Trouble with IPFW or TCP?

Ivan Voras ivoras at freebsd.org
Thu Apr 3 23:34:36 UTC 2008


In which case would an ipfw ruleset like this:

00100 114872026  40487887607 allow ip from any to any via lo0
00200         0            0 deny ip from any to 127.0.0.0/8
00300         0            0 deny ip from 127.0.0.0/8 to any
00600      1585       112576 deny ip from table(0) to me
01000     90279      7325972 allow icmp from any to any
05000 475961039 334422494257 allow tcp from me to any setup keep-state
05100    634155     65779377 allow udp from me to any keep-state
06022    409604     69177326 allow tcp from any to me dst-port 22 setup 
keep-state
06080  52159025  43182548092 allow tcp from any to me dst-port 80 setup 
keep-state
06443   6392366   2043532158 allow tcp from any to me dst-port 443 setup 
keep-state
07020    517065    292377553 allow tcp from any to me dst-port 8080 
setup keep-state
65400  12273387    629703212 deny log ip from any to any
65535         0            0 deny ip from any to any

Generate syslog messages like these:

Apr  4 01:02:06 my.ip kernel: ipfw: 65400 Deny TCP xx.xx.xx.xx:60725 
my.ip.my.ip:443 in via em0
Apr  4 01:02:06 my.ip kernel: ipfw: 65400 Deny TCP xx.xx.xx.xx:57387 
my.ip.my.ip:443 in via em0
Apr  4 01:02:06 my.ip kernel: ipfw: 65400 Deny TCP xx.xx.xx.xx:57387 
my.ip.my.ip:443 in via em0
Apr  4 01:02:06 my.ip kernel: ipfw: 65400 Deny TCP xx.xx.xx.xx:61998 
my.ip.my.ip:443 in via em0
Apr  4 01:02:06 my.ip kernel: ipfw: 65400 Deny TCP xx.xx.xx.xx:61998 
my.ip.my.ip:443 in via em0
Apr  4 01:02:06 my.ip kernel: ipfw: 65400 Deny TCP xx.xx.xx.xx:64288 
my.ip.my.ip:443 in via em0
Apr  4 01:02:06 my.ip kernel: ipfw: 65400 Deny TCP xx.xx.xx.xx:64288 
my.ip.my.ip:443 in via em0
Apr  4 01:02:06 my.ip kernel: ipfw: 65400 Deny TCP xx.xx.xx.xx:50212 
my.ip.my.ip:443 in via em0
Apr  4 01:02:06 my.ip kernel: ipfw: 65400 Deny TCP xx.xx.xx.xx:50212 
my.ip.my.ip:443 in via em0
Apr  4 01:02:06 my.ip kernel: ipfw: 65400 Deny TCP xx.xx.xx.xx:58149 
my.ip.my.ip:443 in via em0
Apr  4 01:02:06 my.ip kernel: ipfw: 65400 Deny TCP xx.xx.xx.xx:58149 
my.ip.my.ip:443 in via em0
Apr  4 01:02:13 my.ip kernel: ipfw: 65400 Deny TCP yy.yy.yy.yy:61919 
my.ip.my.ip:443 in via em0
Apr  4 01:02:13 my.ip kernel: ipfw: 65400 Deny TCP yy.yy.yy.yy:61919 
my.ip.my.ip:443 in via em0
Apr  4 01:02:13 my.ip kernel: ipfw: 65400 Deny TCP yy.yy.yy.yy:56792 
my.ip.my.ip:443 in via em0
Apr  4 01:02:13 my.ip kernel: ipfw: 65400 Deny TCP yy.yy.yy.yy:56792 
my.ip.my.ip:443 in via em0
Apr  4 01:02:14 my.ip kernel: ipfw: 65400 Deny TCP yy.yy.yy.yy:53795 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:58314 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:63204 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:58314 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:52125 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:53386 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:63626 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:63204 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:51376 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:61880 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:49319 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:52125 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:62381 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:53386 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:63626 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:51376 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:54109 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:56945 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:61880 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:50800 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:49319 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:53347 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:58735 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:58732 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:62381 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:56837 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:54109 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:65318 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:56945 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:50800 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:53347 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:58735 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:58732 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:56837 
my.ip.my.ip:443 in via em0
Apr  4 01:02:16 my.ip kernel: ipfw: 65400 Deny TCP zz.zz.zz.zz:65318 
my.ip.my.ip:443 in via em0

?

