Re: [ovs-discuss] does ovs 2.5.0 support openflow 1.5?

On Sat, Jul 02, 2016 at 09:36:22AM +0800, Big Strong wrote: > > > > The FAQ explains how to enable OpenFlow 1.5. > > ovs-vsctl set bridge br0 protocols=OpenFlow10,OpenFlow11,OpenFlow12, > > OpenFlow13,OpenFlow14,OpenFlow15 > > > Yes, I did exactly as FAQ says. But just as my last post tells, th

Re: [ovs-discuss] does ovs 2.5.0 support openflow 1.5?

> > The FAQ explains how to enable OpenFlow 1.5. ovs-vsctl set bridge br0 protocols=OpenFlow10,OpenFlow11,OpenFlow12, > OpenFlow13,OpenFlow14,OpenFlow15 Yes, I did exactly as FAQ says. But just as my last post tells, there is error message generated in controller which shows the switch is unabl

Re: [ovs-discuss] does ovs 2.5.0 support openflow 1.5?

You have two choices. You can enable OpenFlow 1.5 and accept that some features will not work because OpenFlow 1.5 is not fully supported, or you can use the NXM values for TCP flags and other fields, which are supported in any version of OpenFlow that supports OXM. The FAQ explains how to enable

Re: [ovs-discuss] does ovs 2.5.0 support openflow 1.5?

I'm not quite understand. So how can I enable these features such as TCP FLAGS (which is declared to be usable since ovs 2.1.0) while unable to enable openflow 1.5? Following is the error message when I tried to enable openflow 1.5 using "ovs-vsctl set bridge br0 protocols=OpenFlow10,OpenFlow11,Ope

Re: [ovs-discuss] Removing flow table from OVS

On Mon, Jun 27, 2016 at 05:37:58PM +0900, Adonis Congaro wrote: > I have found that this command is used to limit flow entries in user space: > ovs-vsctl -- --id=@ft create Flow_Table flow_limit=100 > overflow_policy=refuse -- set Bridge brflow_tables=0=@ft > > But after executing this command

Re: [ovs-discuss] OVS fdb - MAC learning Subscription

On Mon, Jun 27, 2016 at 07:21:53PM +0530, Kapil Adhikesavalu wrote: > is there a way to get notification on MACs learnt on a OVS bridge ? OVS doesn't provide a way to do that. If you implement MAC learning using the "learn" action, e.g. as described in the OVS tutorial under the tutorial/ directo

Re: [ovs-discuss] OVS - IGMP Support?

On Mon, Jun 27, 2016 at 09:56:29AM -0600, Ken Evensen wrote: > I feel like this an amateur question, but I'll ask anyway. To what extent, > if any is IGMP implemented in OVS? If it is implemented, does it support a > true rendezvous point? OVS has limited support for IGMP in its "normal switchin

Re: [ovs-discuss] change ovs source code

On Tue, Jun 28, 2016 at 10:42:38AM +0430, Mehrshad Shahmohammadi wrote: > i want to send (ofp_flow_stats) packet to controller every 5 second. > > i want to change source code of ovs . > > i don't know where is this packet in the source code of ovs? Instead of changing OVS, it's probably easier

Re: [ovs-discuss] [ovs-dev] Sampling in datapath

On Wed, Jun 29, 2016 at 05:45:43PM +0300, Marcelo Caggiani Luizelli wrote: > I've seen that there is an implementation of sampling function in the ovs > linux datapath (in file datapath/actions.c). The issue is that this > functions is sending the packet to userspace (usually used by ipix/sflow > s

Re: [ovs-discuss] [ovs-dev] Flow actions handling

On Wed, Jun 29, 2016 at 07:52:10PM +0530, Kapil Adhikesavalu wrote: > Hi, > > i have the following flow rule, > > *tun_id=0xfa0,in_port=10 actions=mod_vlan_vid:1,resubmit(0,2),resubmit(0,1)* > > and i need to know in which order actions will be performed. > > will the actions be performed in th

Re: [ovs-discuss] About the QoS of OpenvSwitch

On Thu, Jun 30, 2016 at 06:15:05AM +, Xiao Ma (xima2) wrote: > Hi, All > > I want to use the QoS of OpenvSwitch to control the bandwidth based on the > vlan id. > So I deployed it as the topology attached,and configured the flow, the qos > rule, and used iperf tool to test it. > But the resu

Re: [ovs-discuss] [ovs-dev] The QoS feature minimum guaranteed bandwidth of OpenvSwitch not work

On Fri, Jul 01, 2016 at 03:40:30AM +, Xiao Ma (xima2) wrote: > I want to use the QoS feature of OpenvSwitch to control the bandwidth based > on the vlan id(Scene 1) or port id(Scene 2). > So I deployed it as showed bellow,and configured the qos rules,the flows,and > used iperf tool to test it

Re: [ovs-discuss] c

OK, in that case, can you run the command in circumstances such that it hangs, but adding the "-v" option? That will cause it to print a log of debugging messages to stdout. Give it a couple of seconds, then hit control+C and email all of the output that it printed. Thanks, Ben. On Fri, Jul 01

