Hi,
I have a use-case and I want to
(1) create vhost-user ports on OVS-DPDK running on a host
(2) in a container a dpdk-based application using virtio-user connecting to
vhost-user ports via sockets.
I looked at following 2 links from Intel which shows how-to configure
vhost-user and
On Thu, Jun 07, 2018 at 04:32:10PM -0700, Han Zhou wrote:
> On Thu, Jun 7, 2018 at 11:15 AM, Ben Pfaff wrote:
> >
> > To review the problem we're trying to solve, ovn-controller uses too
> > much CPU in many circumstances because it recomputes everything on every
> > iteration on the main loop.
- Move conntrack lock out of NAT module
- Use spinlock instead of read/write lock for conntrack entry.
- Update 'ctTotalRelatedEntries' using interlocked functions
- Refactor conntrack code to make it more readable.
Testing:
Evaluated TCP performance using iperf3.
Before optimization:
Native:
To review the problem we're trying to solve, ovn-controller uses too
much CPU in many circumstances because it recomputes everything on every
iteration on the main loop. That includes when it wakes up for any
reason, e.g. in response to a timer, or to an "ovs-appctl" command, or
to respond to an
On Thu, Jun 7, 2018 at 3:38 AM, Jakub Sitnicki wrote:
>
> Add a test that performs typical operations of creating & destroying
> logical routers, switches, ports, address sets and ACLs while checking
> if they trigger full logical flow processing in the ovn-controller.
> This way confirm that
On Thu, Jun 7, 2018 at 3:38 AM, Jakub Sitnicki wrote:
>
> Facilitate checking coverage counters from scripts and tests with a new
> "coverage/read-counter" command that gets the total count for a counter.
>
> Same could be achieved by scraping the output of "coverage/show" command
> but the
On 6/7/2018 9:50 AM, Ben Pfaff wrote:
On Wed, Jun 06, 2018 at 03:23:28PM -0700, Greg Rose wrote:
Before attempting to add a gre tunnel to OVS via the vport gre
kernel interface make sure that the openvswitch kernel module has
been able to grab the gre protocol entry point. If OVS does not
own
On Wed, Jun 06, 2018 at 03:23:28PM -0700, Greg Rose wrote:
> Before attempting to add a gre tunnel to OVS via the vport gre
> kernel interface make sure that the openvswitch kernel module has
> been able to grab the gre protocol entry point. If OVS does not
> own the gre protocol then report
Dear Sir,
Kindly quote your best price for attached requisitions for our June 2018
supply for captioned vessel.
Your prompt action is highly appreciated.
Warmest Regards,
Mr.Bhupendra Chaudhary
Supply Chain Department
GLOBAL STAINLESS STEEL
17, Jagnath estate, Nr. Gujarat Bottling,
On 07/06/2018 13:48, Eelco Chaudron wrote:
> I'm planning on reviewing this patchset, but when I applied the patch to
> master and tried to start OVS it crashed:
>
> #0 eth_compose (b=b@entry=0x7ffebedf7b00, eth_dst=..., eth_src=...,
> eth_type=, size=size@entry=0) at lib/packets.c:965
> #1
On 07/06/18 16:10, Flavio Leitner wrote:
Assume the device is present if it can be opened.
Reported-by: Eelco Chaudron
Signed-off-by: Flavio Leitner
---
lib/netdev-linux.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/netdev-linux.c b/lib/netdev-linux.c
index d19bd86ac..d2b79e569
On 07/06/18 16:10, Flavio Leitner wrote:
If the 'openvswitch' kernel module is not loaded, the API is not
available and the userspace will keep retrying. This approach is
not ideal for the netdev datapath type.
This patch disables network netns support if the error code returned
indicates that
On 07/06/18 16:11, Flavio Leitner wrote:
Tap device is not added to the kernel datapath, so there is
no way to get netns information.
Reported-by: Eelco Chaudron
Signed-off-by: Flavio Leitner
---
lib/netdev-linux.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git
Tap device is not added to the kernel datapath, so there is
no way to get netns information.
Reported-by: Eelco Chaudron
Signed-off-by: Flavio Leitner
---
lib/netdev-linux.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/lib/netdev-linux.c b/lib/netdev-linux.c
index
If the 'openvswitch' kernel module is not loaded, the API is not
available and the userspace will keep retrying. This approach is
not ideal for the netdev datapath type.
This patch disables network netns support if the error code returned
indicates that the API is not available.
Reported-by:
Assume the device is present if it can be opened.
