lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  PHC 
Open Source and information security mailing list archives
 
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Mon, 21 Jan 2008 01:07:50 -0600
From: "Nate McFeters" <nate.mcfeters@...il.com>
To: secreview <secreview@...hmail.com>
Cc: full-disclosure@...ts.grok.org.uk
Subject: Re: [Professional IT Security Providers -
	Exposed] PlanNetGroup ( F )

SecReview,
My 2 cents on your review, although I will try to be nicer then you were to
the reviewee.  I'm completely skipping your section where you talked to the
non-technical person, that's not even fair... sorta like reviewing a
consulting group based on their website alone... oh shit, I forgot you guys
do that too.

Your comments on Question 1:

We're not impressed with Michael's answer. First off we have no idea what
the hell this means: "Depending on time and availability, we will work on
finding any new vulnerability if we generate an anomaly of interest." And we
totally disagree with "Currently, the focus is primarily on discovering new
Oracle vulnerabilities - as MS SQL 2K5 is more difficult to beat on,
compared to Oracle." In fact, whatever is being described above doesn't
sound anything like a vulnerability assessment, we're not sure what kind of
service it is.

The first portion "Depending on time and availability..." I don't understand
what your confusion is.  Basically the responder is saying that he's willing
to do what the client will pay him for.  Consulting is not a cookie-cutter
gig, so sometimes clients want you to spend 5 minutes running scans, some
want you to fuzz a proprietary protocol for as long as it takes.  I
personally don't think either end of the extreme is of value to the client,
but you can hardly fault the respondent for delivering what the client asks
for.

The second, I don't agree the overall focus is on Oracle, but if you read
the new (ZDnet, eWeek), or if you follow the conferences (HITB Malaysia 2007
great Oracle presnetation), then you will know that Oracle is catching a bit
of the limelight.  Besides that, I don't think you are qualified to say what
exactly a vulnerability assessment is... if the client is paying you to
assess their database servers, then that is a vulnerability assessment of
their database servers and that is what the work is.  Different clients have
different needs, and their are different specialty consulting groups to help
meet those... can hardly fault him if his specialty is databases.

Your Comments on Question 2:

>>trying to be cute with your "Again, carefully!" bullshit?

Come on guys... imagine you get called by a group of people asking to assess
your company and you don't know who they are, wouldn't you try to befriend
them if possible?  A little professionalism would go a long way to improving
your reviews.

>>A penetration test is not "Anything Goes!"

Umm... sorry guys, there is plenty of cause for performing a Denial of
Service test.  Keep in mind that availability is a large portion of what
security is about.  I don't think he's talking about using a bot net to try
to take them down.

>>it doesn't sound like Michael knows how to perform IDS evasion testing.
Using a proxy is >>not going to help anyone evade detection, it will just
help them to hide their IP address.

Hmm... well, you're partially right.  I suppose that if he had enough proxy
servers and kept his scans very focused, he "might" be able to get around an
IDS.  In any case, not all clients want IDS evasion performed... for
instance, they may want to test their incident response, or, they may allow
the consulting group through the IPS/IDS in an effort to save on time and
costs.

Your response to question 3:

>>>From the answer above, it looks like they like the same tools as most
people. That said, >>we've seen no proof of talent from anyone at
PlanNetGroup yet. So we're near certain that >>their deliverables ARE the
product of automation.

If they are the same tools that everyone use, how can you knock them for
that?  It seems to me that a group starts with a score of 0 in your book,
and then if they impress you they get points.  If you don't ask the right
questions, I don't see how they could impress you.  I concede, it is
certainly possible that they have no skills, and that they use automation,
but I don't think it is fair to say that at this point of the review.

Your response to question 4:

>>Woha, it takes too much time to create a fake deliverable? Well that's one
way to get out >>of it, but we don't buy it. Either way, at this point we
don't feel that a sample report would >>help this review, we've seen nothing
impressive yet.

Ever tried to do so?  It does take awhile, and it is risky.  If you miss
sanitization and release results of one of your clients you could get sued.
 Perhaps given the context of the investigation he didn't want to give you
an old report and it would take to long and too much of his billable time to
actually get this to you.  That's not unreasonable.  You aren't paying him.
 Again with the comments of nothing impressive yet.  You are asking generic
questions, how could anything be impressive?  It's a phone call or email and
you are asking questions that almost all consulting groups should have
relatively the same answers to... I see nothing impressive in that at all.

Your response to question 5:

>>It sounds like Michael has a difficult time sticking to the scope of work.
Any time anyone >>performs Distributed Metastasis it should be built into a
scope of work first. If it is not, >>then do not perform the testing because
it is invasive and will get you into trouble. This is >>a big negative point
in our eyes as its critical that providers are able to adhere to the scope
>>of work for each specific engagement.

I actually agree with most of this, but then again, as long as he doesn't go
over the clients budgetary and time constraints and is providing the
customer with value, I have no problem with going outside of scope as long
as the client does not.  Also, I don't know that it is a big negative as you
say.

