<div dir="ltr"><br><div class="gmail_extra"><br><div class="gmail_quote">On 22 October 2015 at 01:21, yujie <span dir="ltr"><<a href="mailto:judy_yujie@126.com" target="_blank">judy_yujie@126.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">I used ixgbe and vlan, passthrough a VF to vm.<br>
After the VM created, it could not connect to VM on the same compute node without use sriov.<div class=""><div class="h5"><br></div></div></blockquote><div><br></div><div>Not sure if this is the same conversation happening on Launchpad, but if not, this might be relevant: <br></div><div><br></div><div><a href="https://bugs.launchpad.net/neutron/+bug/1506003">https://bugs.launchpad.net/neutron/+bug/1506003</a><br></div><div><br></div><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 class=""><div class="h5">
<br>
在 2015/10/22 10:58, Alexander Duyck 写道:<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">
I assume by Intel cards you mean something that is running ixgbe? If so<br>
and you are trying to use SR-IOV with OVS and VLANs running on top of<br>
the PF it will fail. The issue is that OVS requires the ability to place<br>
the PF in promiscuous mode to support VLAN trunking, and ixgbe driver<br>
prevents that when SR-IOV is enabled.<br>
<br>
The "bridge fdb add" approach mentioned should work as long as ixgbe PF<br>
is used on a flat network.<br>
<br>
- Alex<br>
<br>
On 10/19/2015 07:33 PM, yujie 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">
Hi Moshe Levi,<br>
Sorry for replying to this message after so long time. The testing<br>
environment was unavailable before.<br>
I use Intel cards, but could only tested base kilo and vlan. Could<br>
it work?<br>
<br>
在 2015/9/22 13:24, Moshe Levi 写道:<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">
Hi Yujie,<br>
<br>
There is a patch <a href="https://review.openstack.org/#/c/198736/" rel="noreferrer" target="_blank">https://review.openstack.org/#/c/198736/</a> which I<br>
wrote to add the mac of the normal instance to<br>
the SR-IOV embedded switch so that the packet will go to the PF<br>
instead of going to the wire.<br>
This is done by using bridge tool with the command "bridge fdb add<br>
<mac> dev <interface>"<br>
<br>
I was able to test it on Mellanox ConnectX3 card with both vlan and<br>
flat network and it worked fine.<br>
I wasn't able to test it on any of the Intel cards, but I was told<br>
the it only working on flat network, in vlan network the Intel card<br>
is dropping the tagged packets and they are not go up to the VF.<br>
<br>
What NIC are you using? Can you try using "bridge fdb add <mac> dev<br>
<interface>" where <mac> is the mac of the normal vm and <interface><br>
is the PF<br>
and see if that resolve the issue.<br>
Also can you check it with flat and vlan networks.<br>
<br>
<br>
-----Original Message-----<br>
From: yujie [mailto:<a href="mailto:judy_yujie@126.com" target="_blank">judy_yujie@126.com</a>]<br>
Sent: Tuesday, September 22, 2015 6:28 AM<br>
To: <a href="mailto:openstack-dev@lists.openstack.org" target="_blank">openstack-dev@lists.openstack.org</a><br>
Subject: [openstack-dev] [neutron][sriov] SRIOV-VM could not work<br>
well with normal VM<br>
<br>
Hi all,<br>
I am using neutron kilo without dvr to create sriov instance VM-A,it<br>
works well and could connect to its gateway fine.<br>
But when I let the normal instance VM-B which in the same<br>
compute-node with VM-A ping its gateway, it failed. I capture the<br>
packet on the network-node, find the gateway already reply the<br>
ARP-reply message to VM-B. But compute-node which VM-B lives could<br>
not send the package to VM-B.<br>
If delete VM-A and set : echo 0 ><br>
/sys/class/enp5s0f0/device/sriov_numvfs, the problem solved.<br>
<br>
Is it a same question with the bug: SR-IOV port doesn't reach OVS<br>
port on same compute node ?<br>
<a href="https://bugs.launchpad.net/neutron/+bug/1492228" rel="noreferrer" target="_blank">https://bugs.launchpad.net/neutron/+bug/1492228</a><br>
Any suggestions will be grateful.<br>
<br>
Thanks,<br>
Yujie<br>
<br>
<br>
__________________________________________________________________________<br>
<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe:<br>
<a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br>
__________________________________________________________________________<br>
<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe:<br>
<a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br>
</blockquote>
<br>
<br>
__________________________________________________________________________<br>
<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe:<br>
<a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
</blockquote>
<br>
<br>
__________________________________________________________________________<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
</blockquote>
<br>
<br>
__________________________________________________________________________<br>
OpenStack Development Mailing List (not for usage questions)<br>
Unsubscribe: <a href="http://OpenStack-dev-request@lists.openstack.org?subject:unsubscribe" rel="noreferrer" target="_blank">OpenStack-dev-request@lists.openstack.org?subject:unsubscribe</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" rel="noreferrer" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
</div></div></blockquote></div><br></div></div>