Can any one please help me understand the invisible link that carries
traffic from gre0 through birdge with NIC attached.
On Thu, Dec 5, 2013 at 10:10 PM, kevin parker wrote:
> Hi All,
> I have a working GRE connected between one kvm host ovs 1.11 and
> xenserver ovs 1.4.6
>
> I have cr
Hi Kevin
I'm trying the below link for the same functionality
http://linuxmantra.com/2013/08/gre-tunneling-with-open-vswitch.html
I'm able to reach tep0 of Hypervisor1 from both the VMs but I wonder why
I'm not able to reach tep0 of Hypervisor2 from VM1 and VM2.
Could you help regarding thi
Thanks Jesse,
yes,it is a driver issue,Since other servers don't
have this driver issue they are not showing this in dmesg,but does that
mean that ovs send packets with out considering MTU of interface.
Thanks,
kevin
On Mon, Dec 2, 2013 at 11:54 PM, Jesse Gross wrote:
> It'
It seems you have to pass switch CA certificate
"--ca-cert=/switchca/cacert.pem" while starting test-controller.
Arun
From: Kelvin keros
Date: Thu, 5 Dec 2013 19:51:19 +0530
To:
Subject: [ovs-discuss] ovs-vsctl set-controller fails with test-controller
using SSL.
Hi OVS Team,
I was tryin
Thank you very much, Joe!
The way I am sending actions from Ryu:
actions.append(ofp_parser.OFPActionPushVlan(ether.ETH_TYPE_8021Q))
actions.append(ofp_parser.OFPActionSetField(vlan_vid=4))
actions.append(ofp_parser.OFPActionPushMpls(0x8847))
actions.append(ofp_parser.OFPActionSetField(mpls_label=
Hi All,
currently I'm using OVS together with OpenVPN.
Now I want to restrict access of some (not all) VPN clients to DNS and RDP.
I've seen that iptables is not working.
Is here any other way to apply firewall rules?
Thanks,
Klaus
___
discuss mailing
On Fri, Dec 06, 2013 at 04:32:44PM +, Klaus Hochlehnert wrote:
> currently I'm using OVS together with OpenVPN.
> Now I want to restrict access of some (not all) VPN clients to DNS and RDP.
>
> I've seen that iptables is not working.
> Is here any other way to apply firewall rules?
OpenFlow?
On Fri, Dec 6, 2013 at 3:43 AM, kevin parker wrote:
> Thanks Jesse,
> yes,it is a driver issue,Since other servers don't have
> this driver issue they are not showing this in dmesg,but does that mean that
> ovs send packets with out considering MTU of interface.
Layer 2 switc
Please don't drop the list.
You can avoid the bug by specifying an out_group of OFPP_ANY in your
flow_mod.
On Fri, Dec 06, 2013 at 06:59:49PM +0800, chen zhang wrote:
> thx for your kind reply.so how should i change to make it what i want?
>
>
> 2013/12/6 Ben Pfaff
>
> > Please don't drop the
Never used this before.
Maybe any hint (link, ...)?
-Original Message-
From: Ben Pfaff [mailto:b...@nicira.com]
Sent: Friday, December 06, 2013 5:43 PM
To: Klaus Hochlehnert
Cc: 'discuss@openvswitch.org'
Subject: Re: [ovs-discuss] OVS + OpenVPN + firewall
On Fri, Dec 06, 2013 at 04:32:44
ovs-ofctl(8)?
On Fri, Dec 06, 2013 at 04:52:26PM +, Klaus Hochlehnert wrote:
> Never used this before.
> Maybe any hint (link, ...)?
>
> -Original Message-
> From: Ben Pfaff [mailto:b...@nicira.com]
> Sent: Friday, December 06, 2013 5:43 PM
> To: Klaus Hochlehnert
> Cc: 'discuss@open
On Fri, Dec 06, 2013 at 11:16:47AM +0400, Vasiliy Tolstov wrote:
> Does it possible to isolate one network from another by adding mpls
> label (for example user id) and on recieve side accept packets only
> with specific label?
I believe that this is possible given a solid MPLS implementation, but
On 6 December 2013 07:39, Nataliia Trifonova
wrote:
> ...
> should push MPLS label and then VLAN tag. Results are: when I just PushMpls,
> SetField(mpls_label) and PushVlan, then VLAN tag 0 is added, but when I want
> to add SetField(vlan_vid), VLAN tag is not added at all. But the rule looks
> li
2013/12/6 Ben Pfaff :
> I believe that this is possible given a solid MPLS implementation, but I
> doubt that the OVS implementation of MPLS is good enough yet.
Thanks! What parts of MPLS missing to do such thing?
--
Vasiliy Tolstov,
e-mail: v.tols...@selfip.ru
jabber: v...@selfip.ru
__
On Fri, Dec 06, 2013 at 09:55:04PM +0400, Vasiliy Tolstov wrote:
> 2013/12/6 Ben Pfaff :
> > I believe that this is possible given a solid MPLS implementation, but I
> > doubt that the OVS implementation of MPLS is good enough yet.
>
> Thanks! What parts of MPLS missing to do such thing?
I've rec
Just in case anyone else is interested. After some testing it works for me...
# Delete old rules for source ip
ovs-ofctl del-flows br0
"in_port=51,table=0,ip,vlan_tci=0x,nw_src=10.134.38.194/32"
# Default action for source ip
ovs-ofctl add-flow br0
"in_port=51,table=0,ip,vlan_tci=0x,idle
2013/12/6 Ben Pfaff :
> I've received informal reports that basic functionality like popping off
> an MPLS label doesn't work. Because MPLS is still under heavy
> development, I haven't bothered to follow up on those reports (maybe
> they are wrong or misinformed) but I still wouldn't rely on it.
Thanks Ben.
Why would dhclient requests destined to eth0 would be seen at eth1?
And why would br0 (with mac address A) would use a dhcp release destined to
mac address B?
Y.
On Thu, Dec 5, 2013 at 9:41 PM, Ben Pfaff wrote:
> On Wed, Dec 04, 2013 at 04:01:16PM -0800, Yiannis Yiakoumis wrote:
>
18 matches
Mail list logo