I can connect with plain telnet from the reported addresses without 
problems. One thing that is suspicious is that the messages come in 
these bursts (which I can't explain) but the Apache's listen backlog 
should handle those. In any case, I don't think they are connection 
requests:

Here's output of "tcpdump -v host xx.xx.xx.xx and port 443":

01:13:07.654677 IP (tos 0x0, ttl 58, id 54089, offset 0, flags [none], 
proto TCP (6), length 40) xx.xx.xx.xx.58789 > my.ip.my.ip.https: ., 
cksum 0x54ca (correct), ack 3708282724 win 0
01:13:07.654764 IP (tos 0x0, ttl 58, id 54095, offset 0, flags [none], 
proto TCP (6), length 40) xx.xx.xx.xx.61579 > my.ip.my.ip.https: ., 
cksum 0xf60d (correct), ack 610863831 win 0
01:13:07.654810 IP (tos 0x0, ttl 58, id 54099, offset 0, flags [none], 
proto TCP (6), length 40) xx.xx.xx.xx.61852 > my.ip.my.ip.https: ., 
cksum 0xab18 (correct), ack 1491048554 win 0
01:13:07.654854 IP (tos 0x0, ttl 58, id 54103, offset 0, flags [none], 
proto TCP (6), length 40) xx.xx.xx.xx.63950 > my.ip.my.ip.https: ., 
cksum 0x1e51 (correct), ack 2955921131 win 0
01:13:07.654897 IP (tos 0x0, ttl 58, id 54107, offset 0, flags [none], 
proto TCP (6), length 40) xx.xx.xx.xx.53299 > my.ip.my.ip.https: ., 
cksum 0xa141 (correct), ack 2339864417 win 0
01:13:07.654940 IP (tos 0x0, ttl 58, id 54121, offset 0, flags [none], 
proto TCP (6), length 40) xx.xx.xx.xx.50521 > my.ip.my.ip.https: ., 
cksum 0x2c55 (correct), ack 216576745 win 0
01:13:07.654984 IP (tos 0x0, ttl 58, id 54123, offset 0, flags [DF], 
proto TCP (6), length 52) xx.xx.xx.xx.58789 > my.ip.my.ip.https: ., 
cksum 0x882d (correct), ack 1 win 33304 <nop,nop,timestamp 140692997 
4077078528>
01:13:07.655026 IP (tos 0x0, ttl 58, id 54126, offset 0, flags [DF], 
proto TCP (6), length 52) xx.xx.xx.xx.61579 > my.ip.my.ip.https: ., 
cksum 0x0617 (correct), ack 1 win 33304 <nop,nop,timestamp 140692997 
3172245833>
01:13:07.655069 IP (tos 0x0, ttl 58, id 54128, offset 0, flags [DF], 
proto TCP (6), length 52) xx.xx.xx.xx.61852 > my.ip.my.ip.https: ., 
cksum 0x7006 (correct), ack 1 win 33304 <nop,nop,timestamp 140692997 
3472415360>
01:13:07.655112 IP (tos 0x0, ttl 58, id 54130, offset 0, flags [DF], 
proto TCP (6), length 52) xx.xx.xx.xx.63950 > my.ip.my.ip.https: ., 
cksum 0x7ade (correct), ack 1 win 33304 <nop,nop,timestamp 140692997 
415365400>
01:13:07.655155 IP (tos 0x0, ttl 58, id 54132, offset 0, flags [DF], 
proto TCP (6), length 52) xx.xx.xx.xx.53299 > my.ip.my.ip.https: ., 
cksum 0x4087 (correct), ack 1 win 33304 <nop,nop,timestamp 140692997 
2999393370>
01:13:07.655197 IP (tos 0x0, ttl 58, id 54139, offset 0, flags [DF], 
proto TCP (6), length 52) xx.xx.xx.xx.50521 > my.ip.my.ip.https: ., 
cksum 0x13e0 (correct), ack 1 win 33304 <nop,nop,timestamp 140692997 
3427580559>

There are no SYNs here, so it looks to me like something mid-traffic.