Re: [ovs-discuss] c

ovs-vsctl: cannot create a bridge named br0 because a bridge named br0 already exists On Fri, Jul 1, 2016 at 5:22 PM, Ben Pfaff wrote: > If you hit control+C and run the ovs-vsctl command again, what happens? > > On Fri, Jul 01, 2016 at 05:06:27PM -0500, Charlie Li wrote: > > I don't think so -

Re: [ovs-discuss] c

If you hit control+C and run the ovs-vsctl command again, what happens? On Fri, Jul 01, 2016 at 05:06:27PM -0500, Charlie Li wrote: > I don't think so - "ps -ef" shows the following processes > > root 1458 1 0 16:17 ?00:00:00 ./ovsdb-server > --remote=punix:/usr/local/var/run/op

Re: [ovs-discuss] c

I don't think so - "ps -ef" shows the following processes root 1458 1 0 16:17 ?00:00:00 ./ovsdb-server --remote=punix:/usr/local/var/run/openvswitch/db.sock --remote=db:Open_vSw root 1464 1 0 16:18 ?00:00:09 ./ovs-vswitchd --dpdk -c 0x1 -n 4 -- unix:/usr/local/v

Re: [ovs-discuss] c

Does ovs-vswitchd die when you run that? On Fri, Jul 01, 2016 at 04:24:15PM -0500, Charlie Li wrote: > If I don't re-build DPDK and OVS, the following command will hang > > sudo ./ovs-vsctl add-br br0 -- set bridge br0 datapath_type=netdev > > Looks like Bala reported the same issue (not sure if

Re: [ovs-discuss] c

If I don't re-build DPDK and OVS, the following command will hang sudo ./ovs-vsctl add-br br0 -- set bridge br0 datapath_type=netdev Looks like Bala reported the same issue (not sure if it is caused by the same reason - not re-building OVS/DPDK) On Fri, Jul 1, 2016 at 3:21 PM, Ben Pfaff wrote:

Re: [ovs-discuss] etcd for OVN status update (was: Re: more about etcd (can it support big transactions and many monitors?))

On Wed, Jun 29, 2016 at 03:38:26PM -0700, Andy Zhou wrote: > This is one possible way to implement reference integrity with etcd: > > * DB wide versioning. > > Assign a key db/version that stores db wide transaction id. Assume the id > starts with 0. Any client issued transaction on the DB should

Re: [ovs-discuss] c

On Fri, Jul 01, 2016 at 01:57:38PM -0500, Charlie Li wrote: > By now I am successful in running openvswitch-2.5.0 with DPDK-2.2.0 by > following "INSTALL.DPDK.md ". > > I understand that I have to build DPDK and OVS for the very first time. > > # Build DPDK > cd $DPDK_DIR

Re: [ovs-discuss] Bug in Openvswitch 2.5

That really doesn't make sense. I hope you can figure out the problem. On Fri, Jul 01, 2016 at 02:09:40PM -0500, Charlie Li wrote: > Hi Bala, > > I have a similar problem if I don't build DPDK and OVS every time after > reboot. > > My workaround is to build DPDK and OVS again every time after a

Re: [ovs-discuss] Bug in Openvswitch 2.5

Hi Bala, I have a similar problem if I don't build DPDK and OVS every time after reboot. My workaround is to build DPDK and OVS again every time after a system reboot. # Build DPDK cd $DPDK_DIR make install T=x86_64-native-linuxapp-gcc # Configure & build OVS cd $OVS_DIR ./boot.sh ./configure -

[ovs-discuss] c

Hello, By now I am successful in running openvswitch-2.5.0 with DPDK-2.2.0 by following "INSTALL.DPDK.md ". I understand that I have to build DPDK and OVS for the very first time. # Build DPDK cd $DPDK_DIR make install T=x86_64-native-linuxapp-gcc # Configure & build OV

Re: [ovs-discuss] etcd for OVN status update (was: Re: more about etcd (can it support big transactions and many monitors?))

On Fri, Jul 1, 2016 at 11:07 AM, Russell Bryant wrote: > > > On Fri, Jul 1, 2016 at 1:59 PM, Andy Zhou wrote: > >> >>> A single, dedicated etcd client handling every other transaction sounds >>> like it could be a scale bottleneck. What do you think? >>> >> >> Yeah, this is not ideal. Suggesti

Re: [ovs-discuss] etcd for OVN status update (was: Re: more about etcd (can it support big transactions and many monitors?))

On Fri, Jul 1, 2016 at 1:59 PM, Andy Zhou wrote: > >> A single, dedicated etcd client handling every other transaction sounds >> like it could be a scale bottleneck. What do you think? >> > > Yeah, this is not ideal. Suggestions are welcome. > I don't have any suggestions right now... > In O

Re: [ovs-discuss] etcd for OVN status update (was: Re: more about etcd (can it support big transactions and many monitors?))

> > > A single, dedicated etcd client handling every other transaction sounds > like it could be a scale bottleneck. What do you think? > Yeah, this is not ideal. Suggestions are welcome. In OVN-SB use case, transactions are usually only issued by northd, write update should not, at least in th

Re: [ovs-discuss] etcd for OVN status update (was: Re: more about etcd (can it support big transactions and many monitors?))

