Hi,
I am trying to test the default maximum value of flow_limit in a
flow_table.
To achieve this, I executed the following commands:
[root@user]# ovs-ofctl dump-tables br0 | less OFPST_TABLE reply (xid=0x1):
255 tables
0: classifier: wild=0x3f, max=100, active=9
looku
ok. I got your point.
thanks for the response.
Regards,
Akash
On Thu, May 16, 2013 at 12:04 PM, Ben Pfaff wrote:
> That does not make much sense. First, inactivity probes are by default
> disabled for Unix sockets. Second, a disconnected socket has no traffic.
> On May 15, 2013 11:30 PM, "akash
That does not make much sense. First, inactivity probes are by default
disabled for Unix sockets. Second, a disconnected socket has no traffic.
On May 15, 2013 11:30 PM, "akash sharma" wrote:
> I want to check that inactivity probe message is not sent by the switch to
> the controller after the u
I want to check that inactivity probe message is not sent by the switch to
the controller after the unix socket is disconnected.
On Thu, May 16, 2013 at 11:47 AM, Ben Pfaff wrote:
> OK. But what is your higher level goal? The best way to observe the
> inactivity probe message depends on how you
OK. But what is your higher level goal? The best way to observe the
inactivity probe message depends on how you plan to react to it.
On May 15, 2013 11:07 PM, "akash sharma" wrote:
> Thanks for the response.
> It resolves my two queries.
> Regarding query#2, I need to capture the "inactivity-prob
Thanks for the response.
It resolves my two queries.
Regarding query#2, I need to capture the "inactivity-probe" message sent by
the switch to the controller.
But I am not able to understand how can I capture the messages/packets
exchanged between the switch and the unix controller.
Regards,
Akash
On Thu, May 16, 2013 at 11:23:23AM +0530, akash sharma wrote:
> 1. I need to disconnect the unix socket.
>
> I deleted the link created above using the following command:
>
> [root@user]rm /usr/local/var/run/openvswitch/br0.mgmt
>
> Am i correct or is there any other way to disconnect the uni
Hi,
I have created a bridge br0.
then I created a unix domain socket on bridge br0 using the following
command:
[root@user]# ln -s /usr/local/var/run/openvswitch/br0.mgmt
/usr/local/var/run/openvswitch/br0.controller
[root@user]# ovs-vsctl set-controller br0
unix:/usr/local/var/run/openvswit
On Wed, May 15, 2013 at 03:19:54AM +, Jing Ai wrote:
> I've noticed the REAME.gcov provide a way to enable coverage (however,
> some coverage integration support for lcov has been removed by Ben
> last year). However, when I used lcov to analyze the generated .gcda
> files, I always got either
That is definitely a good idea and I will try to setup a Wiki soon on github so
that everyone can join to learn and contribute.
--发件人：Volkan
YAZICI 发送时间：2013年5月14日(星期二) 15:55收件人：泠茗
抄 送：discuss 主 题：Re:
[ovs-discuss] The Implementat
Thanks so much! It looks like this is going to be easier than I at first
suspected.
On Tue, May 14, 2013 at 2:08 PM, Ben Pfaff wrote:
> On Tue, May 14, 2013 at 12:07:46PM -0700, Nick wrote:
> > Ok that sounds better. So would it be best to use the same eth0 (or
> > whatever the physical interf
On Tue, May 14, 2013 at 11:25 PM, ananthan wrote:
> Thanks for your reply,but now things are more complicated,please go through
> this,i have also added tcpdump output.
You need to ask a concise question. Nobody is going to analyze your
situation for you.
_
Hi,
I have a setup with linux kernel 3.2, xen 4.1.2 and openvswitch 1.7.3 - on 10G
infrastructure.
The NAT firewall is connected to openvswitch which causes connection delays and
dropped packets under high load.
With packet counts > 150.000/s delay rises and cpu load > 90%.
Packet counts > 400
13 matches
Mail list logo
