Re: IDMEF Draft 15

[email protected] Fri, 24 Feb 2006 14:41:16 +0000
Newsgroups gmane.ietf.idwg
Message-ID <W9067928291325581140792076@webmail4>
----=_vm_0011_W9067928291_32558_1140792076
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
X-MIME-Autoconverted: from 8bit to quoted-printable by mail2.ac.hmc.edu id k1OEfIVJ023984

Hi Herve,

As GuardTower from SCS was largely one of the first SEMs developed in 200=
0 for an American
Telco, we have a vested interest in the proceedings of this group, as doe=
s Anton. I believe that
standardization in data export would go a long way in making ubiquitous i=
nformation sharing.
for IPS incident response reporting.

Unfortunately, recent changes in hardware on commerical appliance IPS tha=
t scale to carrier class
bandwidths ( a necessity for signature comparison at such speeds) have th=
e industry=20
moving back into an exclusive proprietary SIM/SEM reporting models , I su=
spect for driving revenue for their=20
proprietary architecures reporting consoles/ SEMs. This appears to be rea=
lly playing fits with the managed services=20
providers for supporting the new IPS appliances, I hear. This development=
 is really a misguided tangential
direction, and I believe represents.the reason why the attempted standard=
ization for this group and the group itself's charter.
Perhaps getting on with it.. and having IETF acceptance, is better, rathe=
r than perfection or agreement on the
draft.contents. In this way we can have the user community ask ( well doe=
s you IPS support the IDMEF=20
standard for reporting Mr. IPS Vendor?, because that is a Mandatory requi=
rement in our RFP, so we can conttinue=20
to use our current SEM/SIM or choose an SEM/SIM as seperate best of breed=
 solution based on it supporting our
organization's technology set.

What does this mean for IDS/IPS schema and RFC? Well point me to the late=
st draft and I will try to
make an articulate evaluation, but I would hope encrypted HTTPS transmiss=
ion of the sreported chema /incident datagrams
would be in the RFC?

Dan White
CEO- Secure Commerce Systems
www. securecommercesystems.com
866- SEC COMM

>-----Original Message-----
>From: Herve Debar [mailto:[email protected]]
>Sent: Friday, February 24, 2006 03:47 AM
>To: 'Anton Chuvakin, Ph.D.'
>Cc: ''intrusion detection wg''
>Subject: Re: IDMEF Draft 15
>
>Anton Chuvakin, Ph.D. wrote:
>> I am wondering if this message to the list actually got thru. If it di=
d,
>> IDMEF is truly dead since nobody even bothered to respond to the
>> comments about its death...
>
>Well, it did go to the list, but I'm not sure there is much to respond
>to. You are voicing your opinion, which is fair enough, and at the same
>time not being specific enough to enable a response, IMO. And I'm really
>not interested in a flamefest at this stage.
>
>>> All,
>>>
>>>> By 25 February, we would like to get all applicable comments. Again,=
 we
>>> are not making changes to the substance of the document, but rather j=
ust
>>> cleaning up an typo type problems.
>>>> Once we are done with the document, we will ask that it be published
>>>> as an
>>> experimental rfc.
>>>
>>> So, IDMEF is officially "dead" then? IDS might not be dead, but IDMEF
>>> largely is... I haven't posted to this list that much, but I've been
>>> watching IDMEF for many years. In fact, netForensics nFX SIM solution=
s's
>>> early XML event tranfer protocol, defined back in 2000, was loosely
>>> inspired
>>> by the IDMEF. However, IDMEF always was and remains largely unsuitabl=
e
>>> for
>>> our purposes, and I suspect other vendors in the SIM/SIEM/correlation
>>> space
>>> would agree with me. And so would the IDS (now IPS, for the most part=
)
>>> vendors. The issues are too numerous to list here. Even something as
>>> basic
>>> as alert vs hearbeart distinction raises some questions.
>
>There are SIM environments using IDMEF, commercially.
>
>>> One area where I can see shreds of IDMEF surviving is
>>> cross-organizational
>>> data sharing. A little problem with this is that it largely does not
>>> exist... I can see how a hybrid of IDMEF and IODEF with inherent and
>>> well-defined data sanitization might come handy. Other than that,
>>> utilization of IDMEF in four other use cases (from the page 4 of the
>>> doc) is
>>> totally irrelevant and is being better addressed by other means.
>
>As far as I can tell, INCH is also nearing completion, so again you may
>be late bringing in comments.
>
>Herv=E9
>--=20
>Herv=E9 Debar <mailto:[email protected]>
>Tel: +33 (0)2 31 75 92 61 GSM: +33 (0)6 74 09 09 66
>France T=E9l=E9com R&D (new)Fax: +33 (0)2 31 37 83 43
>42 rue des Coutures (--) BP 6243 (--) F-14066 Caen Cedex 4
>


