<div dir="ltr"><div dir="ltr"><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Thu, Sep 24, 2020 at 10:14 AM Tony Liu <<a href="mailto:tonyliu0592@hotmail.com">tonyliu0592@hotmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Hi,<br>
<br>
I read through this long discussion [1].<br>
<br>
Here is what I am doing.<br>
<br>
+------------------------------------------+<br>
| external logical switch |<br>
+-+-------------+--------------------+-----+<br>
| | |<br>
+--+--+ +--+--+ +---+----+<br>
|dgp1 | |dgp2 | ... |dgp1000 |<br>
+--+--+ +--+--+ +---+----+<br>
| | |<br>
+-+-+ +-+-+ +---+---+<br>
|LR1| |LR2| |LR1000 |<br>
+---+ +---+ +-------+<br>
<br>
First of all, I see the same flow explosion in lr_in_arp_resolve<br>
table. I'd like to confirm the patch [2] will also avoid explosion<br>
in my case?<br>
<br></blockquote><div><br></div><div>I think so. Maybe Han or Dumitru can confirm. I suggest that you test it out yourself.</div><div>You can stop the neutron server and run a script which sets this option on each logical router.</div><div><br></div><div>something like</div><div><br></div><div>for i in $(ovn-nbctl --bare --columns __uuid list logical_router)</div><div>do</div><div> ovn-nbctl set logical_router $i options:always_learn_from_arp_request=false</div><div>done</div><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
In my case, LRs are not bound to any specific compute chassis.<br>
All DGPs are bound on the central set of gateway chassis.<br>
It's central SNAT and FIP.<br>
<br>
I am looking for the possibility to do distributed SNAT and FIP to<br>
avoid central gateway nodes. With OpenStack integration,<br>
distributed FIP is supported, but not distributed SNAT. because<br>
there is not chassis specific address can be used as the source<br>
IP for SNAT.<br>
<br></blockquote><div><br></div><div>I don't think OVN supports distributed SNAT.</div><div><br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
Given the idea in [3], DPG can be bound on compute chassis.<br>
I don't need the support to have multiple DPGs on one LR.<br>
Then is that going to work for distributed SNAT?<br>
Any details, like how to allocate chassis specific address<br>
as the source IP for SNAT, and how ARP works for that address?<br></blockquote><div><br></div><div>I am not sure how easy is it going to support this.</div><div><br></div><div>Thanks</div><div>Numan</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">
<br>
[1] <a href="https://www.mail-archive.com/ovs-discuss@openvswitch.org/msg06948.html" rel="noreferrer" target="_blank" class="md-opjjpmhoiojifppkkcdabiobhakljdgm_doc">https://www.mail-archive.com/ovs-discuss@openvswitch.org/msg06948.html</a><br>
[2] <a href="https://www.mail-archive.com/ovs-dev@openvswitch.org/msg45681.html" rel="noreferrer" target="_blank" class="md-opjjpmhoiojifppkkcdabiobhakljdgm_doc">https://www.mail-archive.com/ovs-dev@openvswitch.org/msg45681.html</a><br>
[3] <a href="https://www.mail-archive.com/ovs-discuss@openvswitch.org/msg06987.html" rel="noreferrer" target="_blank" class="md-opjjpmhoiojifppkkcdabiobhakljdgm_doc">https://www.mail-archive.com/ovs-discuss@openvswitch.org/msg06987.html</a><br>
<br>
Thanks!<br>
Tony<br>
<br>
_______________________________________________<br>
discuss mailing list<br>
<a href="mailto:discuss@openvswitch.org" target="_blank" class="md-opjjpmhoiojifppkkcdabiobhakljdgm_doc">discuss@openvswitch.org</a><br>
<a href="https://mail.openvswitch.org/mailman/listinfo/ovs-discuss" rel="noreferrer" target="_blank" class="md-opjjpmhoiojifppkkcdabiobhakljdgm_doc">https://mail.openvswitch.org/mailman/listinfo/ovs-discuss</a><br>
<br>
</blockquote></div></div>