[ovs-dev] [PATCH 2/5] vtep: Initial checkin of vtep schema.

Justin Pettit jpettit at nicira.com
Thu Oct 17 21:36:45 UTC 2013


On Oct 17, 2013, at 9:15 AM, Ben Pfaff <blp at nicira.com> wrote:

> On Fri, Oct 11, 2013 at 04:56:23PM -0700, Justin Pettit wrote:
>> From: Bruce Davie <bdavie at vmware.com>
>> 
>> The hardware VTEP OVSDB schema specifies relations that a VTEP can use
>> to integrate physical ports into logical switches maintained by a
>> network virtualization controller such as NVP.
>> 
>> Co-authored-by: Ben Pfaff <blp at nicira.com>
>> Co-authored-by: Kenneth Duda <kduda at aristanetworks.com>
>> Co-authored-by: Justin Pettit <jpettit at nicira.com>
>> Signed-off-by: Justin Pettit <jpettit at nicira.com>
> 
> This commit and change log has a few references to NVP.  I think that
> NVP is now marketed under the name NSX, so perhaps we should update the
> references.

Good point.  Updated.  (Also, I updated another reference added in the later ovs-vtep patch.)

> This week we changed how we generate the E-R diagrams for the vswitchd
> schema.  Can you update the commit to use the same pattern for the vtep
> schema?  (That means that the commit would no longer include vtep.gv or
> vtep.pic.)

Ansis was mistakenly locked out of the repo, so he hasn't been able to push his changes.  I thought I'd be able to slide by on a technicality.  Shoot.  Okay, I'll send out v2 momentarily.

> vtep.xml still talks about port 6632 as the default OVSDB port.  Do we
> have a plan for transitioning vtep to use the IETF standardized port?

That's a good point.  I'll bring it up with Bruce and others involved in this.  Unless you feel passionately, I'm leaving it as-is, though, since this is what the existing implementors are currently using, and we'll need to come up with some sort of transition plan.

> vtep.xml has a few references to Open vSwitch (rather that to the Open
> vSwitch database or database server) that seem a little out of line,
> since VTEP doesn't involve Open vSwitch proper.


I fixed those to be "database server".  (I also removed some description of OVS's implementation of BFD, which is not relevant.)

--Justin





More information about the dev mailing list