----=_vm_0011_W9067928291_32558_1140792076
Content-Type: text/html; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<html><P>Hi Herve,</P>
<P><BR>As GuardTower from SCS was largely one of the first SEMs developed=
 in 2000 for an American</P>
<P>Telco, we have a vested interest in the proceedings of this group, as =
does Anton. I believe that</P>
<P>standardization in data export would go a long way in making ubiquitou=
s information sharing.</P>
<P>for IPS incident response reporting.</P>
<P>&nbsp;</P>
<P>Unfortunately, recent changes in hardware on commerical&nbsp; applianc=
e IPS that scale to carrier class</P>
<P>bandwidths ( a necessity for signature comparison at such speeds) have=
 the industry </P>
<P>moving back into an exclusive&nbsp;proprietary SIM/SEM reporting model=
s , I suspect for driving revenue for their </P>
<P>proprietary architecures reporting consoles/ SEMs. This appears to be =
really playing fits with the managed services </P>
<P>providers for supporting the new IPS appliances, I hear. This developm=
ent is really a misguided tangential</P>
<P>direction, and I believe represents.the reason why the attempted stand=
ardization for this group and the group itself's charter.</P>
<P>Perhaps getting on with it.. and having IETF acceptance, is better, ra=
ther than perfection or agreement on the</P>
<P>draft.contents. In this way we can have the user community ask ( well =
does you IPS support the IDMEF </P>
<P>standard for reporting Mr. IPS Vendor?, because that is a Mandatory re=
quirement in our RFP, so we can conttinue </P>
<P>to use our current SEM/SIM or choose an SEM/SIM&nbsp; as seperate best=
 of breed&nbsp;solution based on it supporting our</P>
<P>organization's technology set.</P>
<P>&nbsp;</P>
<P>What does this mean for IDS/IPS schema and RFC? Well point me to the l=
atest draft and I will try to</P>
<P>make an articulate evaluation, but I would hope encrypted HTTPS transm=
ission of the sreported chema&nbsp;/incident datagrams</P>
<P>would be in the RFC?</P>
<P>&nbsp;</P>
<P>Dan White</P>
<P>CEO- Secure Commerce Systems</P>
<P>www. securecommercesystems.com</P>
<P>866- SEC COMM<BR><BR>&gt;-----Original Message-----<BR>&gt;From: Herve=
 Debar [mailto:[email protected]]<BR>&gt;Sent: Friday, Februa=
ry 24, 2006 03:47 AM<BR>&gt;To: 'Anton Chuvakin, Ph.D.'<BR>&gt;Cc: ''intr=
usion detection wg''<BR>&gt;Subject: Re: IDMEF Draft 15<BR>&gt;<BR>&gt;An=
ton Chuvakin, Ph.D. wrote:<BR>&gt;&gt; I am wondering if this message to =
the list actually got thru. If it did,<BR>&gt;&gt; IDMEF is truly dead si=
nce nobody even bothered to respond to the<BR>&gt;&gt; comments about its=
 death...<BR>&gt;<BR>&gt;Well, it did go to the list, but I'm not sure th=
ere is much to respond<BR>&gt;to. You are voicing your opinion, which is =
fair enough, and at the same<BR>&gt;time not being specific enough to ena=
ble a response, IMO. And I'm really<BR>&gt;not interested in a flamefest =
at this stage.<BR>&gt;<BR>&gt;&gt;&gt; All,<BR>&gt;&gt;&gt;<BR>&gt;&gt;&g=
t;&gt; By 25 February, we would like to get all applicable comments. Agai=
n, we<BR>&gt;&gt;&gt; are not making changes to the substance of the docu=
ment, but rather just<BR>&gt;&gt;&gt; cleaning up an typo type problems.<=
BR>&gt;&gt;&gt;&gt; Once we are done with the document, we will ask that =
it be published<BR>&gt;&gt;&gt;&gt; as an<BR>&gt;&gt;&gt; experimental rf=
c.<BR>&gt;&gt;&gt;<BR>&gt;&gt;&gt; So, IDMEF is officially "dead" then? I=
DS might not be dead, but IDMEF<BR>&gt;&gt;&gt; largely is... I haven't p=
osted to this list that much, but I've been<BR>&gt;&gt;&gt; watching IDME=
F for many years. In fact, netForensics nFX SIM solutions's<BR>&gt;&gt;&g=
t; early XML event tranfer protocol, defined back in 2000, was loosely<BR=
>&gt;&gt;&gt; inspired<BR>&gt;&gt;&gt; by the IDMEF. However, IDMEF alway=
s was and remains largely unsuitable<BR>&gt;&gt;&gt; for<BR>&gt;&gt;&gt; =
our purposes, and I suspect other vendors in the SIM/SIEM/correlation<BR>=
&gt;&gt;&gt; space<BR>&gt;&gt;&gt; would agree with me. And so would the =
IDS (now IPS, for the most part)<BR>&gt;&gt;&gt; vendors. The issues are =
too numerous to list here. Even something as<BR>&gt;&gt;&gt; basic<BR>&gt=
;&gt;&gt; as alert vs hearbeart distinction raises some questions.<BR>&gt=
;<BR>&gt;There are SIM environments using IDMEF, commercially.<BR>&gt;<BR=
>&gt;&gt;&gt; One area where I can see shreds of IDMEF surviving is<BR>&g=
t;&gt;&gt; cross-organizational<BR>&gt;&gt;&gt; data sharing. A little pr=
oblem with this is that it largely does not<BR>&gt;&gt;&gt; exist... I ca=
n see how a hybrid of IDMEF and IODEF with inherent and<BR>&gt;&gt;&gt; w=
ell-defined data sanitization might come handy. Other than that,<BR>&gt;&=
gt;&gt; utilization of IDMEF in four other use cases (from the page 4 of =
the<BR>&gt;&gt;&gt; doc) is<BR>&gt;&gt;&gt; totally irrelevant and is bei=
ng better addressed by other means.<BR>&gt;<BR>&gt;As far as I can tell, =
INCH is also nearing completion, so again you may<BR>&gt;be late bringing=
 in comments.<BR>&gt;<BR>&gt;Herv=E9<BR>&gt;-- <BR>&gt;Herv=E9 Debar &lt;=
mailto:[email protected]&gt;<BR>&gt;Tel: +33 (0)2 31 75 92 61=
 GSM: +33 (0)6 74 09 09 66<BR>&gt;France T=E9l=E9com R&amp;D (new)Fax: +3=
3 (0)2 31 37 83 43<BR>&gt;42 rue des Coutures (--) BP 6243 (--) F-14066 C=
aen Cedex 4<BR>&gt;</P></html>

----=_vm_0011_W9067928291_32558_1140792076--