On 4/29/14 3:58 PM, Ben Pfaff wrote:
On Tue, Apr 29, 2014 at 03:51:45PM -0400, Mark Haywood wrote:
On 4/28/14 11:53 PM, Ben Pfaff wrote:
On Mon, Apr 28, 2014 at 07:01:02PM -0400, Mark Haywood wrote:
On 4/28/14, 6:36 PM, Ben Pfaff wrote:
On Mon, Apr 28, 2014 at 03:06:50PM -0400, Mark Haywood w
I'm not sure that I exactly understand your test setup. Performance is
fine if you go from the hypervisor but is 1.5G (85% reduction on 10G)
if it is between VMs and tunneling is used in both cases?
It sounds like the lack of offloading could be a problem (I believe
those NICs support VXLAN offloa
On Wed, Apr 30, 2014 at 06:37:27PM +0200, Christian Stigen Larsen wrote:
> Ben Pfaff :
> > It looks like branch-2.1 needs a commit that only made it onto master.
> > The log message didn't make it clear that it was a bug fix.
> >
> > Please try applying commit cfc50ae514f805dcd9c14589f21158185424d
On Tue, Apr 29, 2014 at 11:51:12AM -0500, navaz wrote:
> I would like to limit the flow entries in the OVS. ( For ex: say 100 flow
> entries) .
ovs-vsctl -- --id=@ft create Flow_Table flow_limit=100 overflow_policy=refuse
-- set Bridge br0 flow_tables=0=@ft
__
Ben Pfaff :
> It looks like branch-2.1 needs a commit that only made it onto master.
> The log message didn't make it clear that it was a bug fix.
>
> Please try applying commit cfc50ae514f805dcd9c14589f21158185424daf6 to
> your local copy of OVS 2.1.
I cherry-picked it onto origin/branch-2.1 (ov
Justin Pettit :
>> 2.2 is definitely fine -- I had only tested 2.0 and 2.1, and the patch
>> doesn't
>> apply on 2.1. But you do point-releases on 2.1?
>
> We do bug releases even on non-LTS versions. In fact, we just released
> 2.1.1 yesterday, but I don't know that it would address your issu
On Wed, Apr 30, 2014 at 04:45:00PM +0200, Christian Stigen Larsen wrote:
>
> Ben Pfaff :
> >> RTTs were up from an average of 87 ms to 541 ms.
> >
> > Hmm. I thought about this for a minute longer.
> >
> > Can you try applying this patch and see if it makes any difference?
>
> Ah, I hadn't tes
On April 30, 2014 at 7:45:32 AM, Christian Stigen Larsen (cslar...@gmail.com)
wrote:
>
> 2.2 is definitely fine -- I had only tested 2.0 and 2.1, and the patch doesn't
> apply on 2.1. But you do point-releases on 2.1?
We do bug releases even on non-LTS versions. In fact, we just released 2.1.1
On Wed, Apr 30, 2014 at 04:49:49PM +0200, Christian Stigen Larsen wrote:
>
> Ben Pfaff :
> >> RTTs were up from an average of 87 ms to 541 ms.
> >
> > An RTT of 87 ms is very high to begin with. What does mininet do that
> > could cause such a high RTT?
>
> Fair point. Honestly, I don't know,
Hi, Eugene,
I am using KVM. I set the vlan tag on the OVS bridge by
"ovs-vsctl set port vnet1 tag=1 # vnet1 is the vport for a VM"
So inside the VM, you actually can not see the vlan tag by tcpdump. The
vlan tag is inserted by the OVS in the host kernel.
Thanks for pointing to the patches. I wi
Ben Pfaff :
>> RTTs were up from an average of 87 ms to 541 ms.
>
> An RTT of 87 ms is very high to begin with. What does mininet do that
> could cause such a high RTT?
Fair point. Honestly, I don't know, but perhaps I can ask on mininet-discuss.
I thought it was simply due to the Python cont
Ben Pfaff :
>> RTTs were up from an average of 87 ms to 541 ms.
>
> Hmm. I thought about this for a minute longer.
>
> Can you try applying this patch and see if it makes any difference?
Ah, I hadn't tested 2.2! (Isn't that in-development, as in "don't rely on it"?)
2.2 is definitely fine --
On Wed, Apr 30, 2014 at 01:53:39PM +0200, Christian Stigen Larsen wrote:
> I'm using Open vSwitch with Mininet, and noticed a large increase in ICMP
> ping RTTs when using POX OpenFlow controllers _without_ flow entries:
>
> RTTs were up from an average of 87 ms to 541 ms.
Hmm. I thought about t
On Wed, Apr 30, 2014 at 01:53:39PM +0200, Christian Stigen Larsen wrote:
> I'm using Open vSwitch with Mininet, and noticed a large increase in ICMP
> ping RTTs when using POX OpenFlow controllers _without_ flow entries:
>
> RTTs were up from an average of 87 ms to 541 ms.
An RTT of 87 ms is very
Hi,
I'm using Open vSwitch with Mininet, and noticed a large increase in ICMP
ping RTTs when using POX OpenFlow controllers _without_ flow entries:
RTTs were up from an average of 87 ms to 541 ms.
I did a git bisect on origin/branch-2.1 and found that this was introduced
in e79a6c833e0d72370951d
15 matches
Mail list logo
