There are already a number of patches out in the wild that rewrite next hop tracking and probably solve this issue. I would suggest that we work on getting one of those in instead of putting this patch in?
donald On Wed, Aug 26, 2015 at 7:34 PM, Sergii Vystoropskyi <[email protected]> wrote: > It's easier to explain with example. Let's assume that there is a > 1.1.1.1/32 route resolved > throught 0.0.0.0/0. In case 0.0.0.0/0 goes away 1.1.1.1/32 should be > removed from kernel as > inactive. But this will not happen as number of nexthops will be equal to > 0. > > * zebra/rt_netlink.c modified check to delete routes with 0 nexthops > --- > zebra/rt_netlink.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/zebra/rt_netlink.c b/zebra/rt_netlink.c > index 33fbc08..ff1f1dd 100644 > --- a/zebra/rt_netlink.c > +++ b/zebra/rt_netlink.c > @@ -1742,7 +1742,7 @@ netlink_route_multipath (int cmd, struct prefix *p, > struct rib *rib, > } > > /* If there is no useful nexthop then return. */ > - if (nexthop_num == 0) > + if (nexthop_num == 0 && cmd != RTM_DELROUTE) > { > if (IS_ZEBRA_DEBUG_KERNEL) > zlog_debug ("netlink_route_multipath(): No useful nexthop."); > -- > 2.1.2.AMZN > > > _______________________________________________ > Quagga-dev mailing list > [email protected] > https://lists.quagga.net/mailman/listinfo/quagga-dev >
_______________________________________________ Quagga-dev mailing list [email protected] https://lists.quagga.net/mailman/listinfo/quagga-dev
