Le 03/06/2015 19:40, Paul Jakma a écrit :
On Wed, 3 Jun 2015, Paul Jakma wrote:
At present, we are not clear on the design because several people seem to
think the ZServ API added by this patch-set adds still is compatible with
multiple-processes.
Oh, and this isn't about VRFs and zebra generally. I'm fine with that.
This is very specifically about the static, unnegotiated VRF ID added to the
header that allows a client to expect that it can send any command it wants to
the remote zebra for whatever VRF context and expect that the remote zebra must
be able to deal with it.
I'm sorry, but I don't understand your concern. All API, whatever the product
is, have a scope where this API is valid. In that case (no matter we are
playing with multiple daemons or single dameon model), the VRFID is not a
random number but should be valid, ie an existing VRFID.
There is a hook which allows to add/remove/check this VRFID, thus the scope is
cleary defined.
If the user tries to play with a wrong VRFID, for me it's simply a bug or a
configuration problem. And if something else is needed to ensure that
negociation, we can add it later. I really don't show why you think this series
is blocker.
Note that alongside that discussion, I will send the v5 with Donald's comments
addressed.
Regards,
Nicolas
_______________________________________________
Quagga-dev mailing list
[email protected]
https://lists.quagga.net/mailman/listinfo/quagga-dev
