On 6/28/19 2:49 AM, Pablo Neira Ayuso wrote:
> On Mon, Jun 24, 2019 at 11:17:38AM +0800, zhe.he@xxxxxxxxxxxxx wrote:
>> From: He Zhe <zhe.he@xxxxxxxxxxxxx>
>>
>> Since v5.1-rc1, some types of packets do not get unreachable reply with the
>> following iptables setting. Fox example,
>>
>> $ iptables -A INPUT -p icmp --icmp-type 8 -j REJECT
>> $ ping 127.0.0.1 -c 1
>> PING 127.0.0.1 (127.0.0.1) 56(84) bytes of data.
>> — 127.0.0.1 ping statistics —
>> 1 packets transmitted, 0 received, 100% packet loss, time 0ms
>>
>> We should have got the following reply from command line, but we did not.
>> From 127.0.0.1 icmp_seq=1 Destination Port Unreachable
>>
>> Yi Zhao reported it and narrowed it down to:
>> 7fc38225363d ("netfilter: reject: skip csum verification for protocols that don't support it"),
>>
>> This is because nf_ip_checksum still expects pseudo-header protocol type 0 for
>> packets that are of neither TCP or UDP, and thus ICMP packets are mistakenly
>> treated as TCP/UDP.
>>
>> This patch corrects the conditions in nf_ip_checksum and all other places that
>> still call it with protocol 0.
> Looking at 7fc38225363dd8f19e667ad7c77b63bc4a5c065d, I wonder this can
> be fixed while simplifying it...
>
> I think nf_reject_verify_csum() is useless?
>
> In your patch, now you explicitly check for IPPROTO_TCP and
> IPPROTO_UDP to validate the checksum.
Thanks for your review.
I suppose the two main points of 7fc38225363d are valid and I was trying to
align with them and fix them:
1) Skip csum verification for protocols that don't support it.
2) Remove the protocol 0 used to indicate non-TCP/UDP packets, and use actual
types instead to be clear.
1) uses nf_reject_verify_csum to skip those that should be skipped and leaves
the protocols that support csum to the rest of the logic including
nf_ip_checksum. But 2) removes the "0" transition from the rest of the
logic and thus causes this issue. So I add the explicit check against TCP/UDP to
nf_ip_checksum. And nf_reject_verify_csum is still useful.
Zhe
>
