LVS
lvs-devel
Google
 
Web LinuxVirtualServer.org

Re: [PATCH] ipvs does not decrement IP ttl

To: Dwip Banerjee <dwip@xxxxxxxxxxxxxxxxxx>
Subject: Re: [PATCH] ipvs does not decrement IP ttl
Cc: lvs-devel@xxxxxxxxxxxxxxx
From: Julian Anastasov <ja@xxxxxx>
Date: Mon, 24 Oct 2016 22:19:13 +0300 (EEST)
        Hello,

On Mon, 24 Oct 2016, Dwip Banerjee wrote:

> We decrement the IP ttl in all the modes in order to prevent infinite
> route loops. The changes were done based on Julian Anastasov's
> suggestions in a prior thread.
> 
> The ttl based check/discard and the actual decrement are done in
> __ip_vs_get_out_rt() and in __ip_vs_get_out_rt_v6(), for the IPv6
> case. decrement_ttl() implements the actual functionality for the
> two cases.
> 
> 
> Signed-off-by: Dwip Banerjee <dwip@xxxxxxxxxxxxxxxxxx>
> ---
>  net/netfilter/ipvs/ip_vs_xmit.c |   55 
> +++++++++++++++++++++++++++++++++++++++
>  1 files changed, 55 insertions(+), 0 deletions(-)
> 
> diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c
> index 01d3d89..8b11e92 100644
> --- a/net/netfilter/ipvs/ip_vs_xmit.c
> +++ b/net/netfilter/ipvs/ip_vs_xmit.c
> @@ -254,6 +254,55 @@ static inline bool ensure_mtu_is_adequate(struct 
> netns_ipvs *ipvs, int skb_af,
>       return true;
>  }
>  
> +static inline bool decrement_ttl(int skb_af, struct sk_buff *skb)
> +{
> +#ifdef CONFIG_IP_VS_IPV6
> +     if (skb_af == AF_INET6) {
> +             struct ipv6hdr *hdr = ipv6_hdr(skb);
> +             struct dst_entry *dst = skb_dst(skb);
> +             struct net *net = dev_net(dst->dev);

        dev_net should not be used anymore, you can check
the latest changes by Eric W. Biederman:

git log -p -- net/netfilter/ipvs/ip_vs_xmit.c

        Better to use ipvs->net.

> +
> +             /* check and decrement ttl */
> +             if (hdr->hop_limit <= 1) {
> +                     /* Force OUTPUT device used as source address */
> +                     skb->dev = dst->dev;
> +                     icmpv6_send(skb, ICMPV6_TIME_EXCEED,
> +                                      ICMPV6_EXC_HOPLIMIT, 0);
> +                     __IP6_INC_STATS(net, ip6_dst_idev(dst),
> +                                     IPSTATS_MIB_INHDRERRORS);
> +
> +                     return false;
> +             }
> +
> +             /* don't propagate ttl change to cloned packets */
> +             if (!skb_make_writable(skb, sizeof(struct ipv6hdr)))
> +                     return false;

        Now I see that using hdr after skb_make_writable is
risky because skb data can be reallocated. Better to use
ip[v6]_hdr(skb) for all places, i.e. without any hdr/iph ptrs.

> +
> +             hdr->hop_limit--;
> +     } else
> +#endif
> +     {
> +             struct iphdr  *iph = ip_hdr(skb);
> +
> +             if (iph->ttl <= 1) {
> +                     /* Tell the sender its packet died... */
> +                     __IP_INC_STATS(dev_net(skb_dst(skb)->dev),
> +                                    IPSTATS_MIB_INHDRERRORS);
> +                     icmp_send(skb, ICMP_TIME_EXCEEDED, ICMP_EXC_TTL, 0);
> +                     return false;
> +             }
> +
> +             /* don't propagate ttl change to cloned packets */
> +             if (!skb_make_writable(skb, sizeof(struct iphdr)))
> +                     return false;
> +
> +             /* Decrease ttl */
> +             ip_decrease_ttl(iph);
> +     }
> +
> +     return true;
> +}
> +

Regards

--
Julian Anastasov <ja@xxxxxx>
--
To unsubscribe from this list: send the line "unsubscribe lvs-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html

<Prev in Thread] Current Thread [Next in Thread>