Hi Joe,
I can reproduce this problem with the kernel module upstream versione
3.13.1[0-1] and with the version build from the branch 2.1 (using kernel 3.11)
and 2.2 (using kernel 3.13).
Right now I have the kernel 3.13 with the openvswitch module from the upstream
kernel.
Thanks a lot!
Regards
2014-06-24 19:23 GMT+04:00 Lori Jakab :
> Others who know better can correct me, but I think there is more work
> involved than that. For now, the data structures and code supporting
> tunnels is IPv4 specific, and you would need to change more than one file,
> and not just the datapath (user spac
Hi Jesse,
Unfortunately I don’t have any other logs into dmesg, kern.log and syslog.
The problem is that I have a big amount of packet loss and retransmission for
25% of time. So the OVS Rule for 75% time works. It seems like a overflow kind
but I’m not capable of debugging C++ code anymore.
Th
Hi Jesse,
Unfortunately I don’t have any other logs into dmesg, kern.log and syslog.
The problem is that I have a big amount of packet loss and retransmission for
25% of time. So the OVS Rule for 75% time works. It seems like a overflow kind
but I’m not capable of debugging C++ code anymore.
Tha
On Fri, Jun 20, 2014 at 8:01 AM, Federico Iezzi wrote:
> Hi Guys,
>
> I’m in trouble with our staging OpenStack env based on IceHouse Neutron
> and Havana Nova.
> We are using Ubuntu 12.04.4 with Kernel 3.13 mirror release 11.
>
> With the latests OVS versione (2.1.2, 2.1.3, 2.2.0 and 2.3 DEV
I'm still learning OVS, but I'll take a stab:
1) Before the ping starts, an ARP request will be sent which will be
broadcast. It won't match the destination MAC because the purpose of the
ARP is to find out what the MAC is. Normally the VLAN would take care of
broadcasting the packet out to all h
On Tue, Jun 24, 2014 at 10:08:28AM -0700, shruti rao wrote:
> mininet> sh ovs-ofctl dump-flows s1
> NXST_FLOW reply (xid=0x4):
> cookie=0x0, duration=160.318s, table=0, n_packets=0, n_bytes=0,
> idle_age=160, actions=NORMAL
> mininet> sh ovs-ofctl dump-flows s2
> NXST_FLOW reply (xid=0x4):
> cook
Hi,
I am new to Open vSwitch and am trying to create a VLAN with two switches
(s1, s2) having 2 hosts (s1 having red1 and blue1 and s2 having red2 and
blue2) each and no controller. My configuration is as mentioned below:
mininet> net
s1 lo: s1-eth1:red1-eth0 s1-eth2:blue1-eth0 s1-eth3:s2-eth1
s
Dpctl and everything it shows is an implementation detail. Use OpenFlow
port numbers in OpenFlow.
On Jun 24, 2014 9:23 AM, "Tahhan, Maryam" wrote:
> Ok, I'm still a little confused, which would you use when adding a flow
> with ofctl, the port numbers from ofctl or from dpctl, I'm assuming ofctl
Ok, I'm still a little confused, which would you use when adding a flow with
ofctl, the port numbers from ofctl or from dpctl, I'm assuming ofctl
-Original Message-
From: Ben Pfaff [mailto:b...@nicira.com]
Sent: Tuesday, June 24, 2014 5:07 PM
To: Tahhan, Maryam
Cc: discuss@openvswitch.or
On Tue, Jun 24, 2014 at 04:01:02PM +, Tahhan, Maryam wrote:
> I'm just wondering why does ovs-dpctl show different port numbers to
> what is returned from ovs-ofctl show?
There's no reasonable way to have them be the same, since a single
kernel datapath implements multiple OpenFlow switches.
_
Hi
I'm just wondering why does ovs-dpctl show different port numbers to what is
returned from ovs-ofctl show?
[root@silpixa00378251 test_vanilla_ovs]# ovs-dpctl show
system@ovs-system:
lookups: hit:182194988 missed:36415 lost:181
flows: 0
masks: hit:247644295 total:0 hit/p
On 6/24/14, 3:15 PM, Vasiliy Tolstov wrote:
2014-06-24 14:45 GMT+04:00 Lori Jakab :
No, and I haven't seen any patches proposed to implement that on the
developer mailing list either.
Very bad =(. As i understand to get tunnels working in ipv6 (vxlan) i
need to patch vport-vxlan.c ?
kernel dat
Guys,
you committed some great changes like Geneve and DPDK
support: "dpif-netdev: batch packet processing" and "dpif-netdev:
Upcall: Remove an extra memcpy of packet data" is really great
things to speed up SDN/NFV OVS-powered solutions.
Can you share some roadmap? This helps us to keep step
I'm not familiar with mn, but when you tag a port in OpenVswitch as an
access port, it tells openvswitch that packets should _not_ be tagged on
that interface. Instead, traffic is untagged, but presumed to belong to
vlan 20 by the switch. So in that setup, only traffic on eth2 would be
tagged. I
2014-06-24 14:45 GMT+04:00 Lori Jakab :
> No, and I haven't seen any patches proposed to implement that on the
> developer mailing list either.
Very bad =(. As i understand to get tunnels working in ipv6 (vxlan) i
need to patch vport-vxlan.c ?
kernel data path (i'm use 3.12) not needs to be patch
On 6/24/14, 1:43 PM, Vasiliy Tolstov wrote:
2014-06-24 14:42 GMT+04:00 Lori Jakab :
No, only IPv4 addrress literals are accepted.
If i need ipv6 based tunnel? Does openvswitch supports ipv6 based tunnels?
No, and I haven't seen any patches proposed to implement that on the
developer mailin
2014-06-24 14:42 GMT+04:00 Lori Jakab :
> No, only IPv4 addrress literals are accepted.
If i need ipv6 based tunnel? Does openvswitch supports ipv6 based tunnels?
--
Vasiliy Tolstov,
e-mail: v.tols...@selfip.ru
jabber: v...@selfip.ru
___
discuss maili
On 6/24/14, 12:39 PM, Vasiliy Tolstov wrote:
Does it possible to create tunnel (gre, vxlan) providing hostname in
remote_ip and openvswitch resolve ip for this hostname?
No, only IPv4 addrress literals are accepted.
-Lori
___
discuss mailing list
dis
Does it possible to create tunnel (gre, vxlan) providing hostname in
remote_ip and openvswitch resolve ip for this hostname?
--
Vasiliy Tolstov,
e-mail: v.tols...@selfip.ru
jabber: v...@selfip.ru
___
discuss mailing list
discuss@openvswitch.org
http://o
Hi all,
I am trying to implement vlan in mininet open vSwitch. I did the following ..
Open vSwitch version : 1.9
Kernel version : 3.8.0-19-generic
1. Create a topology using sudo mn command as,
mininet@mininet-vm:~$ sudo mn --topo=linear,2
*** Creating network
*** Adding cont
Hello. As i see many commits goes to master after last 2.x release.
When the new release date? Or some roadmap items needs to be completed
for next release?
--
Vasiliy Tolstov,
e-mail: v.tols...@selfip.ru
jabber: v...@selfip.ru
___
discuss mailing list
22 matches
Mail list logo
