* Re: "martian source" mixes up source and destination
[not found] <jnh9tl$rth$1@dough.gmane.org>
@ 2012-04-29 4:52 ` Michael Tokarev
0 siblings, 0 replies; only message in thread
From: Michael Tokarev @ 2012-04-29 4:52 UTC (permalink / raw)
To: Mantas M.; +Cc: linux-kernel, netdev
On 28.04.2012 21:37, Mantas M. wrote:
> After enabling the /proc/sys/net/ipv4/conf/all/log_martians sysctl, I've
> been receiving the following messages on dmesg:
Cc'ing netdev@vger for network-related question.
>> martian source 255.255.255.255 from 192.168.1.115, on dev wlan0
>> ll header: ff:ff:ff:ff:ff:ff:f4:6d:04:85:fc:d7:08:00
>> martian source 192.168.1.255 from 192.168.1.115, on dev wlan0
>> ll header: ff:ff:ff:ff:ff:ff:f4:6d:04:85:fc:d7:08:00
>> martian source 192.168.1.255 from 192.168.1.115, on dev wlan0
>> ll header: ff:ff:ff:ff:ff:ff:f4:6d:04:85:fc:d7:08:00
>> <later>
>> martian source 255.255.255.255 from 192.168.1.183, on dev eth0
>> ll header: ff:ff:ff:ff:ff:ff:48:5d:60:e8:65:8f:08:00
It has been this way since very long time. The meaning of this
message is:
there is a martian source sending packets to 255.255.255.255
from 192.168.1.183 on dev eth0
> However, the packets are *not* martians, but legitimate broadcast
> UDP/IPv4 packets: they have the *destination* as 255.255.255.255 or
> 192.168.1.255, with the source being 192.168.1.115 (wlan0 on this
> device) for the first three packets, and 192.168.1.183 (eth0) on the
> fourth.
log_martians does not check destination address, it checks _source_
address. At least it did so in the past. Please show output of
`ip route' from your machine.
> Linux 3.3.4-1-ARCH on Arch Linux
Thanks,
/mjt
^ permalink raw reply [flat|nested] only message in thread