all the traffic of the VMs will go from the node running nova-network? I guess this is not correct right? </div>
<div><br></div></div></blockquote><div><br>Depends on how your networking is set up. But usually a bridge is created for each virtual interface of your instances. nova-compute will do that for you.<br></div></div></blockquote><div><br></div><div>So with this statement should I understand that the bridge is also required/used by nova-compute… right?</div><div>In this case I thing the traffic will not route through the node that runs the nova-network…. </div><div><br></div><blockquote type="cite"><div class="gmail_quote"><div> </div></div></blockquote><blockquote type="cite"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0.8ex; border-left-width: 1px; border-left-color: rgb(204, 204, 204); border-left-style: solid; padding-left: 1ex; position: static; z-index: auto; ">
<div style="word-wrap:break-word"><div></div><div>If all the traffic from VMs will route through the node running nova-network then should I run nova-network on each compute node as well? Or what is the correct way? </div>
<div><br></div></div></blockquote><div style="word-wrap:break-word"><br>depends on your networking configuration. </div></div></blockquote><div><br></div><div><div>Can you please explain it a bit. I mean in a very simple scenario, </div><div><br></div><div>node-A, controller node running nova-api, nova-network, nova-scheduler, nova-cert, nova-cert, nova-objectstore, nova-volume </div><div><br></div><div>node-B, only nova-compute </div><div><br></div><div>Will the traffic of the booted instances on node-B will route from node-A or not.</div></div><div><br></div><div><br></div><div><br></div><br><blockquote type="cite"><div class="gmail_quote"><div style="word-wrap:break-word"><div>Thanks in advance. </div><br></div></div>
</blockquote></div><br></div><div><br></div><div><br></div></body></html>