For addresses such as those in the ipfw log, I see several messages like:

Mar 24 17:00:01 my.ip kernel: TCP: [xx.xx.xx.xx]:64714 to 
[my.ip.my.ip]:443 tcpflags 0x18<PUSH,ACK>; tcp_do_segment: FIN_WAIT_2: 
Received 198 bytes of data after socket was closed, sending RST and 
removing tcpcb
Mar 24 17:10:57 my.ip kernel: 6110>ipfw: 65400 Deny TCP 
xx.xx.xx.xx:58213 my.ip.my.ip:443 in via em0
Mar 25 00:00:05 my.ip kernel: TCP: [xx.xx.xx.xx]:52233 to 
[my.ip.my.ip]:443 tcpflags 0x10<ACK>; syncache_expand: Segment failed 
SYNCOOKIE authentication, segment rejected (probably spoofed)
Mar 25 01:45:05 my.ip kernel: TCP: [xx.xx.xx.xx]:51120 to 
[my.ip.my.ip]:443 tcpflags 0x10<ACK>; syncache_expand: Segment failed 
SYNCOOKIE authentication, segment rejected (probably spoofed)
Mar 25 01:45:05 my.ip kernel: TCP: [xx.xx.xx.xx]:51120 to 
[my.ip.my.ip]:443 tcpflags 0x10<ACK>; syncache_expand: Segment failed 
SYNCOOKIE authentication, segment rejected (probably spoofed)
Mar 25 14:00:09 my.ip kernel: TCP: [xx.xx.xx.xx]:55665 to 
[my.ip.my.ip]:443 tcpflags 0x10<ACK>; syncache_expand: Segment failed 
SYNCOOKIE authentication, segment rejected (probably spoofed)
Mar 25 14:00:09 my.ip kernel: TCP: [xx.xx.xx.xx]:55665 to 
[my.ip.my.ip]:443 tcpflags 0x10<ACK>; syncache_expand: Segment failed 
SYNCOOKIE authentication, segment rejected (probably spoofed)
Mar 25 17:00:00 my.ip kernel: TCP: [xx.xx.xx.xx]:60048 to 
[my.ip.my.ip]:443 tcpflags 0x18<PUSH,ACK>; tcp_do_segment: FIN_WAIT_1: 
Received 346 bytes of data after socket was closed, sending RST and 
removing tcpcb
Mar 25 17:00:00 my.ip kernel: TCP: [xx.xx.xx.xx]:60048 to 
[my.ip.my.ip]:443 tcpflags 0x11<FIN,ACK>; syncache_expand: Segment 
failed SYNCOOKIE authentication, segment rejected (probably spoofed)
Mar 25 23:00:01 my.ip kernel: TCP: [xx.xx.xx.xx]:50205 to 
[my.ip.my.ip]:443 tcpflags 0x18<PUSH,ACK>; tcp_do_segment: FIN_WAIT_1: 
Received 346 bytes of data after socket was closed, sending RST and 
removing tcpcb
Mar 25 23:00:01 my.ip kernel: TCP: [xx.xx.xx.xx]:50205 to 
[my.ip.my.ip]:443 tcpflags 0x11<FIN,ACK>; syncache_expand: Segment 
failed SYNCOOKIE authentication, segment rejected (probably spoofed)
Mar 26 00:38:49 my.ip kernel: <a10>ipfw: 65400 Deny TCP 
xx.xx.xx.xx:61129 my.ip.my.ip:443 in via em0
Apr  2 09:00:01 my.ip kernel: TCP: [xx.xx.xx.xx]:57248 to 
[my.ip.my.ip]:443 tcpflags 0x18<PUSH,ACK>; tcp_do_segment: FIN_WAIT_1: 
Received 330 bytes of data after socket was closed, sending RST and 
removing tcpcb
Apr  2 09:00:01 my.ip kernel: TCP: [xx.xx.xx.xx]:57248 to 
[my.ip.my.ip]:443 tcpflags 0x11<FIN,ACK>; syncache_expand: Segment 
failed SYNCOOKIE authentication, segment rejected (probably spoofed)

But these messages do *not* occur when the ipfw log and tcpdump record 
the above described behaviour so it might not be connected.

The machine at my.ip is running 7.0-RELEASE i386, the rest are a set of 
machines that send trivial periodic (every 15 minutes) HTTPS messages to 
this machine.

In this set most are 6.2 or 6.3, mixed i386 and amd64. The one 7-STABLE 
machine that does the same thing doesn't generate the behaviour 
described above so it might be something specific to when 6.x machines 
talk to 7.x. Was there a bug like that in 6.x?


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 250 bytes
Desc: OpenPGP digital signature
Url : http://lists.freebsd.org/pipermail/freebsd-net/attachments/20080403/833c780d/signature.pgp


More information about the freebsd-net mailing list