[quagga-dev 7913] Why was passive-interface default removed from RIPNG

Tim Clark olgoat52 at gmail.com
Mon Apr 12 17:19:05 BST 2010

So much of the original ripd_interface.c code was reused for 
ripng_interface.c but for some reason the "passive-interface default" 
capability was removed from the ripng variant.
The vector was changed from non-default representing interfaces that had 
passive/no passive state set explicitly to a vector of  interfaces set 
to passive.

The result of this change is that in RIPNG, every interface you want to 
be passive has to be defined explicitly.  In some configurations, the 
user wants all interfaces to be passive and only one or two to not be.  
If the interfaces you want to be passive are tun interface (started 
dynamically) explicitly defining them is problematic.

Granted the ripd handling for passive interface default was no 100% 
correct in that a no passive-interface default followed by a 
passive-interface default resulted in the loss of information about any 
interface that was explicitly set.  But it was usable in a configuration 
that was not manually changed with respect to passive-interface.

Any thoughts?  I am in the process of hacking the ripd support back into 
ripng for my purposes.

__________ Information from ESET NOD32 Antivirus, version of virus signature database 5021 (20100412) __________

The message was checked by ESET NOD32 Antivirus.


More information about the Quagga-dev mailing list