Hey Adam,
Sorry for the delayed reply, I have digressed by some issue,
Could I know the xen server version? I do not see any mem grow
on kvm, so I'd like to test using the same xen version.
Thanks,
Alex Wang,
On Fri, Dec 5, 2014 at 2:03 AM, Adam Mazur wrote:
> Hi Alex,
>
Hey Rahul,
Please do not cross-post..
I'll have someone dealing with ovs performance test reply to your original
post soon,
Could you also clarify (in your original post) if you are testing using vm
or
bare metal machine?
Thanks,
Alex Wang,
On Thu, Dec 4, 2014 at 10:42 PM, Rahul
4528985512
86805523496241024860881119731815148054128138029000137181138581
Thanks,
Alex Wang,
On Wed, Nov 26, 2014 at 10:35 PM, Rahul Arora
wrote:
> Hi Team,
>
> We are doing comparison of throughput and CPU consumption between OVS
> 2.3.0 and kernel bridge with different packet size.
>
> We are observing hu
Hey Rahul,
The kernel version should not make a difference,
I'm curious, since ovs-2.3.0 is multithreaded, not sure how do you measure
the cpu usage to be 100%?
Also, could you post the full `mpstat -P ALL` output, so we can see the
distribution of cpu usage,
Thanks,
Alex Wang,
On Tue
Hey Adam,
I'll try xen 4.3,
I tested on my xen6.2 build 70446 vm, did not see the memory util surge,
Thanks,
Alex Wang,
On Mon, Dec 8, 2014 at 8:17 AM, Adam Mazur wrote:
> Hi Alex,
>
> I'm using xen 4.3, directly as-is from debian-wheezy distribution.
> Kernel 3.17
I believe it has 16 cores, with hyperthreading disabled,
On Tue, Dec 9, 2014 at 9:37 PM, Rahul Arora
wrote:
> Hi Alex,
>
> The platform which you are using for your performance testing is of how
> many cores??
>
> On Tue, Dec 9, 2014 at 10:45 PM, Alex Wang wrote:
>
&g
'modinfo openvswitch'?
Never seen such warning with 2.3 kernel module,
Thanks,
Alex Wang,
On Tue, Dec 9, 2014 at 10:47 PM, Rahul Arora
wrote:
> Alex,
>
> I am getting these logs when i am getting 100% CPU usage.
>
> Please ha
t did you modify,
3. Repeat your test and show me the full `mpstat -P ALL` output,
4. Run 'modinfo openvswitch' and show me the output,
Also, please do not cross-post anymore without providing more
details about your experiment,
Thanks,
Alex Wang,
On Wed, Dec 10, 2014 at 1:08 AM, Rah
,
Thanks,
Alex Wang,
On Fri, Dec 12, 2014 at 3:33 AM, Saloni Jain wrote:
>
> Hi Team,
>
> We are trying to install ovs with DPDK refering to INSTALL.DPDK.md file.
>
> dpdk 1.7.1 is taken from the link *dpdk.org/download
> <http://dpdk.org/download>*
> cloned the latest m
vs-ofctl (0x1 => OpenFlow 1.0, 0x4 => OpenFlow 1.3)
For now, ovs-ofctl only enables OpenFlow 1.0 by default.
So you must use -O OpenFlow13
Thanks,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
le name=table1
ovs-vsctl set Flow_Table table0 prefixes=ip_dst,ip_src
ovs-vsctl set Flow_Table table1 prefixes=[]
Signed-off-by: Jarno Rajahalme
Acked-by: Ben Pfaff
Thanks,
Alex Wang,
On Tue, Dec 16, 2014 at 12:58 AM, 李伟 wrote:
>
> Hi Justin Pettit
> Thank you for y
Have you encountered similar issue?~ Is there any special
configuration (not included in INSTALL.NetBSD) for make and make
check ovs?
Thanks,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
Hey Padma,
OVS supports BFD, please refer to 'man ovs-vswitchd.conf.db', Interface
Table section for detailed configuration.
Thanks,
Alex Wang,
On Wed, Jan 28, 2015 at 10:04 PM, Padma Jayasankar
wrote:
> Hi,
> For my project i have to send BFD packets between the source and
ets through the tunnel... No OpenFlow rules are involved...
We currently do not support multi-hop bfd.
However, at this time, it is still unclear what kind of link
(tunnel/point-to-point physical link) you want you monitor... maybe you
should clarify first,
Thanks,
A
. This centralizes the decision when to
include VXLAN compat code to a single place which eases further changes.
Reported-by: Pravin Shelar
Signed-off-by: Thomas Graf
Acked-by: Pravin B Shelar
Thanks,
Alex Wang,
On Thu, Feb 5, 2015 at 8:50 AM, eyal wrote:
> OS: Cent
multi-hop. If the intermediate switches just
forward
using flows and do not decrement the ttl, then, it could work but it is not
really
multi-hop. When you have intermediate vendor routers, the ttl of the udp
packets will change. In ovs bfd implementation, the received udp packet
must have TTL=255.
Sorry for this delayed reply,
We found the issue, and working on a fix. Will send more updates soon,
Thanks,
Alex Wang,
On Tue, Feb 17, 2015 at 4:59 PM, Atanu Ghosh wrote:
> Hi,
>
> I have been trying to use OVS on FreeBSD 10.1. I have hit a number of
> issues.
>
> 1) T
Hey Atanu,
We pushed the fixes to master and branch-2.3. I tested on my FreeBSD-9.3
VM. Everything should work fine now,
Thanks,
Alex Wang,
On Fri, Feb 20, 2015 at 9:54 AM, Alex Wang wrote:
> Sorry for this delayed reply,
>
> We found the issue, and working on a fix. Will send mor
Hey,
I think you need to load the 'if_tap.ko'~
Thanks,
Alex Wang,
On Mon, Feb 23, 2015 at 3:29 PM, Atanu Ghosh wrote:
> Hi,
>
> The code in the master branch now passes all the tests on FreeBSD 10.1.
> However, I still see the same errors when trying to create a bridge.
Okay, could you show me the ovs-vswitchd.log?
Usually, there should be some messages explaining the error.
Thanks,
Alex Wang,
On Tue, Feb 24, 2015 at 9:33 AM, Atanu Ghosh wrote:
> Hi,
>
> It seems to be loaded:
>
> k3$ kldstat | grep if_tap
> 131 0x81b93000
According to
https://www.freebsd.org/cgi/man.cgi?query=tap&apropos=0&sektion=4&manpath=FreeBSD+9.3-RELEASE&arch=default&format=html
it seems that freebsd does not support setting SIOCSIFNAME for tap
interfaces.
I think I'll just remove the code in ovs.
Thanks,
Alex Wa
Also, just make sure you did not terminate ovs by 'kill' it, since that way,
the opened tap interfaces are not removed and could cause errors when
you restart ovs.
Thanks,
Alex Wang,
On Tue, Feb 24, 2015 at 3:35 PM, Atanu Ghosh wrote:
> Hi,
>
> If SIOCSIFNAME is used to cha
dd comment to
warn users (that the function should only be called with > 128-bit word)?
The patch is sent out here:
http://openvswitch.org/pipermail/dev/2015-February/051727.html
Thanks,
Alex Wang,
On Tue, Dec 30, 2014 at 10:28 AM, Andrew Mann wrote:
> For what it's worth, Bob Jenki
te) inputs, and nothing smaller. I wonder if it would be useful to
> have a test on smaller input, but larger than 32-bit (perhaps 128 or
> 256 bits).
>
Makes sense to me,
Working on adding it.
Alex Wang,
>
> On 26 February 2015 at 10:44, Alex Wang wrote:
> > Hey All,
> >
the tunneled
> traffic out of the machine. Another option to try out is not to use
> LACP bonds but rather something like active-backup (man
> ovs-vswitchd.conf.db -> Bonding Configuration)
>
This is a good suggestion.
> I have added 2 people familiar with OVS bonding code. They should be
> able to discern better.
>
Thanks,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
el=0/0,proto=112/0,tclass=0/0,hlimit=255/0,frag=no/0xff)),
packets:217320, bytes:20428080, used:0.548s,
actions:pop_vlan,24,29,push_vlan(vid=303,pcp=0),1
"""
Could you verify the output ports using 'ovs-dpctl show'?
Thanks,
Alex Wang,
On Fri, Mar 6, 2015 at 2:08 AM, Tobias
Hey,
I did not see any reason for the second flow to be deleted after 120s.
And you did not config the 'idle_timeout' which could cause the deletion.
So, both flow should be there forever.
Which version of ovs are you using?