Reported-by: Eelco Chaudron
Signed-off-by: Flavio Leitner
---
lib/netdev-linux.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/netdev-linux.c b/lib/netdev-linux.c
index d19bd86ac..d2b79e569 100644
--- a/lib/netdev-linux.c
+++
Manohar Krishnappa Chidambaraswamy
writes:
> Jan,
>
> Have addressed your comments/suggestions. Please take a look and let me
> know if you have any more comments.
>
> Signed-off-by: Manohar K C
>
> CC: Jan Scheurich
> ---
Hi Jan, and Manohar,
Have you considered making this token bucket
On Fri, 1 Jun 2018 11:23:12 -0700, William Tu wrote:
> Looking at the dpif_netlink_rtnl_probe_oot_tunnels(), since now we
> added ERSPAN feature, instead of probing geneve module,
> we should probe ip_gre module with a nlattr of ERSPAN (ex: HWID).
> If it does not return -ENOSUPPORT, then use the
I'm planning on reviewing this patchset, but when I applied the patch to
master and tried to start OVS it crashed:
#0 eth_compose (b=b@entry=0x7ffebedf7b00, eth_dst=..., eth_src=...,
eth_type=, size=size@entry=0) at lib/packets.c:965
#1 0x0074bfda in flow_compose
Add a test that performs typical operations of creating & destroying
logical routers, switches, ports, address sets and ACLs while checking
if they trigger full logical flow processing in the ovn-controller.
This way confirm that incremental processing is taking effect when we
expect it to.
Place
Facilitate checking coverage counters from scripts and tests with a new
"coverage/read-counter" command that gets the total count for a counter.
Same could be achieved by scraping the output of "coverage/show" command
but the difficulties there are that output is in human readable format
and
(This patch set depends on v3 of the "ovn-controller incremental processing"
patch set [1].)
At the moment we lack tests that confirm that incremental logical flow
processing is working as expected. This patch set aims to address this.
The approach to testing is, in short, to observe if we hit
On Wed, 6 Jun 2018 10:27:23 -0700
Han Zhou wrote:
> On Wed, Jun 6, 2018 at 7:19 AM, Jakub Sitnicki wrote:
> >
> > Add a test that performs typical operations of creating & destroying
> > logical routers, switches, ports, address sets and ACLs while checking
> > if they trigger full logical flow
From: Venkata Anil
We have added new flows for using vlans instead of tunnels for
redirecting VLAN packets to a gateway chassis. This patch documents
these flows in ovn-northd.8.xml and ovn-architecture.7.xml.
Signed-off-by: Venkata Anil
---
v4->v5:
* This patch is added to document the
From: Venkata Anil
External switches should learn the distributed gateway port MAC address
as they have to forward the packet tagged with tenant vlan network but
with this MAC as destination MAC address. So router has to send ARP
reply and gARP for this MAC address through router internal patch
From: Venkata Anil
When a vm on a vlan tenant network sends traffic to an external network,
it is tunneled from host chassis to gateway chassis. In the earlier
discussion [1], Russel (also in his doc [2]) suggested if we can figure
out a way for OVN to do this redirect to the gateway host over a
From: Venkata Anil
This patch avoids tunneling and instead uses source tenant vlan network
across hypervisors for traffic from vlan network on local hypervisor
towards gateway hypervisor hosting redirect chassiss port.
On the local hypervisor, when the packet enters logical router ingress
On 06/06/2018 08:49, Ansis Atteka wrote:
> On Tue, 5 Jun 2018 at 20:32, Ben Pfaff wrote:
>>
>> On Tue, Jun 05, 2018 at 07:48:26PM -0700, Ansis Atteka wrote:
>>> Fix following compilation error when building rpm packages
>>> with rhel/openvswitch.spec file.
>>>
>>> error: Installed (but
Thanks for the additional tests. It might be nice to have one that tests
overlapping fragments, too. Also, it would be great if there could be IPv6
versions of these "corner cases".
Sorry it took so long to review these patches. I'll be faster on v7.
--Justin
> On May 2, 2018, at 12:34
> On Apr 8, 2018, at 7:54 PM, Darrell Ball wrote:
>
> diff --git a/lib/ct-dpif.c b/lib/ct-dpif.c
> index 60c8986..adcf42b 100644
> --- a/lib/ct-dpif.c
> +++ b/lib/ct-dpif.c
> @@ -209,6 +209,30 @@ int ct_dpif_ipf_get_status(struct dpif *dpif, bool
> *ipf_v4_enabled,
> :
30 matches
Mail list logo