[Snort-users] Everything being triggered as 1:486:4 ICMP unreachable

Todd Wease twease at ...1935...
Wed May 30 08:15:29 EDT 2007


David Ryan wrote:
>
> All,
>
>  
>
> To partly reply to my own question - one of my flaws was keeping
> $HOME_NET as 10.0.0.0/8 - this prevented most of my test traffic from
> triggering since my tests were coming from $HOME_NET and not
> $EXTERNAL_NET.  This was significant where the rule I was testing was
> of the form $EXTERNAL_NET -> $HOME_NET
>
>  
>
> I'm still getting the ICMP messages as well, so I have to presume that
> those packets are actually out there also, but I don't understand why
> they are being generated since as observeved from the fact that I can
> SSH and ping the host it is not actually unreachable.
>
>  
>
> David
>
>  
>
> *From:* David Ryan
> *Sent:* 30 May 2007 11:10
> *To:* 'snort-users at lists.sourceforge.net'
> *Subject:* Everything being triggered as 1:486:4 ICMP unreachable
>
>  
>
> Hi all,
>
>  
>
> I built and tested a number of machines and shipped them out to remote
> sites recently.  Now that they are at the remote site I am trying to
> tweak them a bit but I have run into some pretty basic problems that I
> hadn't come across before.  I'm sure the devices were logging events
> properly before I shipped them, but now every time I try to test a
> rule I get back the following line -
>
>  
>
> May 30 10:35:14 <hostname> snort[5017]: [1:486:4] ICMP Destination
> Unreachable Communication with Destination Host is Administratively
> Prohibited [Classification: Misc activity] [Priority: 3]: {ICMP}
> <probe address> -> <my test machine address>
>
>  
>
> There are a couple of relevant points here -
>
> $HOME_NET is defined as 10.0.0.0/8 and $EXTERNAL_NET is defined as
> !$HOME_NET
>
> The probe has an ip address assigned in a /24 10.x.x subnet
>
> My test machine has an ip address assigned in a different /23 10.x.x
> subnet
>
> I know there is no routing problem since I am SSH'd in to the probe
> from my test machine and I can ping in both directions
>
>  
>
> I am trying to read through the rule generating this message to see
> what is being triggered, but it looks pretty much like it is triggered
> on what it describes - an ICMP destination unreachable notification
> packet.
>
>  
>
> I looked back through the old logs and see the following from before I
> shipped it off so I know it was logging correctly -
>
> May 15 12:08:10 <hostname> snort[2331]: [1:2189:3] BAD-TRAFFIC IP
> Proto 103 PIM [Classification: Detection of a non-standard protocol or
> event] [Priority: 2]: {PIM} <old probe address> -> 224.0.0.13
>
>  
>
> The only thing that changed is the probe IP address, but both the old
> and new configs used DHCP and I'm pretty sure that when I was testing
> before the IP address changed and it still continued to log OK.
>
>  
>
> Any ideas what is causing this trigger ?  As above, I am reasonably
> sure that it is not an actual ICMP destination unreachable packet
> because the destination is reachable . . .
>
>  
>
> David
>

Sounds like it might be a firewall issue.  This response is usually sent
by a firewall to say that it's filter won't allow communication with
that host/port.  It may be allowing ssh to that host (port 22) and may
be allowing pings, but not allowing the host/port combo used in your test.

Todd




More information about the Snort-users mailing list