Thanks,
Alex Wang,
On Mon, Mar 23, 2015 at 9:03 PM, Vi
vs-vsctl show' output of Box-B? For traffic sent from
untagged VLAN port of Box-B, the vmbr0 of Box-A should not forward it
to tagged VLAN ports.
Thanks,
Alex Wang,
>
> Here is my basic setup
> ##
> ovs-vsctl show
> 8031569b-5210-4334-88fd-36704af3a667
> Br
s, one of them
> being 1115.
>
>
> [root@box-b /root $ ovs-vsctl show
> 872e809c-3121-4a4f-9fe8-8e436a7cf101
> ovs_version: "2.3.1"
>
>
> Any Ideas where I could look at?
>
>
Thx for providing the info~
Could you try applying the s
...
`ovs-dpctl show`
`ovs-appctl dpif/dump-flows vmbr0`
Thanks,
Alex Wang,
On Tue, Apr 7, 2015 at 11:33 AM, Ben Schmidt wrote:
>
>
> On Thu, Apr 2, 2015 at 7:01 PM, Alex Wang wrote:
>
>>
>>>> What is the 'ovs-vsctl show' output of Box-B? For traffic sent fr
10"
tag: 10
Interface "vlan10"
type: internal
ovs_version: "2.3.1"
Still could not reproduce the issue...
May be worth double check the kernel module version, and the exact commit
of your ovs.
Thanks,
Alex Wang,
On Mon, Apr
You may want to check the '--with-flows file' option if you want to hack
something out.
~
On Fri, May 15, 2015 at 9:18 AM, Ben Pfaff wrote:
> On Wed, May 13, 2015 at 10:49:50PM +0200, Krzysztof Wojciechowski wrote:
> > I am searching for this for 2 days. Is it possible to change flows
> > hard
Fwd to Andy,~
On Sun, May 17, 2015 at 4:29 AM, zhangzhiming
wrote:
> Hi,
>
> I found one patch was omitted to be pushed to branch-2.3, which leads to
> double freed skb.
> Could someone to confirm the patch and submit it to branch-2.3?
> Thanks!
>
> Here is the patch information:
>
>
> commit
,
Alex Wang,
On Wed, Jun 3, 2015 at 10:37 PM, Tony van der Peet <
tony.vanderp...@gmail.com> wrote:
> I use OpenVSwitch and occasionally upgrade from the tip of master. My
> previous upgrade was in Sep/2014, and I have just upgraded to last Friday's
> tip. A number of previousl
gt;
This seems to indicate that the ovs-vswitchd process is dead locked...
I could not find comment in branch-2.3 that relates to dead lock. It would
be
helpful if you can gdb into the running ovs-vswitchd process and provide
the backtraces when there is a failure.
Or could I know what your set
Thx for the info, I'll try reproduce it on my local setup using
active-backup
bond. And use it as managment interface,
Will update if I run into the same issue~
Thanks,
Alex Wang,
On Wed, Jun 3, 2015 at 11:52 PM, Chris wrote:
> Hello Alex,
>
>
>
> I will do the gdb d
Glad to hear you figured it out,
Curious about:
1. Do you need to do flow remove in the loop after the "for in_port..." loop
in you test?
2. Where in the outer loop do you add the sleep.?
Thanks,
Alex Wang,
On Fri, Jun 5, 2015 at 12:02 AM, Tony van der Peet <
tony.vanderp...@gma
nably long
"""
And post the output, if a thread holding the rwlock for very long time,
there will be such log in ovs-vswitchd.log
Thanks,
Alex Wang,
On Tue, Jun 23, 2015 at 8:51 PM, Chris wrote:
> Hello Alex,
>
>
>
> It happened again and I got some gdp traces, plea
On Tue, Jun 30, 2015 at 9:48 AM, Xuemei Liu wrote:
> Hi,
>
> I want to generate forwarding rules for ovs, and need to get the port id
> (0, 1, ...) at the
>
Assume you are talking about OpenFlow flows, then please use 'ovs-ofctl
show '
> ovs from the port name (eth0, ...). I tried ovs-vsctl
do not need to care about it,
please refer to
function bfd_put_packet() in bfd.c module for how bfd packet is composed,
Thanks,
Alex Wang,
> thanks!
>
>
>
> ___
> discuss mailing list
>
there... and we may fail
to include the final stats (between notifying flow deletion and actually
deleting the flow in
datapath)...
Ben, do you think we should address this?
