On 4/2/24 16:32, Ales Musil wrote:
> On Tue, Apr 2, 2024 at 4:25 PM Ilya Maximets <[email protected]> wrote:
> 
>> On 4/2/24 16:17, Ales Musil wrote:
>>>
>>>
>>> On Tue, Apr 2, 2024 at 4:03 PM Ilya Maximets <[email protected]
>> <mailto:[email protected]>> wrote:
>>>
>>>     On 4/2/24 15:36, Ales Musil wrote:
>>>     >
>>>     >
>>>     > On Tue, Apr 2, 2024 at 3:02 PM Ilya Maximets <[email protected]
>> <mailto:[email protected]> <mailto:[email protected] <mailto:
>> [email protected]>>> wrote:
>>>     >
>>>     >     On 3/28/24 09:53, Ales Musil wrote:
>>>     >     > The br-int connection is hardcoded to use unix socket, which
>> requires
>>>     >     > for the socket to be visible for ovn-controller. This is
>> achievable in
>>>     >     > container by mounting the socket, but in turn the container
>> requires
>>>     >     > additional privileges.
>>>     >     >
>>>     >     > Add option to ovn-controller that allows to specify remote
>> target for
>>>     >     > br-int. This gives the user possibility to connect to br-int
>> in different
>>>     >     > manner than unix socket, defaulting to the unix socket when
>> not specified.
>>>     >     > In addition, there is an option to specify inactivity probe
>> for this
>>>     >     > connection, disabled by default.
>>>     >     >
>>>     >     > Reported-at: https://issues.redhat.com/browse/FDP-243 <
>> https://issues.redhat.com/browse/FDP-243> <
>> https://issues.redhat.com/browse/FDP-243 <
>> https://issues.redhat.com/browse/FDP-243>>
>>>     >     > Signed-off-by: Ales Musil <[email protected] <mailto:
>> [email protected]> <mailto:[email protected] <mailto:[email protected]>>>
>>>     >     > ---
>>>     >     >  NEWS                            |  6 +++
>>>     >     >  controller/ofctrl.c             | 10 +----
>>>     >     >  controller/ofctrl.h             |  5 ++-
>>>     >     >  controller/ovn-controller.8.xml | 12 ++++++
>>>     >     >  controller/ovn-controller.c     | 68
>> ++++++++++++++++++++++++++-------
>>>     >     >  controller/pinctrl.c            | 56
>> ++++++---------------------
>>>     >     >  controller/pinctrl.h            |  6 ++-
>>>     >     >  controller/statctrl.c           | 66
>> ++++++--------------------------
>>>     >     >  controller/statctrl.h           |  3 +-
>>>     >     >  include/ovn/features.h          |  2 +-
>>>     >     >  lib/features.c                  | 35 +++++------------
>>>     >     >  lib/ovn-util.c                  | 26 +++++++++++++
>>>     >     >  lib/ovn-util.h                  |  4 ++
>>>     >     >  lib/test-ovn-features.c         |  6 +--
>>>     >     >  tests/ovn-controller.at <http://ovn-controller.at> <
>> http://ovn-controller.at <http://ovn-controller.at>>         | 31
>> +++++++++++++++
>>>     >     >  utilities/ovn-ctl               | 10 +++++
>>>     >     >  16 files changed, 192 insertions(+), 154 deletions(-)
>>>     >     >
>>>     >     > diff --git a/NEWS b/NEWS
>>>     >     > index 4d6ebea89..4979bb806 100644
>>>     >     > --- a/NEWS
>>>     >     > +++ b/NEWS
>>>     >     > @@ -12,6 +12,12 @@ Post v24.03.0
>>>     >     >      flow table id.
>>>     >     >      "lflow-stage-to-oftable STAGE_NAME" that converts stage
>> name into OpenFlow
>>>     >     >      table id.
>>>     >     > +  - Add option to ovn-controller called
>> "--br-int-remote=REMOTE" that allows
>>>     >     > +    to specify connection method to integration bridge for
>> ovn-controller,
>>>     >     > +    defaulting to the unix socket.
>>>     >     > +  - Add option to ovn-controller called
>> "--br-int-probe-interval=INTERVAL"
>>>     >     > +    that sets probe interval for integration bridge
>> connection,
>>>     >     > +    disabled by default.
>>>     >
>>>     >     I didn't review the code, but I don't think the names should
>> be changed.
>>>     >     There is nothing user-facing called "br-int" and any bridge
>> name can be
>>>     >     chosen for a deployment.  It should be called
>> "ovn-bridge-mgmt-remote" or
>>>     >     something like that.
>>>     >
>>>     >
>>>     > I'm fine either way, I just went with br-int because the code
>> around that is actually
>>>     > called br-int even if we have different management bridge name.
>>>
>>>     The point is - users do not know how the code looks like
>>>     and they may have nothing called 'br-int' in their setup.
>>>
>>>
>>> Sure that is fine I can change the user facing part.
>>>
>>>
>>>
>>>     Also, maybe these should be a database configuration instead
>>>     of command line arguments?  Alongside the 'ovn-bridge'.
>>>
>>>
>>> So having in DB means we have to react to any change, the question is,
>> do we expect
>>> uset to change this configuration often?
>>
>> Do we react to 'ovn-bridge' changes?  I guess, ovn-controller can react to
>> connection changes the same way.  This configuration should not change
>> frequently, IMO.
>>
> 
> I'll let others chime in, again I don't have a strong preference for this
> one, it can be either of those.
> 

If I may, I vote for storing the OF connection method as external ID in
the OVS database.  I agree with Ilya, the "br-int" bridge name
(ovn-bridge) is already there, it makes the most sense to keep
everything together in one place.

> 
>>
>>> Also "ovs-database" is also just an argument to the ovn-controller.
>>
>> Database connection and OF connection are different things, and we have
>> to know how to connect to the database.  The model until now was to have
>> a single command line argument - database connection method - and store
>> all the other configuration in the database.
>>
>>>
>>>
>>>
>>>     Best regards, Ilya Maximets.
>>>
>>>
>>> Thanks,
>>> Ales
>>> --

Regards,
Dumitru

_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Reply via email to