On Thu, Jul 29, 2021 at 3:14 PM Ihar Hrachyshka wrote:
>
> When vlan-passthru is true for LS, for untagged localnet ports, tagged
> VLAN traffic originating from VIFs should be passed through intact since
> the VLAN header belongs to VIF user, not the localnet port fabric.
>
> This patch adds mult
On Wed, Jul 28, 2021 at 8:18 AM Mark Gray wrote:
>
> On 28/07/2021 11:28, Dumitru Ceara wrote:
> > Currently ovn-controller logs (at level INFO) when a logical switch
> > port claimed or released but we have no default logs about when its
> > corresponding openflows have been successfully installe
On Fri, Jul 30, 2021 at 10:42 AM Dumitru Ceara wrote:
>
> Logical_Flows are immutable, that is, when the match or action change
> ovn-northd will actually remove the old logical flow and add a new one.
>
> Commit ecc941c70f16 added flows to flood ARPs to routers for unreachable
> addresses.
>
> In
On Mon, Jul 26, 2021 at 4:38 PM Numan Siddique wrote:
>
> On Mon, Jul 26, 2021 at 10:01 AM Dumitru Ceara wrote:
> >
> > Commit 6aab704fb67d ("controller: Avoid unnecessary load balancer flow
> > processing.") handled the case when load balancers are updated but
> > addition/removal of load balanc
On Fri, Jul 30, 2021 at 8:13 AM Lorenzo Bianconi
wrote:
>
> Signed-off-by: Lorenzo Bianconi
Thanks. Applied.
Numan
> ---
> NEWS | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/NEWS b/NEWS
> index eefdae9bc..8b5c52dda 100644
> --- a/NEWS
> +++ b/NEWS
> @@ -1,6 +1,8 @@
> Post-v21.06
On Tue, Jul 27, 2021 at 7:11 PM Han Zhou wrote:
>
> On Tue, Jul 27, 2021 at 3:03 PM Numan Siddique wrote:
> >
> > On Tue, Jul 27, 2021 at 4:00 PM Han Zhou wrote:
> > >
> > > On Tue, Jul 27, 2021 at 11:43 AM Numan Siddique wrote:
> > > >
> > > > On Tue, Jul 27, 2021 at 3:36 AM Han Zhou wrote:
>
On Thu, Jul 29, 2021 at 10:51 AM Priyankar Jain
wrote:
>
> Allow adding label to an ACL to identify which ACL allowed a particular
> flow in the connection tracking table.
>
> The ACL label covers 32 bits at the end of ct_label. Since only allowed
> connections are committed, only "allow" and "all
On 7/29/21 10:52 PM, Mark Michelson wrote:
> I'm fine with this tradeoff. As was stated by someone else, this is
> relieving load on many ovn-controllers to instead slightly load
> ovn-northd more. As you've noted, there are potential additional
> optimizations we could make in ovn-northd to offset
Logical_Flows are immutable, that is, when the match or action change
ovn-northd will actually remove the old logical flow and add a new one.
Commit ecc941c70f16 added flows to flood ARPs to routers for unreachable
addresses.
In the following topology (2 switches connected to a router with a load
I finally got a chance to look through the code.
Acked-by: Mark Michelson
On 7/29/21 5:59 AM, Dumitru Ceara wrote:
Logical_Flows are immutable, that is, when the match or action change
ovn-northd will actually remove the old logical flow and add a new one.
Commit ecc941c70f16 added flows to f
Signed-off-by: Lorenzo Bianconi
---
NEWS | 2 ++
1 file changed, 2 insertions(+)
diff --git a/NEWS b/NEWS
index eefdae9bc..8b5c52dda 100644
--- a/NEWS
+++ b/NEWS
@@ -1,6 +1,8 @@
Post-v21.06.0
-
- Added Control Plane Protection support (control plane traffic metering)
For a fully distributed virtual network dataplane, ovn-controller
flood-fills datapaths that are connected through patch ports. This
creates scale problems in ovn-controller when the connected datapaths
are too many.
In a particular situation, when distributed gateway ports are used to
connect log
Move the logical of handling patch port changes for updating local_datapaths
to a separate function, which will be reused later. Also split the
switch-case for the 3 cases: LP_PATCH, LP_LOCALPORT and LP_VTEP, because
there are more differences than common, to avoid the extra if-blocks.
Signed-off-
The extra get_local_datapath() call seems unnecessary. Remove it and
combine the two if-blocks to an if-else block.
Signed-off-by: Han Zhou
---
controller/binding.c | 13 -
1 file changed, 4 insertions(+), 9 deletions(-)
diff --git a/controller/binding.c b/controller/binding.c
index
Note: This patch series is on top of a pending patch that is still under
review:
http://patchwork.ozlabs.org/project/ovn/patch/20210729080218.1235041-1-hz...@ovn.org/
It is RFC because: a) it is based on the unmerged patch. b) DDlog
changes are not done yet. Below is a copy of the commit message
When there is a big number of ha_chassis_groups (e.g. for distributed
gateway ports), the calculation of ha_ref_chassis can take the major
part of ovn-northd CPU as shown in perf.
However, when there is only one chassis in ha_chassis_group, no BFD
sessions are needed, so ha_ref_chassis calculation
16 matches
Mail list logo