Thanks,
Alex Wang,
> My test script is as attached as the attachement. An example output is as
> follows:
>
Hey Kris,
There was a bug in the "add_vsp()" function in "ofproto/ofproto-dpif.c".
And the fix is in this patch
http://git.openvswitch.org/cgi-bin/gitweb.cgi?p=openvswitch;a=commit;h=deea120099d23fac3f687ec302351e38a21ee353
.
I think this may be the reason for your problem.
> tap2
> statistics : {stp_error_count=0, stp_rx_count=0, stp_tx_count=55}
> status : {stp_port_id="8003", stp_role=designated,
> stp_sec_in_state="66", stp_state=forwarding}
>
> Acutally i dont' know what meaning of ab
Thanks Ben, I just realized that before branch-1.10, odp and ofp use same
port number.
Very sorry for the misleading comments, Kris.
You problem should not be caused by that bug.
On Fri, Jun 7, 2013 at 9:57 AM, Ben Pfaff wrote:
> On Fri, Jun 07, 2013 at 09:52:54AM -0700, Alex Wang wr
Hey,
You could check the tutorial/Tutorial doc. There are plenty of examples
about flow priority.
Regards,
Alex Wang,
On Mon, Jul 1, 2013 at 8:43 AM, vis reddy wrote:
> Hello All
> I am trying to figure out the exact meaning of priority of flow, could
> some one explain with a
Hey,
Actually, you can have multiple gre tunnels with same src/dst ip by using
different 'key's.
e.g.
"""
ovs-vsctl add-port br0 p1 -- set Interface p1 type=gre
options:remote_ip=10.24.122.78 options:key=1
"""
Kind Regards,
Alex Wang,
On Fri, Jul
Hey,
Also, could you check if there is traffic on the link?
e.g. the "ovs-dpctl dump-flows" output, or "tcpdump" output
There may be some broadcast packets looping and jamming the link.
Kind Regards,
Alex Wang,
On Mon, Jul 1, 2013 at 10:31 AM, Jesse Gross wrote:
> On
Hey JinHwan,
Could you also provide the "ovs-dpctl dump-flows" and "tcpdump" output?
Kind Regards,
Alex Wang,
On Fri, Jul 5, 2013 at 12:23 AM, JinHwan Hwang wrote:
> I've followed ovs gre setting described at
> this<http://blog.scottlowe.org/2013/05/07/u
Hey JinHwan,
Thanks for providing more info,
Seem to me that you captured those flow dumps and tcpdump at different
time, so they may not correspond to each other.
I'll spend some time and try implementing this and see if I have the same
issue. You used KVM, right?
Below are my observations fro
stands for action that encapsulates the packet based on the tunnel
specification and then outputs packet to datapath port 4 and port 5.
Great pleasure chatting with you. ;D
Kind Regards,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
igured as:
"actions=strip_vlan,2", for striping the VLAN header before outputting the
packet.
Kind Regards,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
Hey Ben,
The bug is that "netdev_get_in4()" and "netdev_get_in6" return 0 when there
is ip address.
Fix is here:
"""
if (!netdev_open(vlan_dev->name, "system", &netdev)) {
if (!netdev_get_in4(netdev, NULL, NULL) ||
!netdev_get_in6(net
No problem,
This is probably not easy to test. Since the vlandev.c read the vlan config
from "/proc/net/vlan/config" file, seems we must need root privilege to
modify the config and test.
On Tue, Jul 23, 2013 at 11:24 AM, Ben Pfaff wrote:
> On Mon, Jul 22, 2013 at 05:46:20PM -07
interface,
is not physical.
To know more about interface types, do "man ovs-vswitchd.conf.db" and check
"Interface TABLE"
Also this post can be helpful:
http://blog.scottlowe.org/2013/05/07/using-gre-tunnels-with-open-vswitch/
Kind Regards,
Alex Wang,
_
each other?
>
What do you mean by setting gre tunnel to br1? Is this like below: ?
ovs-vsctl add-port br1 gre_br1 -- set interface gre_br1 type=gre
options:remote_ip=
If it is, it will not work. Because the machine does not know how to get to
"ip address of br1 on the other
tch.org/pipermail/discuss/2013-July/010440.html
Kind Regards,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
No you do not need to do that.
On Thu, Aug 1, 2013 at 11:35 PM, Prashant Chougule
wrote:
> do i need to add route for br1 as per above example, because in my route
> table gateway is 0.0.0.0 for both bridges.
