Hope, bug in next hop calculation for stub routes fixed.

This commit is contained in:
Ondrej Filip 2001-08-21 15:00:29 +00:00
parent e43ae6330e
commit e8ab168030

View file

@ -235,7 +235,15 @@ again:
a0.flags=0;
a0.aflags=0;
a0.iface=en->nhi;
if(ipa_in_net(en->nh,en->nhi->addr->ip,en->nhi->addr->pxlen))
{
a0.gw=IPA_NONE;
}
else
{
a0.gw=en->nh;
}
ne=net_get(p->table, nf->fn.prefix, nf->fn.pxlen);
e=rte_get_temp(&a0);
e->u.ospf.metric1=nf->metric;
@ -604,13 +612,12 @@ calc_next_hop(struct top_hash_entry *en, struct top_hash_entry *par,
u32 myrid=p->cf->global->router_id;
DBG(" Next hop called.\n");
if(ipa_to_u32(par->nh)==0)
if(ipa_equal(par->nh,IPA_NONE))
{
neighbor *nn;
DBG(" Next hop calculating for id: %I rt: %I type: %u\n",
en->lsa.id,en->lsa.rt,en->lsa.type);
/* xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx */
if(par==oa->rt)
{
if(en->lsa.type==LSA_T_NET)
@ -637,6 +644,7 @@ calc_next_hop(struct top_hash_entry *en, struct top_hash_entry *par,
{
if((neigh=find_neigh_noifa(po,en->lsa.rt))==NULL) return;
en->nhi=neigh->ifa->iface;
en->nh=neigh->ip; /* Yes, neighbor is it's own next hop */
return;
}
}
@ -651,6 +659,7 @@ calc_next_hop(struct top_hash_entry *en, struct top_hash_entry *par,
else /* Parent is some RT neighbor */
{
/* FIXME: I should probably hold ospf_iface in top_hash_entry */
/* FIXME: Isn't this useless */
neigh=NULL;
WALK_LIST(ifa,po->iface_list)
{