On Thu, 21 Apr 2016, Martin Winter wrote:

It was discussed about changing defaults a few times. And in all cases, none of the choices on defaults violated RFCs.

When it comes to Quagga, the precedent a few times before has been to try handle major behavioural changes in a careful way. Even if those networks are broken according to the RFC (e.g. the RIP auth bug). If they're working, an upgrade shouldn't break them.

So a staged way, with a config knob, default to the existing behaviour and write it out explicitly. A later release can change the default safely then. A network that upgrades through each release will keep working, and the risk of an unsuspecting admin finding routers not talking to older ones correctly after an upgrade or behaving differently is minimised.

And that's fine. I actually agree with it.

I'm happy to do that for this bug too.

As far as I know there's no RFC police who'll knock on our doors and drag us away.

regards,
--
Paul Jakma, HPE Aruba, Advanced Technology Group
Fortune:
Depend on the rabbit's foot if you will, but remember, it didn't help
the rabbit.
                -- R.E. Shay

_______________________________________________
Quagga-dev mailing list
Quagga-dev@lists.quagga.net
https://lists.quagga.net/mailman/listinfo/quagga-dev

Reply via email to