<div dir="ltr">I think can say... "YAY!!" :-D<div><br></div><div>With "LibvirtOpenVswitchDriver" my internal communication is the double now! It goes from ~200 (with LibvirtHybridOVSBridgeDriver) to <b><u>400Mbit/s</u></b> (with LibvirtOpenVswitchDriver)! Still far from 1Gbit/s (my physical path limit) but, more acceptable now.</div>
<div><br></div><div>The command "ethtool -K eth1 gro off" still makes no difference.</div><div><br></div><div>So, there is only 1 remain problem, when traffic pass trough L3 / Namespace, it is still useless. Even the SSH connection into my Instances, via its Floating IPs, is slow as hell, sometimes it just stops responding for a few seconds, and becomes online again "out-of-nothing"...</div>
<div><br></div><div>I just detect a weird "behavior", when I run "apt-get update" from instance-1, it is slow as I said plus, its ssh connection (where I'm running apt-get update), stops responding right after I run "apt-get update" AND, <u>all my others ssh connections also stops working too!</u> For a few seconds... This means that when I run "apt-get update" from within instance-1, the SSH session of instance-2 is affected too!! There is something pretty bad going on at L3 / Namespace.</div>
<div><br></div><div>BTW, do you think that a ~400MBit/sec intra-vm-communication (GRE tunnel) on top of a 1Gbit ethernet is acceptable?! It is still less than a half...</div><div class="gmail_extra"><br></div><div class="gmail_extra">
Thank you!</div><div class="gmail_extra">Thiago<br><br><div class="gmail_quote">On 25 October 2013 12:28, Darragh O'Reilly <span dir="ltr"><<a href="mailto:dara2002-openstack@yahoo.com" target="_blank">dara2002-openstack@yahoo.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div><div style="font-size:12pt;font-family:HelveticaNeue,'Helvetica Neue',Helvetica,Arial,'Lucida Grande',sans-serif">
<div>Hi Thiago,</div><div><br></div><div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica Neue',Helvetica,Arial,'Lucida Grande',sans-serif">for the VIF error: you will need to change qemu.conf as described here:</div>
<div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica Neue',Helvetica,Arial,'Lucida Grande',sans-serif"><a href="http://openvswitch.org/openstack/documentation/" target="_blank">http://openvswitch.org/openstack/documentation/</a></div>
<div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica Neue',Helvetica,Arial,'Lucida Grande',sans-serif"> <br></div><div style="font-style:normal;font-size:16px;background-color:transparent;font-family:HelveticaNeue,'Helvetica Neue',Helvetica,Arial,'Lucida Grande',sans-serif">
Re, Darragh.<br></div><div><div class="h5"><div style="display:block"> <br> <br> <div style="font-family:HelveticaNeue,'Helvetica Neue',Helvetica,Arial,'Lucida Grande',sans-serif;font-size:12pt"> <div style="font-family:HelveticaNeue,'Helvetica Neue',Helvetica,Arial,'Lucida Grande',sans-serif;font-size:12pt">
<div dir="ltr"> <font face="Arial"> On Friday, 25 October 2013, 15:14, Martinx - $B%8%'!<%`%:(B <<a href="mailto:thiagocmartinsc@gmail.com" target="_blank">thiagocmartinsc@gmail.com</a>> wrote:<br> </font> </div> <blockquote style="border-left-width:2px;border-left-style:solid;border-left-color:rgb(16,16,255);margin-left:5px;margin-top:5px;padding-left:5px">
<div><div><div><div dir="ltr">Hi Darragh,<div><br clear="none"></div><div>Yes, Instances are getting MTU 1400.</div><div><br clear="none"></div><div>I'm using LibvirtHybridOVSBridgeDriver at my Compute Nodes. I'll check BG 1223267 right now! </div>
<div><br clear="none">
</div><div><br clear="none"></div><div>The <span style="font-family:arial,sans-serif;font-size:12.800000190734863px">LibvirtOpenVswitchDriver doesn't work, look:</span></div><div><span style="font-family:arial,sans-serif;font-size:12.800000190734863px"><br clear="none">
</span></div><div><a rel="nofollow" shape="rect" href="http://paste.openstack.org/show/49709/" target="_blank">http://paste.openstack.org/show/49709/</a><span style="font-family:arial,sans-serif;font-size:12.800000190734863px"><br clear="none">
</span></div><div><br clear="none"></div><div>
<a rel="nofollow" shape="rect" href="http://paste.openstack.org/show/49710/" target="_blank">http://paste.openstack.org/show/49710/</a><br clear="none">
</div><div><span style="font-family:arial,sans-serif;font-size:12.800000190734863px"><br clear="none"></span></div><div><span style="font-family:arial,sans-serif;font-size:12.800000190734863px"><br clear="none"></span></div>
