[ovs-discuss] questions about recirculation

2016-05-20 Thread hanxueluo
why use recirculation action for TCP-mode bond? what's the benefits, compared to the old implement? Is there any other situation using recirculation action?___ discuss mailing list discuss@openvswitch.org http://openvswitch.org/mailman/listinfo/discuss

Re: [ovs-discuss] ovn_l3: should vm get replies from pings to router interface (of non-local subnet)?

2016-05-20 Thread Flaviof
Hi Justin (et all!), As we talked in the last ovn meeting, I took a look at the code that populates the logical rules for doing the icmp response. It is here [1], in ovn-northd.c, function build_lrouter_flows(). Thanks Russell for the valuable pointers! Unlike ARP, I come to the thinking that les

Re: [ovs-discuss] Instrumenting OVN

2016-05-20 Thread Ryan Moats
Ben Pfaff wrote on 05/20/2016 12:03:29 PM: > From: Ben Pfaff > To: Ryan Moats/Omaha/IBM@IBMUS > Cc: discuss > Date: 05/20/2016 12:03 PM > Subject: Re: [ovs-discuss] Instrumenting OVN > > On Fri, May 20, 2016 at 10:50:31AM -0500, Ryan Moats wrote: > > However, both of the above are currently opt

Re: [ovs-discuss] In ovs-userconntrack_20151115 Branch - ICMP Blocked port can be hacked, if same icmp request id is used while sending the packet from the blocked side of the firewall.

2016-05-20 Thread Daniele Di Proietto
On 20/05/2016 10:35, "Daniele Di Proietto" wrote: >Hi Soumyadeep, > > > >On 18/05/2016 22:58, "soumyadeep.chowdh...@wipro.com" > wrote: > >>Hi Daniele, >> >>Its much cleaner approach. >>Thanks for the update. > >Thanks again for reporting this > >> >>One more question: >>Is your latest user

Re: [ovs-discuss] In ovs-userconntrack_20151115 Branch - ICMP Blocked port can be hacked, if same icmp request id is used while sending the packet from the blocked side of the firewall.

2016-05-20 Thread Daniele Di Proietto
Hi Soumyadeep, On 18/05/2016 22:58, "soumyadeep.chowdh...@wipro.com" wrote: >Hi Daniele, > >Its much cleaner approach. >Thanks for the update. Thanks again for reporting this > >One more question: >Is your latest userconntrack branch merged with ovs-v2.5.0 already ? OVS 2.5.0 has already b

Re: [ovs-discuss] Instrumenting OVN

2016-05-20 Thread Ben Pfaff
On Fri, May 20, 2016 at 10:50:31AM -0500, Ryan Moats wrote: > However, both of the above are currently optional things that *could* > be added by hand. Because of our CI/CD requirements, we'll need to feed > it with fully decorated files. Maintaining a local branch manually > in this scenario is no

Re: [ovs-discuss] Collecting NetFlow data

2016-05-20 Thread Ben Pfaff
Yes, that's how UDP works. On Fri, May 20, 2016 at 05:40:34PM +0200, Syed Faraz Ali wrote: > The connection is established even when the other machine is offline. Is it > still normal? > > # netstat -an | grep 2055 > udp0 0 192.168.56.104:49403192.168.56.103:2055 > ESTABLISHED >

[ovs-discuss] adding path processing and table constraints

2016-05-20 Thread oualid hjira
Hello, i want to configure the path processing and the table search constraints(pipelining and table entries and search type) from the code of the ovs, i want to access the code that make these configurations thanks Hjira Oualid ___ discuss mailing list

Re: [ovs-discuss] Issue with vlan tag being changed after reboot of KVM hypervisor.

2016-05-20 Thread Madko
If you don't need the port-group abstraction (and the switch definition in libvirt), just create your ovs switch (ovsbr0) with ovs-vsctl and use this in the VM definition: I don't find a lot of documentation about this easier solution. Hope it helps. Le ven. 20 mai

Re: [ovs-discuss] Instrumenting OVN

2016-05-20 Thread Ryan Moats
Ben Pfaff wrote on 05/20/2016 12:13:46 AM: > From: Ben Pfaff > To: Ryan Moats/Omaha/IBM@IBMUS > Cc: discuss > Date: 05/20/2016 12:14 AM > Subject: Re: [ovs-discuss] Instrumenting OVN > > On Thu, May 19, 2016 at 05:51:19PM -0500, Ryan Moats wrote: > > > > As I mentioned at today's IRC meeting -

Re: [ovs-discuss] Collecting NetFlow data

2016-05-20 Thread Syed Faraz Ali
The connection is established even when the other machine is offline. Is it still normal? # netstat -an | grep 2055 udp0 0 192.168.56.104:49403192.168.56.103:2055 ESTABLISHED # ping 192.168.56.103 PING 192.168.56.103 (192.168.56.103) 56(84) bytes of data. >From 192.168.56.104 icm

Re: [ovs-discuss] Collecting NetFlow data

2016-05-20 Thread Ben Pfaff
It seems OK. When I use a similar configuration, I see an appropriate line in "netstat -an" output: udp0 0 192.168.42.143:52897192.168.56.103:2055 ESTABLISHED Do you get anything like that? On Fri, May 20, 2016 at 05:11:07PM +0200, Syed Faraz Ali wrote: > >OVS isn't even op

[ovs-discuss] Flow analytics with Mininet

2016-05-20 Thread Peter Phaal
Enabling sFlow using ovs-vsctl can be tricky. The following article describes simple Mininet --custom sflow.py script that automatically enables sFlow on all switches and exports the metadata needed to relate the sFlow measurements to the topology: http://blog.sflow.com/2016/05/mininet-flow-ana

Re: [ovs-discuss] Collecting NetFlow data

2016-05-20 Thread Ben Pfaff
On Fri, May 20, 2016 at 12:50:13PM +0200, Syed Faraz Ali wrote: > > It's probably easier to debug if you only enable NetFlow for now. > > Only NetFlow has been enabled. > > > Next, I'd use "strace" to see whether ovs-vswitchd is really trying to > > send data to the collector. It would be using

Re: [ovs-discuss] Not hitting flow rules based on udp ports

2016-05-20 Thread Matthew . Stevens
Cancel that, using nping to create the udp packets works which blows this theory out the water. Cheers > Hi, I suspect this will become a bug report. > > I wish to stop all traffic from a specific udp port. I can do so with tcp, > but not with udp. > > The setup is VirtualBox 5, running Ubuntu 14