<p dir="ltr">Read the names carefully again :) <br>
I was suggesting what I used to do in the past when on a new OpenStack install I had this problem.</p>
<div class="gmail_quote">El 16/12/2014 20:35, "Rick Jones" <<a href="mailto:rick.jones2@hp.com">rick.jones2@hp.com</a>> escribió:<br type="attribution"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On 12/16/2014 11:28 AM, Adrián Norte Fernández wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I use Neutron so..<br>
</blockquote>
<br>
I thought you said you were using Nova legacy networking:<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
In my OpenStack installation (Icehouse and use nova legacy networking)<br>
</blockquote>
<br>
In any event what I am suggesting is you install either netperf or iperf into both of the VMs and run a "pure" networking benchmark between them.  That way you do not have to worry about anything like the 4GB file transfer filling the destination VM's file cache and then getting stuck behind slow flushing out to storage, or any concerns with the encryption overhead in scp etc. etc.<br>
<br>
If you install netperf into the two VMs, I can talk you through the process of how to run some of the more "interesting" tests.<br>
<br>
rick<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
El 16/12/2014 20:27, "Rick Jones" <<a href="mailto:rick.jones2@hp.com" target="_blank">rick.jones2@hp.com</a><br>
<mailto:<a href="mailto:rick.jones2@hp.com" target="_blank">rick.jones2@hp.com</a>>> escribió:<br>
<br>
    On 12/16/2014 11:09 AM, Georgios Dimitrakakis wrote:<br>
<br>
        Changing<br>
<br>
        gso on<br>
        tso on<br>
        gro off<br>
<br>
<br>
        got me back to the initial status.<br>
<br>
<br>
        Although now it starts with approximately 65-70MB/s  for a few<br>
        seconds<br>
        but then it drops down to 30MB/s<br>
<br>
<br>
    What do you see if you use a "pure" networking benchmark such as<br>
    netperf or iperf?<br>
<br>
    rick jones<br>
<br>
    ______________________________<u></u>___________________<br>
    Mailing list:<br>
    <a href="http://lists.openstack.org/__cgi-bin/mailman/listinfo/__openstack" target="_blank">http://lists.openstack.org/__<u></u>cgi-bin/mailman/listinfo/__<u></u>openstack</a><br>
    <<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack" target="_blank">http://lists.openstack.org/<u></u>cgi-bin/mailman/listinfo/<u></u>openstack</a>><br>
    Post to     : <a href="mailto:openstack@lists.openstack.org" target="_blank">openstack@lists.openstack.org</a><br>
    <mailto:<a href="mailto:openstack@lists.openstack.org" target="_blank">openstack@lists.<u></u>openstack.org</a>><br>
    Unsubscribe :<br>
    <a href="http://lists.openstack.org/__cgi-bin/mailman/listinfo/__openstack" target="_blank">http://lists.openstack.org/__<u></u>cgi-bin/mailman/listinfo/__<u></u>openstack</a><br>
    <<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack" target="_blank">http://lists.openstack.org/<u></u>cgi-bin/mailman/listinfo/<u></u>openstack</a>><br>
<br>
</blockquote>
<br>
</blockquote></div>