>
>
> On Fri, Aug 2, 2013 at 10:41 AM, Alex Wang wrote:
>
>&g
ch pair of nodes?
Kind Regards,
Alex Wang,
On Tue, Aug 27, 2013 at 12:43 AM, Soomyung Park wrote:
> Hello.
>
> I need your help.
>
> There are no problem in configure 2 node vxlan connections.
> But, I have the problem(loop error) in 3 nodes vxlan connections.
>
> ovs'
t; output at any pair of nodes?
Also, could you try setting the 'fail-mode' of bridges to 'secure'? using
"ovs-vsctl set bridge [bridge_name] fail-mode=secure" and see if there is
still loop?
Thanks,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
,
Alex Wang,
On Thu, Aug 29, 2013 at 4:43 PM, Soomyung Park wrote:
>
> 2013. 8. 30., 오전 6:57, Alex Wang 작성:
>
> Hi Park,
>
>
> My answer is your question as below;.
>> My assumption is that 3 nodes are in a same tenant network, so are
>> connected with the full
On Wed, Sep 4, 2013 at 2:19 AM, linaishan wrote:
> **
> Hi all,
> I have a question about ovs with no controller.
> At the beginning, I find the hosts linked to ovs with no
> controller couldn't ping each other, which just like the situation in
> mininet, if you don't set a controller, the host1
sible to achieve full-mesh topology by specifying controller
to 'unicast' all multicast packets to each node in the full-mesh logical
network.
Thanks,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
Hey,
On Wed, Sep 4, 2013 at 6:37 PM, linaishan wrote:
> **
> Thanks Alex.
> The command "ovs-vsctl set bridge br0 fail-mode=secure" works well.
> I tried to find the description of the command in the ovs-vsctl user
> document.
> The result is I don't find the command format you given, but a fo
Hey,
I just tried the lacp on my ubuntu machines (ovs 1.4.2). The balance tcp
works fine.
>
> I initiated *several flows to differents destination* (differents mac /
> IP / subnet) and flows never split over the 2 NICs
> (Yes I am aware that a single flow will use only 1 NIC, the issue is with
Hey,
>From FAQ:
"""
Q: I created a bridge and added my Ethernet port to it, using commands
like these:
ovs-vsctl add-br br0
ovs-vsctl add-port br0 eth0
and as soon as I ran the "add-port" command I lost all connectivity
through eth0. Help!
A: A physical Ethernet device
Hey Nan,
Could you explain more the following points?
But my problem is that I can add ports to br0 while keeping connection
> works, I cannot make bridge to forward the packet to the correct place
>
> like in the bridge I have 4 ports, eth0 to 192.168.55.1, eth1 to
> 192.168.2.2
>
> eth3, eth4 a
Well, as the FAQ says,
"""
A physical Ethernet device that is part of an Open vSwitch bridge
should not have an IP address. If one does, then that IP address
will not be fully functional.
"""
The eth1 on br0 still has address "192.168.2.1".
If you still want eth0 on br0 and eth1 having address
> Nan
>
> On Friday, 18 October, 2013 at 7:15 PM, Alex Wang wrote:
>
> Well, as the FAQ says,
>
> """
> A physical Ethernet device that is part of an Open vSwitch bridge
> should not have an IP address. If one does, then that IP address
> will
sorry for this delayed reply,
there is one thing I noticed below:
On Sat, Oct 19, 2013 at 8:23 AM, Nan wrote:
> Hi, Alex,
>
> I tried your suggestion, it seems that the problem is still there
>
> *In host 1: I have the tunnel ready*
>
> zhunan@zhunan-PowerEdge-R210-II:~$ sudo ovs-vsctl show
>
-tunnels-with-open-vswitch/
Thanks,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
ports which beyond to one bridge.
You can use "mod-port no-flood" to achieve it.
Thanks,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
hanks very much!
> Exactly,I need the function similar with "split-horizon".That is when
> one port receives broadcast packets,then it does not forward these packets
> to other ports.
> In addition, could you give me the example of configure? I am not
> very do at o
On Fri, Nov 1, 2013 at 6:22 AM, Jayanth Silesh wrote:
> Hello All,
>
> I need to insert a code where the OVS is finished initializing and my
> function has to be executed.
> My implementation should be the last thing that has to be executed as a
> Part of "Initialization" of the switch.
