[PATCH v3 nf] netfilter: seqadj: Fix one possible panic in seqadj when mem is exhausted

[PATCH v3 nf] netfilter: seqadj: Fix one possible panic in seqadj when mem is exhausted

[fgao]

From: Gao Feng <fgao@ikuai8.com>

When memory is exhausted, nfct_seqadj_ext_add may fail to add the seqadj
extension. But the function nf_ct_seqadj_init doesn't check if get valid
seqadj pointer by the nfct_seqadj, while other functions perform the
sanity check.

So the system would be panic when nfct_seqadj_ext_add failed.

Signed-off-by: Gao Feng <fgao@ikuai8.com>
---
 v3: Remove the warning log when seqadj is null;
 v2: Remove the unnessary seqadj check in nf_ct_seq_adjust
 v1: Initial patch

 net/netfilter/nf_conntrack_seqadj.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/net/netfilter/nf_conntrack_seqadj.c b/net/netfilter/nf_conntrack_seqadj.c
index dff0f0c..7f8d814 100644
--- a/net/netfilter/nf_conntrack_seqadj.c
+++ b/net/netfilter/nf_conntrack_seqadj.c
@@ -16,9 +16,12 @@ int nf_ct_seqadj_init(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
 	if (off == 0)
 		return 0;
 
+	seqadj = nfct_seqadj(ct);
+	if (unlikely(!seqadj))
+		return 0;
+
 	set_bit(IPS_SEQ_ADJUST_BIT, &ct->status);
 
-	seqadj = nfct_seqadj(ct);
 	this_way = &seqadj->seq[dir];
 	this_way->offset_before	 = off;
 	this_way->offset_after	 = off;
-- 
1.9.1

Re: [PATCH v3 nf] netfilter: seqadj: Fix one possible panic in seqadj when mem is exhausted

[Pablo Neira Ayuso]

On Sat, Sep 03, 2016 at 07:51:50PM +0800, fgao@ikuai8.com wrote:
> From: Gao Feng <fgao@ikuai8.com>
> 
> When memory is exhausted, nfct_seqadj_ext_add may fail to add the seqadj
> extension. But the function nf_ct_seqadj_init doesn't check if get valid
> seqadj pointer by the nfct_seqadj, while other functions perform the
> sanity check.
> 
> So the system would be panic when nfct_seqadj_ext_add failed.
> 
> Signed-off-by: Gao Feng <fgao@ikuai8.com>
> ---
>  v3: Remove the warning log when seqadj is null;
>  v2: Remove the unnessary seqadj check in nf_ct_seq_adjust
>  v1: Initial patch
> 
>  net/netfilter/nf_conntrack_seqadj.c | 5 ++++-
>  1 file changed, 4 insertions(+), 1 deletion(-)
> 
> diff --git a/net/netfilter/nf_conntrack_seqadj.c b/net/netfilter/nf_conntrack_seqadj.c
> index dff0f0c..7f8d814 100644
> --- a/net/netfilter/nf_conntrack_seqadj.c
> +++ b/net/netfilter/nf_conntrack_seqadj.c
> @@ -16,9 +16,12 @@ int nf_ct_seqadj_init(struct nf_conn *ct, enum ip_conntrack_info ctinfo,
>  	if (off == 0)
>  		return 0;
>  
> +	seqadj = nfct_seqadj(ct);
> +	if (unlikely(!seqadj))
> +		return 0;

I think we should handle this from init_conntrack() by simply dropping
the packet as we do under similar circunstances (too stress to deal).