ICMP and conntrack

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



I do not know if this issue has been raised yet.

On a TCP or UDP protocol, if there are communication problems a ICMP
"responses" are generated.
Does conntrack handle all ICMP "responses" correctly?

The responses that I can think of are:
3 Destination Unreachable
11 Time to live Exceeded
4 Source Quench

The reason why I am asking is that I have a rules such as:

-A INPUT   -m state --state ESTABLISHED,RELATED -j ACCEPT
-A OUTPUT  -m state --state ESTABLISHED,RELATED -j ACCEPT
-A FORWARD -m state --state ESTABLISHED,RELATED -j ACCEPT

....

-A OUTPUT  -j RULE_DROP 
-A INPUT  -j RULE_DROP 
-A FORWARD  -j RULE_DROP 
-A RULE_DROP   -m limit --limit 5/second -j LOG  --log-level info
--log-prefix "RULE DROP -- "
-A RULE_DROP  -j DROP 

And I still see ICMP types which should be handled by conntrack in the log
file as, such as:

May 16 12:04:23 fw RULE DROP -- IN=eth0 OUT=eth1 SRC=68.121.146.134
DST=62.53.192.175 LEN=56 TOS=0x00 PREC=0xE0 TTL=42 ID=44796 PROTO=ICMP
TYPE=3 CODE=3 [SRC=62.53.192.175 DST=68.121.146.134 LEN=85 TOS=0x00
PREC=0x00 TTL=98 ID=47689 PROTO=UDP SPT=45585 DPT=65000 LEN=65 ]
May 16 12:07:12 fw RULE DROP -- IN=eth0 OUT=eth1 SRC=68.0.95.18
DST=62.53.192.175 LEN=76 TOS=0x00 PREC=0xE0 TTL=49 ID=65337 PROTO=ICMP
TYPE=3 CODE=1 [SRC=62.53.192.175 DST=192.168.1.104 LEN=48 TOS=0x00 PREC=0x80
TTL=110 ID=52247 DF PROTO=TCP SPT=45783 DPT=6881 WINDOW=16384 RES=0x00 SYN
URGP=0 ]
May 16 12:07:20 fw RULE DROP -- IN=eth0 OUT=eth1 SRC=10.49.4.6
DST=62.53.192.143 LEN=56 TOS=0x00 PREC=0xE0 TTL=230 ID=47467 PROTO=ICMP
TYPE=11 CODE=0 [SRC=62.53.192.143 DST=219.100.31.227 LEN=40 TOS=0x00
PREC=0xE0 TTL=1 ID=54789 DF PROTO=TCP INCOMPLETE [8 bytes] ]

Shouldn't these messages be handled by conntrack as "RELATED" or
"ESTABLISHED" and be let through the firewall and therefore not be logged
(i.e. dropped)?
Can anybody help or explain?

Clemente




[Index of Archives]     [Linux Netfilter Development]     [Linux Kernel Networking Development]     [Netem]     [Berkeley Packet Filter]     [Linux Kernel Development]     [Advanced Routing & Traffice Control]     [Bugtraq]

  Powered by Linux