>
You m
commands like "ovs-dpctl dump-flows" or "ovs-appctl dpif/dump-flows
[bridge_name]" will show the flow in_port and the datapath actions.
On Thu, Nov 7, 2013 at 12:25 AM, Gopi Krishna B wrote:
> Hi
> Is there a way to identify traffic direction (ingress or egress) at a
> particular port ?
> a)
You can check the ovs-bugtool command,
On Tue, Nov 19, 2013 at 12:47 AM, abhishek jain wrote:
> Hi all,
>
> I need to run OVS in dubug mode.
> CAn you people provide me any method of doing this.
>
>
>
> Thanks
> Abhishek Jain
>
> ___
> discuss mailing
Hey,
Want to try answering your question,
1. If your hardware uses hyperthreading, I recommend do not specify
'n-handler-threads' more than # of real cores.
> For the the upcall dispatcher, currently it's using hash function to find
> a miss-handler, it's some kind of load balance selection f
>
> B.T.W. how did you test the short-lived connections?
>
>
hping3, and netperf TCP_CRR,
> Regards,
> CY.
>
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
Hey Han,
> The miss_handlers are woken up frequently by the dispatcher outside of the
> for (;;) loop, because the for loop breaks when dpif_recv() returns EAGAIN,
> which happens far more frequently than there are 50 pending upcalls for a
> handler. I am trying to figure out a quick but safe sol
)
port 1: br0 (internal)
We'll investigate & fix it.
Thanks,
Alex Wang,
On Sun, Dec 29, 2013 at 11:54 PM, Li, Chen wrote:
> Hi list,
>
>
>
> I’m trying to enable gre/vxlan for Openstack, but unfortunately, I get
> failed. L
>
>
>
> So, I try to
3 09:43:15
[ 172.575046] openvswitch: cannot register gre protocol handler
This will prevent the gre module from functioning. Also, please refer to
the INSTALL file for the correct way to install kernel module.
Thanks,
Alex Wang,
On Mon, Dec 30, 2013 at 9:51 AM, Alex Wang wrote:
> Hey Chen,
&
Sorry for the delayed reply, just back from vacation,
Could you run `modinfo openvswitch` and make sure you have the 1.11 kernel
module installed?
Also, here is a link for installing it in CentOS6.4 :
http://n40lab.wordpress.com/2013/06/03/centos-6-4-openvswitch-installation/
Thanks,
Alex Wang
Hey Ben,
Yeah, this is introduced by my changes. I'm working on a fix.
Alex Wang,
Thx,
On Mon, Apr 21, 2014 at 4:48 PM, Ben Pfaff wrote:
> Yes, I'm seeing these too on current master. I bisected it to the
> following commit:
>
> commit 9a159f748866136a12
saw 60+ flows in datapath that should be
deleted. Did those 'Zombie' flows get hit? Could you provide more info
about how to reproduce it?
Thanks,
Alex Wang,
___
discuss mailing list
discuss@openvswitch.org
http://openvswitch.org/mailman/listinfo/discuss
Sorry for this very delayed reply,
After talking within ovs-team, we think it is best to use branch-2.1
which is LTE branch of OVS. If you still observe the issue, we will
work on a fix.
Thanks,
Alex Wang,
___
discuss mailing list
discuss
Sorry for using the wrong term (LTS, not LTE) and misclaiming it
(branch-2.1
is not a LTS branch).
We just recommend you to use branch-2.1, that branch should work fine.
There is no new LTS branch for ovs.
Thanks,
Alex Wang,
On Mon, Jun 16, 2014 at 5:10 PM, Alex Wang wrote:
> Sorry for t
Hey Han,
We found the issue, fix has been sent here:
http://openvswitch.org/pipermail/dev/2014-June/042333.html
Thanks again for reporting,
Alex Wang,
On Wed, Jun 11, 2014 at 3:54 AM, Han Zhou wrote:
> Hello folks,
>
> We encountered a problem on a hypervisor with OVS2.0, that
can be deleted? If so, I can have
> a workaround without updating OVS version: I can find out the more
> inclusive megaflow and manually delete flows with ovs-dpctl in order
> (A -> B), then warning logs should stop, right?
Right. Note you should prevent the relevant traffic from instal
The tcp field is missing, could you try 'ovs-dpctl dump-flow -m' to get
all fields. And use it as input to 'ovsdpctl flow-del'
Thanks,
Alex Wang,
On Wed, Jul 2, 2014 at 1:22 AM, Han Zhou wrote:
> Hi Alex,
>
> On Wed, Jul 2, 2014 at 2:30 PM, Alex Wang wrote:
I'm not exactly sure about the dp flow you have... but yes, del-flows
will purge all flows.
