Commit df4d9254 authored by Hannes Frederic Sowa's avatar Hannes Frederic Sowa Committed by David S. Miller

ipv4: try to cache dst_entries which would cause a redirect

Not caching dst_entries which cause redirects could be exploited by hosts
on the same subnet, causing a severe DoS attack. This effect aggravated
since commit f8864972 ("ipv4: fix dst race in sk_dst_get()").

Lookups causing redirects will be allocated with DST_NOCACHE set which
will force dst_release to free them via RCU.  Unfortunately waiting for
RCU grace period just takes too long, we can end up with >1M dst_entries
waiting to be released and the system will run OOM. rcuos threads cannot
catch up under high softirq load.

Attaching the flag to emit a redirect later on to the specific skb allows
us to cache those dst_entries thus reducing the pressure on allocation
and deallocation.

This issue was discovered by Marcelo Leitner.

Cc: Julian Anastasov <ja@ssi.bg>
Signed-off-by: default avatarMarcelo Leitner <mleitner@redhat.com>
Signed-off-by: default avatarFlorian Westphal <fw@strlen.de>
Signed-off-by: default avatarHannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: default avatarJulian Anastasov <ja@ssi.bg>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent 412d2907
...@@ -39,11 +39,12 @@ struct inet_skb_parm { ...@@ -39,11 +39,12 @@ struct inet_skb_parm {
struct ip_options opt; /* Compiled IP options */ struct ip_options opt; /* Compiled IP options */
unsigned char flags; unsigned char flags;
#define IPSKB_FORWARDED 1 #define IPSKB_FORWARDED BIT(0)
#define IPSKB_XFRM_TUNNEL_SIZE 2 #define IPSKB_XFRM_TUNNEL_SIZE BIT(1)
#define IPSKB_XFRM_TRANSFORMED 4 #define IPSKB_XFRM_TRANSFORMED BIT(2)
#define IPSKB_FRAG_COMPLETE 8 #define IPSKB_FRAG_COMPLETE BIT(3)
#define IPSKB_REROUTED 16 #define IPSKB_REROUTED BIT(4)
#define IPSKB_DOREDIRECT BIT(5)
u16 frag_max_size; u16 frag_max_size;
}; };
......
...@@ -129,7 +129,8 @@ int ip_forward(struct sk_buff *skb) ...@@ -129,7 +129,8 @@ int ip_forward(struct sk_buff *skb)
* We now generate an ICMP HOST REDIRECT giving the route * We now generate an ICMP HOST REDIRECT giving the route
* we calculated. * we calculated.
*/ */
if (rt->rt_flags&RTCF_DOREDIRECT && !opt->srr && !skb_sec_path(skb)) if (IPCB(skb)->flags & IPSKB_DOREDIRECT && !opt->srr &&
!skb_sec_path(skb))
ip_rt_send_redirect(skb); ip_rt_send_redirect(skb);
skb->priority = rt_tos2priority(iph->tos); skb->priority = rt_tos2priority(iph->tos);
......
...@@ -1554,11 +1554,10 @@ static int __mkroute_input(struct sk_buff *skb, ...@@ -1554,11 +1554,10 @@ static int __mkroute_input(struct sk_buff *skb,
do_cache = res->fi && !itag; do_cache = res->fi && !itag;
if (out_dev == in_dev && err && IN_DEV_TX_REDIRECTS(out_dev) && if (out_dev == in_dev && err && IN_DEV_TX_REDIRECTS(out_dev) &&
skb->protocol == htons(ETH_P_IP) &&
(IN_DEV_SHARED_MEDIA(out_dev) || (IN_DEV_SHARED_MEDIA(out_dev) ||
inet_addr_onlink(out_dev, saddr, FIB_RES_GW(*res)))) { inet_addr_onlink(out_dev, saddr, FIB_RES_GW(*res))))
flags |= RTCF_DOREDIRECT; IPCB(skb)->flags |= IPSKB_DOREDIRECT;
do_cache = false;
}
if (skb->protocol != htons(ETH_P_IP)) { if (skb->protocol != htons(ETH_P_IP)) {
/* Not IP (i.e. ARP). Do not create route, if it is /* Not IP (i.e. ARP). Do not create route, if it is
...@@ -2303,6 +2302,8 @@ static int rt_fill_info(struct net *net, __be32 dst, __be32 src, ...@@ -2303,6 +2302,8 @@ static int rt_fill_info(struct net *net, __be32 dst, __be32 src,
r->rtm_flags = (rt->rt_flags & ~0xFFFF) | RTM_F_CLONED; r->rtm_flags = (rt->rt_flags & ~0xFFFF) | RTM_F_CLONED;
if (rt->rt_flags & RTCF_NOTIFY) if (rt->rt_flags & RTCF_NOTIFY)
r->rtm_flags |= RTM_F_NOTIFY; r->rtm_flags |= RTM_F_NOTIFY;
if (IPCB(skb)->flags & IPSKB_DOREDIRECT)
r->rtm_flags |= RTCF_DOREDIRECT;
if (nla_put_be32(skb, RTA_DST, dst)) if (nla_put_be32(skb, RTA_DST, dst))
goto nla_put_failure; goto nla_put_failure;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment