On Thu, 21 Apr 2016, David Lamparter wrote:
My review of your patch argues that updating a Quagga-only setup can
transition through the same path that any mixed setup already is on. The
extra complexity of a special-case pure-Quagga network seems to provide
very little gain at a large cost -- especially since the default remains
at non-standard-compliance *and* will be saved as such in the config.
That means that, 5 years from now, some admin will be shooting
themselves in the foot with it just because some Quaggas on their
network have a saved config, and they're then using 6987 for the first
time.
No, because then they must have an upgraded Quagga.
If we include the H-bit support, then new Quagga will unambiguously
support signalling both cases clearly. Also, (missing so far from the
proposal) it should have RFC7770 support to give area wide visibility of
any routers missing that support, and we can just do the right thing.
Again, the H-bit SPF support can and should be independent of what happens
for non-H-bit routers.
From that perspective, Cumulus' previous patch is perfectly fine. If I
haven't Ack'd it before, let me Ack it here.
Great. Very constructive as always.
Deadlock. Nothing goes in.
regards,
--
Paul Jakma, HPE Aruba, Advanced Technology Group
Fortune:
CF&C stole it, fair and square.
-- Tim Hahn
_______________________________________________
Quagga-dev mailing list
[email protected]
https://lists.quagga.net/mailman/listinfo/quagga-dev
