Hi Gross,
That works!
Thank you very much!
Best regards,
ZHANG Zhiming
Yunshan Networks
From: Jesse Gross
Date: 2015-11-09 20:20
To: zhangzhiming
CC: discuss
Subject: Re: [ovs-discuss] ipfix leads to kernel crash
This is the relevant commit:
> -Original Message-
> From: discuss [mailto:discuss-boun...@openvswitch.org] On Behalf Of Gayathri
> Sent: Thursday, November 5, 2015 5:01 AM
> To: discuss@openvswitch.org
> Subject: [ovs-discuss] Fwd: Run dpdk application in a VM
>
> Hi *,
>
> I am new to dpdk please do help me.
>
>
In the old poc (march - may 2015) the way the service chains worked was a
custom flow programmed based on an nsh key per chain. The nsh header was
sort of the service chain key. The flow was programmed from ovs agent. For
instance you would have video compression and url filter on a chain. They
When Open vSwitch is running on DPDK, we know that the OVS is in userspace. I
read the source code and found that the userspace has a cache like microflow.
My question is that is there a version that also has megaflow in userspace ?___
discuss mailing
Hi,
We're going to have an OVS with DPDK Meetup on Wednesday,
November 18th (day after OVS Conference) in San Jose.
While there will be some OVS with DPDK presentations in the
OVS conference, having a lot of the people who contribute to
OVS with DPDK in the same place gives a good opportunity
Hi Kevin,
Thank you for the reply. With vhost-user it was not working and vswitchd
crashed in host, aswell as guest. So was thinking if that was the right
mode to have dpdk applications working in the VM.
Regards,
Gayathri
On Mon, Nov 9, 2015 at 5:40 AM, Traynor, Kevin
> My original thought was a new table of chains. Each chain has a list of
> service endpoints (originally i had this as logical ports, but it'd need
> to be IP or MAC addresses, I guess). A chain would also have a match
> defined in the same syntax used by ACLs. I imagined the implementation
>
On 11/09/2015 01:18 PM, v_d...@dell.com wrote:
>> My original thought was a new table of chains. Each chain has a list of
>> service endpoints (originally i had this as logical ports, but it'd need
>> to be IP or MAC addresses, I guess). A chain would also have a match
>> defined in the same
Hello everyone:
After testing, there may be a memory lead problem in function
ovsthread_key_destruct__,for details,see e-mail attachment.
Thanks!
5ABUW%U08Y]VMP{}1DK%553.png
Description: Binary data
___
discuss mailing list