Also, if you stop to traffic that hitting the 'more inclusive' flow, then,
both flows ('inclusive' and 'phantom') should be deleted.
Thanks,
Alex Wang,
On Thu, Jul 3
Hey Srinivas,
Right now, ovs has only one dpdk polling thread. We are working on
creating multiple polling threads and pinning polling threads to the same
cpu socket as dpdk interface.
Thanks,
Alex Wang,
On Mon, Jul 28, 2014 at 9:15 AM, Ben Pfaff wrote:
> On Mon, Jul 28, 2014 at 07:33:3
Hey Han,
Sorry for this delayed reply,
Please see my comments inline,
On Wed, Jul 30, 2014 at 3:11 AM, Han Zhou wrote:
> Hi Alex,
>
> In our environment this counter subfacet_install_fail increases
> quickly about 100/sec.
>
Could you show me some ovs-vswitchd.log ? there should be warnin
Hey Han,
On Thu, Jul 31, 2014 at 1:33 AM, Han Zhou wrote:
> Hi Alex,
>
> We had to upgrade to OVS2.1, and everything looks much better.
> The flow miss rate decreased from 20K/sec to 1.5K/sec, which means in
> OVS2.0 there was problem of creating datapath flows (indicated by the
> counter subfa
On Mon, Aug 18, 2014 at 1:37 PM, Ben Pfaff wrote:
> Can you provide a backtrace?
>
>
Or can you tell us what is the value of 'wait' in "recvmsg(sock->fd, &msg,
wait ? 0:MSG_DONTWAIT)"?
If it is 'true' then I guess you encountered a bug which should already been
fixed on master.
The fix is one commit after, commit
1738803acda21425c19d1549c0c1e6586ef0c64a
On Mon, Aug 18, 2014 at 1:59 PM, Xu (Simon) Chen wrote:
> I am running "HEAD-6a92c6f".
>
>
> On Mon, Aug 18, 2014 at 4:52 PM, Alex Wang wrote:
>
>>
>>
>>
>>
Hey Chris,
May I know which release are you using?
Did a similar test on my local setup using scapy. The arp entry is
refreshed expectedly with Gratuitous ARP.
Thanks,
Alex Wang,
On Sat, Sep 27, 2014 at 9:54 PM, Chris Bennett wrote:
> Hi Ben,
>
> > If you're not using an Op
If you mean port bonding,
Please check the Port table -> Bonding Configuration, in `man
ovs-vswitchd.conf.db`
On Sun, Sep 28, 2014 at 10:54 AM, Andrew Niteesh <
afrancisxaviertha...@uh.edu> wrote:
> Hi,
>
> Can anyone help, how to do Ether Channel in OVS.
>
> Thanks & Regards,
> Andrew Niteesh
>
until next reboot
> 15 0 fe:dd:3f:1e:71:f61
> 15 0 fe:dd:3f:1e:71:f61
>
> I'm not heavily experienced with OVS to know whether the above is
> 'normal' but it seems strange :)
>
May I know what is port 1 and port 15? (via `ovs-ofctl show
Hey Chris,
On Mon, Sep 29, 2014 at 5:27 PM, Chris Bennett wrote:
> Hi Alex,
>
> > May I know what is port 1 and port 15? (via `ovs-ofctl show
> > br-inet`)
>
> The 'base' is a VM on VMWare btw, maybe that has something to do with
> the weird behaviour?
>
This should not be related.
> [root@
your './vswitchd/ovs-vswitchd' is the one you build
with dpdk? (e.g. checking './vswitchd/ovs-vswitchd --version' to confirm
the
compile time and version number)
Thanks,
Alex Wang,
On Tue, Sep 30, 2014 at 3:56 AM, Vinod, Chegu wrote:
>
>
>
> Hello,
>
>
I have never experienced this before. Is there any warning in `make
install T=x86_64-native-linuxapp-gcc`?
What is your kernel version?
Thanks,
Alex Wang,
On Wed, Oct 1, 2014 at 2:29 PM, MCGRATH, GIL wrote:
> I am following the ovs install instructions for using DPDK and OV
> use
1 - 100 of 118 matches
Mail list logo