<div><span style="font-family:arial,sans-serif;font-size:12.800000190734863px">My NICs are "</span><font face="arial, sans-serif">RTL8111/8168/8411 PCI Express Gigabit Ethernet", Hypervisors motherboard are MSI-890FXA-GD70.</font></div>
<div><font face="arial, sans-serif"><br clear="none"></font></div><div><font face="arial, sans-serif">The command "ethtool -K eth1 gro off" did not had any effect on the communication between instances on different hypervisors, still poor, around 248Mbit/sec, when its physical path reach 1Gbit/s (where GRE is built).</font></div>
<div><font face="arial, sans-serif"><br clear="none"></font></div><div><font face="arial, sans-serif">My Linux version is "Linux hypervisor-1 3.8.0-32-generic #47~precise1-Ubuntu", same kernel on Network Node" and others nodes too (Ubuntu 12.04.3 installed from scratch for this Havana deployment).</font></div>
<div><font face="arial, sans-serif"><br clear="none"></font></div><div><font face="arial, sans-serif">The only difference I can see right now, between my two hypervisors, is that my second is just a spare machine, with a slow CPU but, I don't think it will have a negative impact at the network throughput, since I have only 1 Instance running into it (plus a qemu-nbd process eating 90% of its CPU). I'll replace this CPU tomorrow, to redo this tests again but, I don't think that this is the source of my problem. The MOBOs of two hypervisors are identical, 1 3Com (manageable) switch connecting the two.</font></div>
<div><font face="arial, sans-serif"><br clear="none"></font></div><div><font face="arial, sans-serif">Thanks!</font></div><div><font face="arial, sans-serif">Thiago</font></div></div><div><br clear="none"><br clear="none">
<div><div>
On 25 October 2013 07:15, Darragh O'Reilly <span dir="ltr"><<a rel="nofollow" shape="rect" href="mailto:dara2002-openstack@yahoo.com" target="_blank">dara2002-openstack@yahoo.com</a>></span> wrote:<br clear="none">
<blockquote style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
Hi Thiago,<br clear="none">
<br clear="none">
you have configured DHCP to push out a MTU of 1400. Can you confirm that the 1400 MTU is actually getting out to the instances by running 'ip link' on them?<br clear="none">
<br clear="none">
There is an open problem where the veth used to connect the OVS and Linux bridges causes a performance drop on some kernels - <a rel="nofollow" shape="rect" href="https://bugs.launchpad.net/nova-project/+bug/1223267" target="_blank">https://bugs.launchpad.net/nova-project/+bug/1223267</a> . If you are using the LibvirtHybridOVSBridgeDriver VIF driver, can you try changing to LibvirtOpenVswitchDriver and repeat the iperf test between instances on different compute-nodes.<br clear="none">
<br clear="none">
What NICs (maker+model) are you using? You could try disabling any off-load functionality - 'ethtool -k <iface-used-for-gre>'.<br clear="none">
<br clear="none">
What kernal are you using: 'uname -a'?<br clear="none">
<br clear="none">
Re, Darragh.<br clear="none">
<div><br clear="none">
> Hi Daniel,<br clear="none">
<br clear="none">
><br clear="none">
> I followed that page, my Instances MTU is lowered by DHCP Agent but, same<br clear="none">
> result: poor network performance (internal between Instances and when<br clear="none">
> trying to reach the Internet).<br clear="none">
><br clear="none">
> No matter if I use "dnsmasq_config_file=/etc/neutron/dnsmasq-neutron.conf +<br clear="none">
> "dhcp-option-force=26,1400"" for my Neutron DHCP agent, or not (i.e. MTU =<br clear="none">
> 1500), the result is almost the same.<br clear="none">
><br clear="none">
> I'll try VXLAN (or just VLANs) this weekend to see if I can get better<br clear="none">
> results...<br clear="none">
><br clear="none">
> Thanks!<br clear="none">
> Thiago<br clear="none">
<br clear="none">
</div><div><div>_______________________________________________<br clear="none">
Mailing list: <a rel="nofollow" shape="rect" href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack</a><br clear="none">
Post to : <a rel="nofollow" shape="rect" href="mailto:openstack@lists.openstack.org" target="_blank">openstack@lists.openstack.org</a><br clear="none">
Unsubscribe : <a rel="nofollow" shape="rect" href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack</a><br clear="none">
</div></div></blockquote></div><br clear="none"></div></div></div></div><br><br></div> </blockquote> </div> </div> </div> </div></div></div></div></blockquote></div><br></div></div>