On Wed, Jun 29, 2016 at 6:38 PM, Andy Zhou wrote: > > > On Wed, Jun 22, 2016 at 10:43 AM, Ben Pfaff wrote: > >> On Wed, Jun 22, 2016 at 01:56:17AM -0700, Andy Zhou wrote: >> > 3. How should the OVN databases be arranged within etcd? There are >> > >multiple possibilities: >> > > >> > >-

Re: [ovs-discuss] pci_map_resource(): cannot mmap(..): Invalid argument (0xffffffffffffffff)

Thanks Ian, That was exactly my problem. After commenting out "pci_request_regions(dev, "igb_uio");" in igb_uio.c, the issue was resolved. Charlie On Fri, Jul 1, 2016 at 6:07 AM, Stokes, Ian wrote: > Hi, > > > > This sounds similar to an issue discussed on the DPDK mailing list, > essentially

Re: [ovs-discuss] Bug OpenVSwitch 2.5.0 - Enabling IPFIX in OpenVSwitch breaks VXLAN tunneling

I clipped out the rest of the original message but you can see the full set of commands that he gave here: http://openvswitch.org/pipermail/discuss/2016-July/021875.html On Fri, Jul 1, 2016 at 9:05 AM, Daniel Ye wrote: > Hi Lluís, > > Which IPFIX do you use? As there are two kinds of IPFIX, one i

Re: [ovs-discuss] Bug OpenVSwitch 2.5.0 - Enabling IPFIX in OpenVSwitch breaks VXLAN tunneling

On Fri, Jul 1, 2016 at 5:19 AM, Lluís Gifre wrote: > Dear all, > > I'm experiencing a problem when enabling IPFIX on an OpenVSwitch where VXLAN > tunnels are configured to interconnect 2 VMs (Virtual Box). Wenyu and Daniel, this is an area that you guys have been working in, can you please take a

Re: [ovs-discuss] Throughput issue

Ok thanks.   Sent: Friday, July 01, 2016 at 4:51 AM From: "Justin Pettit" To: "james hopper" Cc: discuss Subject: Re: [ovs-discuss] Throughput issue > On Jul 1, 2016, at 1:44 PM, james hopper wrote: > > Hey Everyone, > I am using OVS-2.5 on CentOS 6.7. The CentOS box (S1) has 4 interfaces. I

Re: [ovs-discuss] Throughput issue

> On Jul 1, 2016, at 1:44 PM, james hopper wrote: > > Hey Everyone, > I am using OVS-2.5 on CentOS 6.7. The CentOS box (S1) has 4 interfaces. I > have connected 2 end hosts with it and two other interfaces are being used as > uplinks. Each uplink is also connected with another OVS switch runni

Re: [ovs-discuss] [CudaMailTagged] rules in the flow table

You may want to check out our NSDI paper on OVS: http://openvswitch.org/support/papers/nsdi2015.pdf The subtables are part of the tuple space search described there. --Justin > On Jul 1, 2016, at 8:56 AM, Amrane Ait Zeouay wrote: > > Hi, > > I was looking in the source code, and i f

[ovs-discuss] Throughput issue

Hey Everyone, I am using OVS-2.5 on CentOS 6.7. The CentOS box (S1) has 4 interfaces. I have connected 2 end hosts with it and two other interfaces are being used as uplinks. Each uplink is also connected with another OVS switch running on CentOS (S2, S3). Both, S2 and S3, have 2 interfaces each a

Re: [ovs-discuss] pci_map_resource(): cannot mmap(..): Invalid argument (0xffffffffffffffff)

Hi, This sounds similar to an issue discussed on the DPDK mailing list, essentially if CONFIG_IO_STRICT_DEVMEM is turned on in the kernel this behavior occurs. More details in the link below. http://dpdk.org/ml/archives/dev/2016-July/043122.html Are you using a kernel 4.5+ with fedora 23? Th

Re: [ovs-discuss] does ovs 2.5.0 support openflow 1.5?

As Joe mentioned those later versions are considered experimental, but many of the features are supported. By the way, 0x4 maps to OpenFlow 1.3, not 1.4. --Justin > On Jul 1, 2016, at 11:23 AM, Big Strong wrote: > > I checked the release notes at http://openvswitch.org/releases/NEWS-2.5.0.

Re: [ovs-discuss] does ovs 2.5.0 support openflow 1.5?

On 1 July 2016 at 11:23, Big Strong wrote: > I checked the release notes at http://openvswitch.org/releases/NEWS-2.5.0. > And it seems that openflow 1.5 is supported since ovs 2.4.0. But after > install ovs 2.5.0, I checked the version and find it only supports openflow > 1.4, is there any problem

[ovs-discuss] does ovs 2.5.0 support openflow 1.5?

I checked the release notes at http://openvswitch.org/releases/NEWS-2.5.0. And it seems that openflow 1.5 is supported since ovs 2.4.0. But after install ovs 2.5.0, I checked the version and find it only supports openflow 1.4, is there any problem? $ sudo ovs-ofctl --version > ovs-ofctl (Open vSwi