Re: IPS vs. Firewalls (why vs. ?)

[Dave Piscitello <dave () corecom com>]

Interesting taxonomy, but many firewalls today combine the three kinds 
of security services you mention. My firewall's pretty GUI has proxies 
and IPS and stateful packet filters (oh my!). Actually, two of the other 
firewalls I have lying on the floor in my office have at least two of 
the three, and I suspect that one really does have a proxy but the 
marketing people don't want anyone to know about it.

Conclusion? I don't think the traditional arguments over proxy vs. DPI, 
signatures vs. NBS (0-day), etc. are all that relevant nor interesting.

While it's true that we go to war with the equipment we have and not the 
equipment we wish we had, it's also true that we can be more successful 
if we use *all* the equipment we have, and use each where it is most 
effective and efficient.

Proxies get the job done in a lot more situations than they are given 
credit for, and perform well. Claims that proxies unilaterally impede 
performance are marketing drivel. If you take issue with this, consider 
that some companies who bash proxies as being performance inhibitors 
bolt SSL VPNs onto their firewalls. So far as I know, there are no 
stateful inspection SSL VPN implementations.

There are situations where IPS may indeed provide relief from certain 
classes of attacks. There are probably situations where they will suck 
more cycles from your firewall than you can afford. If you want to prune 
and tune an IPS, God Bless You, there's the door, thanks for the visit.

Vendors will forever seek to innovate attack recognition and blocking. 
The hard part for us all is distinguishing one-off clever programming 
tricks from landmark and disruptive technology. Try before you buy, 
caveat emptor, etc.

Like most other aspects of network technology, security technology has 
no choice but to concede to the overwhelming pressures of "convergence". 
Products may begin as a pure-play IDS, IPS, firewall, SSL VPN, IdM, 
anti-malware/spam/spyware gateway, but to survive, they are forced to 
meld all these "point" solutions into a single offering. This really is 
no different from how routing, switching, and WAN access (muxes) evolved.



Gabriele Buratti wrote:
> Parental advisory: explicit vendor opinions may occour in this message !
> Let me show show how IPS firewall market is seen from a IPS firewall
> vendor perspective. I've been following this mailing list for 3 years
> and few vendor opinions popped up. I don't know if this is because it's 
> considered a kind of advertising (thus unpolite) or what ... (in this 
> case list admins, please drop this mail)
> Let me invite my competitors in a friendly discussion about this layer 7 
> thing :)
>
> Here's the thing:
> 1) Proxy firewalls: Proxy firewalls are in theory good because they can
> do rfc compliance checks and "strange things won't be forwarded"
> approach aka the marketing "day-0 protection". More, they'll do fragment
> reassembly. The problems about proxies are:
> - performance decreased due to complete session rewrite
> - when used as reverse proxies for incoming connections you always have
> that listening ports on the proxy-firewall. Listening ports means
> attackable ports.
>
> 2) Firewalls with signatures: just the old IDS signatures, but now
> inline. The problems with signatures are:
> - keep the number of signatures low or it'll be a bottleneck thing
> (false negatives)
> - false positives
> - any variation of a know attack signature will be a new signature
>
> 3) new technologies:
> - reassemble the fragments in a separate space, do the checks, then if
> ok send the fragments (no session rewriting).
> - focus on the "strange things won't be forwarded", rather than
> signatures: faster, sharp, you can use the marketing wizard's "0-day
> protection" word :)
> - decode recursively to stop blended attacks
> - don't use a proxy: check on the fly and if test is passed then forward 
> the packet (so no session rewrites and no dangerous listening ports)
>
>     Gabriele
>
> Marcus J. Ranum wrote:
> > I'd suggest you have them ask a few of the IPS vendors if they recommend
> > using their product in that manner. Unless you're talking to the IPS 
> > vendors
> > that are basically selling a firewall+signatures (like a "deep packet 
> > inspection"
> > firewall) they will backpedal away from that very rapidly. Perhaps your
> > path of least resistance is to tell them that you want one of the new
> > generation "IPS firewalls" then you can turn off the IPS crap (which
> > won't do anything except slow the firewall down, anyhow) and use the
> > firewall rules.

Attachment: dave.vcf
Description:

Attachment: smime.p7s
Description: S/MIME Cryptographic Signature