Ben, 
 
Here is the output of the physical interface
 
eth1      Link encap:Ethernet  HWaddr 00:0e:c6:88:d2:22
          inet6 addr: fe80::20e:c6ff:fe88:d222/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:1538 errors:0 dropped:12 overruns:0 frame:0
          TX packets:63290 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000
          RX bytes:270154 (270.1 KB)  TX bytes:2966862 (2.9 MB)

and you can see there are no dropped count in there!
 
sam
 
> Date: Fri, 19 Sep 2014 08:33:27 -0700
> From: b...@nicira.com
> To: sam.anda...@hotmail.com
> CC: disc...@openvswitch.org; dev@openvswitch.org
> Subject: Re: [ovs-dev] debugging OVS
> 
> Can you show us the "ovs-dpctl show" output?
> 
> Thanks,
> 
> Ben.
> 
> On Fri, Sep 19, 2014 at 11:30:05AM -0400, samantha Andares wrote:
> > hi ben, 
> >  
> > thanks for your feedback.
> >  
> > I am getting 100% packets drop.. If problem was with the amount of 
> > messages... I presume some would have been going through... as stated in my 
> > latest email.... this is specific to when going out on a USB to Ethernet 
> > adaptor on a specific hardware (beagle bone board). I am using OVS 2.0 and 
> > Ubuntu 14.04. How can we analyse the messages in userspace and why would 
> > the message go to userspace... shouldn't it stay in kernel space if it 
> > matches a flow and just be forwarded to the outgoing physical port as per 
> > the rule is saying? My traffic is simple iperf udp messages.. tried flow of 
> > 10MB and 100KB... problem is the same...
> >  
> > thanks, 
> >  
> > sam
> >  
> > > Date: Fri, 19 Sep 2014 08:20:24 -0700
> > > From: b...@nicira.com
> > > To: sam.anda...@hotmail.com
> > > CC: disc...@openvswitch.org; dev@openvswitch.org
> > > Subject: Re: [ovs-dev] debugging OVS
> > > 
> > > [dropping mininet-discuss because it is a closed mailing list and I am
> > > not a subscriber]
> > > 
> > > On Thu, Sep 18, 2014 at 05:37:43PM -0400, samantha Andares wrote:
> > > > I am getting packets dropped when going out of an OVS bridge. The TX
> > > > dropped is increased in ovs-dpctl but not at the physical interface
> > > > level (ifconfig).
> > > >
> > > > Can anyone tell me how to enable some OVS debugging so I can find the
> > > > reason to fix that problem?
> > > 
> > > Do you mean that ovs-dpctl shows the datapath's "drop" count is going
> > > up?  If so, the problem is that packets that need attention from
> > > userspace are arriving faster than userspace can process them.  The best
> > > way to reduce drops is to upgrade to the latest version of Open vSwitch,
> > > which is 2.3, because we've spent a lot of time the last few years
> > > coming up with ways to reduce the number of packets that have to be sent
> > > to userspace.  If you are already using 2.3, then I am surprised that
> > > you are seeing lots of drops.  Is there anything unusual about your
> > > environment or the traffic that you are using?
> >                                       
                                          
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev

Reply via email to