M, Jing Ai wrote:
>
> > Date: Mon, 27 May 2013 09:10:20 -0700
> > From: Justin Pettit
> > Subject: Re: [ovs-discuss] HIgh CPU usage for ovs-vswitchd with flows
> > 3069and lost: 267425491
> > To: kevin parker
> > Cc: discuss@openvswitch.org
> &
; > Subject: Re: [ovs-discuss] HIgh CPU usage for ovs-vswitchd with flows
> > 3069 and lost: 267425491
> > To: kevin parker
> > Cc: discuss@openvswitch.org
> > Message-ID:
> > Content-Type: text/plain; charset=iso-8859-1
> >
> > We've made a lot of
On May 28, 2013, at 10:16 AM, Jing Ai wrote:
> Good to know there are ongoing efforts to improve flow setup. What are the
> timelines for those features? Any patch sent out for review? Thanks!
Patches for kernel wildcarding should go out for review within the next two
weeks. The multithreadin
> Date: Mon, 27 May 2013 09:10:20 -0700
> From: Justin Pettit
> Subject: Re: [ovs-discuss] HIgh CPU usage for ovs-vswitchd with flows
> 3069and lost: 267425491
> To: kevin parker
> Cc: discuss@openvswitch.org
> Message-ID:
> Content-Type: text/plain; chars
The working set of flows is going to be dependent on your environment. There
are lots of ways to generate traffic, but you should be tuning OVS for your
workload.
I'm not clear on your interest in ovs-controller. It's really just meant to
test OVS's ability to be remotely configured--it's jus
Thank you Justin,
Could you please tell me how can i regenerate these
many number of flows for testing so that i can configure threshold based on
that,when i tried iperf with 1B and udp flow didn't increase and there were
no packet loss.Is ovs-controller generate flow based
I'm guessing the traffic is bursty or ovs-vswitchd was busing doing other work
and the queues overflowed.
--Justin
On May 27, 2013, at 9:34 AM, kevin parker wrote:
> Justin,
> Even at 85-90% CPU usage i am seeing increased lost count,initially
> i faced high lost count when ovs-vswi
Justin,
Even at 85-90% CPU usage i am seeing increased lost
count,initially i faced high lost count when ovs-vswitchd was at 100%.So
what can be the reason for dropped packet even with 10% free cpu available.
regards,
kevin
On Mon, May 27, 2013 at 9:40 PM, Justin Pettit wrote:
> We've
We've made a lot of improvements in flow set up rate since version 1.4, so
upgrading to a more current version (we're on 1.10 now) will likely help.
We're currently working on multithreading the OVS userspace and adding support
for wildcarded flows in the kernel, which should substantially impr
Hi,
Running OVS 1.4 on xenserver 6.0.2 , but its taking very high cpu some
times ~100%.
*ovs-dpctl show*
*
*
system@xenbr5:
lookups: hit:2560723 missed:3742809 lost:0
flows: 5
port 0: xenbr5 (internal)
port 1: eth5
system@xapi2:
lookups: hit:1660559495 missed:1241428 lost:0
flows: 11
port 0:
10 matches
Mail list logo
