RE: IDS causing troubles

Matthew Fitzgerald <[email protected]> Tue, 15 Feb 2011 11:08:16 -0500
Newsgroups gmane.comp.security.ids
Message-ID <18755F9CB505054685466605F331A1B003E71806D8@CAEMEX81.caecorp.cae.com>
Just to chime in about potential problems at the physical layer.  I've seen=
 these type of problems on numerous occasions.  At the trivial extreme ther=
e may exist a NIC duplex mismatch or speed mismatch, or in the case of all =
NICS set to auto-auto, the devices can have issues negotiating the speed/du=
plex.  I think generally the guidance out there will tell you to nail up th=
e ports on both sides but this isn't a solution in all cases.  At the more =
complex extreme there are many port stats that can indicate subtle issues. =
 Corrupt packets, out of sequence packets, retransmits, or dropped packets =
can all mean a field-day for an IPS.

You would think this would be picked up relatively quickly but it's a recur=
ring issue in my world.  It's important to know that this sort of negotiati=
on/renegotiation may only present itself under heavy traffic volume or a sp=
ecific type of traffic (MTU issues and so on).  What's more is that upon in=
vestigation, the stats on a port on one side of the connection may look rel=
atively clean whereas the port on the other side of the connection can be s=
truggling.

It can be tough to get a provider to dig into this when it "seems" to be wo=
rking at least for some or for the majority of the time.  It's even more in=
teresting when the two ends of the link are owned by different companies.



Matt Fitzgerald, P.Eng
Security Architect

CAE Professional Services
36 Solutions Drive
Suite 200
Halifax, NS
B3S1N2
Tel. 902-420-3070 x2127
Fax: 902-420-3087
[email protected]

CONFIDENTIALITY NOTICE=20
This e-mail message is intended only for the above named recipient(s) and m=
ay contain information that is privileged, confidential, and/or exempt from=
 disclosure under applicable law. If you have received this message in erro=
r or are not the named recipient(s), please immediately notify the sender, =
delete this email message without making a copy and do not disclose or rela=
y this e-mail message to anyone.



-----Original Message-----
From: Joel Esler [mailto:[email protected]]=20
Sent: February 15, 2011 11:25 AM
To: JiPi DiNi
Cc: Joel Jaeggli; Matthew Fitzgerald; Andrew Plato; Shang Tsung; focus-ids@=
securityfocus.com
Subject: Re: IDS causing troubles

On Feb 14, 2011, at 1:28 PM, JiPi DiNi wrote:

> If inline it has to be a bypass switch not a tap.
>=20
> an IPS with a TAP is an IDS.
> an IPS with a bypass switch configured inline can block on traffic.

You might want to clarify this statement a bit more, for instance, there ar=
e tap vendors that make devices called "Vmode" taps, which is essentially a=
n inline tap, the traffic goes through the tap, and sent through an IPS, ho=
wever if the IPS fails, the vmode tap "fails open" sending the traffic stra=
ight through. =20

This may be what you meant about a bypass switch, but just clarifying the t=
erminology.


--
Joel Esler
http://www.joelesler.net



-----------------------------------------------------------------
Securing Your Online Data Transfer with SSL.
A guide to understanding SSL certificates, how they operate and their application. By making use of an SSL certificate on your web server, you can securely collect sensitive information online, and increase business by giving your customers confidence that their transactions are safe.
http://www.dinclinx.com/Redirect.aspx?36;5001;25;1371;0;1;946;9a80e04e1a17f194