Your response to question 6:

>>It sounds like Michael is a corporate security guy and has no experience
as a hacker.
Bit of a blanket statement I'd say, but OK, let's assume you are correct
>>Certifications hold little to no water when it comes to real IT security.
Agreed, but you are totally putting words into his mouth.  He basically says
the same thing by calling the CISSP a definition test.  Why do that?  Most
people in security have the certs... most realize they are worth nothing and
don't really test tech knowledge, but instead test business knowledge.
>>What does hold water is experience and from what we can tell, Michael has
no real hacker >>experience.
Please define "no real hacker experience".  If you mean he isn't 31337 like
you guys, then OK.  BTW, most clients aren't just paying for "real hacker
experience" they're also paying for the business side, i.e. what is my risk,
how can I mitigate, etc.  A good team has both people.

On your response to question 7:

Do you resell third party technologies?

>>We don't think that it is a good idea that Professional IT Security
Providers sell third party >>technologies. Specifically because they become
biased towards a specific technology and >>push that technology as a method
of remediation when better methods might already exist.
Agreed.  But that said, what if your third-party tech. has nothing to do
with the main thrust of your consulting work?  The question is pretty vague.

On your response to question 8 and 9:

Ok, I'll buy that you have cookie cutter definitions from google of those
flaws and that his definitions don't fit.  I'll even buy that you make a
good point when you say EIP overwrite is not the only method of exploitation
(especially these days), but I'm wondering what you expected.  Should he
have rattled on and on about how to exploit b0f in an XP SP 2 environment?
 Talk to you at length about DEP?  Bit ridiculous expectations.  Hell, while
your at it, why didn't you ask him about integer overflows?  Off-by
one/few/many exploits?  Heap overflows?  Why not have him recite the Heap
Fung Sheui method to you?  What about double free flaws, dangling pointers,
etc. etc. etc.  Let's be serious here, unless you are contracted by
Microsoft or another major software vendor, you probably don't pay the bills
by doing your own research, so... does this really matter?  Sure, it's
great... I'd like to know that consultants I was paying top dollar to knew
about this, but if he comes on site and spends 3 weeks trying to find an
integer overflow, I'm going to be pissed.

Disclaimer:
I'm not a client of PlanNetGroup.  Also, I don't think what you are trying
to do is a terrible thing, there's lots of snake oil being sold in the
commoditized security market out there, but I disapprove of your
professionalism and your methods.  Also, I believe the list is still waiting
for you to credentialize yourself/yourselves.  That still hasn't seem to be
grasped here.  Look, if you're someone people respect, then maybe people
will buy your reviews, but somehow I doubt that is the case.  I'm basing
that view off of the content of your website and the fact that you still
have not credentialized yourself as the list called for so long ago.  Do
that, and I will re-review my review of your reviews.

Nate

On Jan 20, 2008 7:17 PM, secreview <secreview@...hmail.com> wrote:

> The PlanNetGroup is a Professional IT Security Services Provider located
> at http://www.plannetgroup.com. <http://www.plannetgroup.com/> One of our
> readers requested that we perform a review of the PlanNetGroup, so here it
> is. It is important to state that there isn't all that much information
> available on the web about the PlanNetGroup, so this review is based mostly
> on the interviews that we performed.
>
> The PlanNetGroup was founded by Jim Mazotas of Ohio USA according to this Affirmative
> Action Verification Form<http://odnapps01.odn.state.oh.us/das-eod/EODBMSDev.nsf/d881c0c739c3c9b985257344004f1929/c3e323de1df5162b8525735d00607a6d?OpenDocument>.
> We called Mr. Succotash and spoke with him for about an hour about his
> company, here's what he had to say.
>
> When we spoke with Jim Mazotas we asked him how he defined a Penetration
> Test. His answer wasn't really an answer at all but rather was a bunch of
> technical words strung into sentences that made no sense. Here is what he
> said for the most part. We can't give you an exact quote because he
> requested that some of the information related to clients, etc be kept
> confidential.
>
> "We get to target object, where we go with that is based upon the client's
> comfort level. We grab banner information, backend support information, and
> other kinds of information. During a penetration test we most will not
> penetrate. Most mid level companies will not want penetration." – Sanitized
> Quote from Jim
>
> Not only do we not understand what Jim said, but he'd be better off saying
> "I don't know" next time instead of looking like an idiot and making up an
> answer. This goes for all of you people that get asked technical questions.
> If you say "I don't know" at least you won't look like a fool. Anyway.
>
> When we asked Jim to define a Vulnerability Assessment, we became even
> more flustered. Again his answer was like a politician trying to evade a
> question with a bunch of nonsensical noise. Again, we've sanitized this at
> Jim's request.
>
> " A Vulnerability Assessment is more a lab based environment type test.
> Analyze servers and all nodes that are a true vital asset to the company and
> assess the vulnerability In a very planned out manner. This is done in a lab
> based environment." – Sanitized Quote from Jim
>
> Again, next time say "I don't know" because now you look like an idiot.
> Nobody expects you to know everything, but when you make shit up and try to
> fool people, its insulting. To be fair to Jim, he did say that he was not
> technical, but we didn't get technical here. As the founder of the business
> he should at least know what his different service boundaries are and how
> his services are defined.
>
> When we asked Jim if his team performed Vulnerability Research and
> Development, he said that they did not have the time because they were
> "fully booked". His primary customer base includes state government and a
> few private sector businesses. Unfortunately, we can't disclose who his
> exact customers are. He did say that he provides Network Management Services
> and Wireless Management services for many of his clients. Sounds more IT
> related than Professional Security related.
>
> When we finished with our call to Jim we asked him if he'd be kind enough
> to give us contact information for someone more technical in his company. He
> told us that he'd be happy to arrange a call with someone. At the end, we
> didn't end up calling anyone but instead shot a few emails back and fourth.
> The rest of this review is based on those emails.
>
> We decided to ask the same questions to Jim's technical expert. We know
> who his expert is, but we assume that he wants to stay anonymous because he
> signed his email with "Jason Bourne". So for the sake of this interview
> we'll call him Michael. Here's the email from Michael:
>
> -) How do you perform your vulnerability assessments?
>
> "* Carefully! :) Typically, we will work with the customer to define the
> scope of the assessment; limitations to OS, Network Equipment, Web
> Server, etc. This could be a combination of components (depending on
> scope), the real goal ultimately with this is to assess the patching
> effort of a customer. Depending on time and availability, we will work
> on finding any new vulnerability if we generate an anomaly of interest.
> Currently, the focus is primarily on discovering new Oracle
> vulnerabilities - as MS SQL 2K5 is more difficult to beat on, compared
> to Oracle. Within vulnerability assessments, we disregard any attempts
> to evade IDS, IPS, etc."
>
> We're not impressed with Michael's answer. First off we have no idea what
> the hell this means: "Depending on time and availability, we will work on
> finding any new vulnerability if we generate an anomaly of interest." And we
> totally disagree with "Currently, the focus is primarily on discovering new
> Oracle vulnerabilities - as MS SQL 2K5 is more difficult to beat on,
> compared to Oracle." In fact, whatever is being described above doesn't
> sound anything like a vulnerability assessment, we're not sure what kind of
> service it is.
>
> -) How do you perform your penetration testing?
>
> * Again, carefully! The definition that I use with customers is -
> Anything Goes! In addition to attempting to locate missing patches,
> vulnerable IOS's, applications, etc - we will perform an assortment of
> timed attacks, attempt to spoof trusted connections, or even perform
> social engineering - like dropping a few pre-trojan'd usb data sticks
> outside of a customer service area, a data center, etc. The only thing
> that we do not perform, typically, is denial of service style or type of
> attacks. We have had only one customer that we felt was in the position
> to handle such a test and it was performed against their disaster
> recovery infrastructure, not production."
>
> Michael, why are you trying to be cute with your "Again, carefully!"
> bullshit? A penetration test is not "Anything Goes!", if that's how you
> define it then I don't want you anywhere near any of my networks. And why
> the hell would you perform a Denial of Service attack against anyone?
> Everybody can be knocked off line if you fill up their pipe. You scare us
> man!
>
>
> -) How do you perform evasive IDS testing?
>
> "* We use a series of proxy servers to attempt to perform basic hacking
> techniques; port scans, blatant attacks, etc. We are typically going to
> look for TCP resets as a means to evaluate if IDS is present and
> possibly to find if IDS performs blocking activity. Often times, if a
> system in a trusted DMZ can be compromised and used as a proxy
> (exploiting a relationship or rule within a firewall) or an SSH, SSL,
> encrypted tunnel can be established to a server behind the IDS sensor
> than we can successfully pull off an attack without the customers
> security staff even knowing."
>
> It doesn't sound like Michael knows how to perform IDS evasion testing.
> Using a proxy is not going to help anyone evade detection, it will just help
> them to hide their IP address. If the target network or application is being
> protected by an IPS device, then the IP that they are attacking from will be
> shunned just the same. So, we understand that the PlanNetGroup's expert
> hasn't a clue as to how to evade IDS. (Michael, did you get your answer from
> Google?)
>
> -) What tools do you favor?
>
> "* We really do not favor any tools. The focus of our effort (Assuming we
> are performing a pen-test or assessment) is to analyze a situation and
> choose the best tool for the end result or compromise. I will use commercial
> applications, such as AppScan, WebInspect, even ISS. There are however
> plenty of freeware, low-cost tools that we use; nmap, nessus, metasploit -
> ultimately, I find that an internet browser and a telnet prompt will suffice
> for much of the testing. It ultimately gets back to interpreting the results
> and adjusting the testing accordingly. We make it a point to try out new
> freeware tools on every assignment. The more tools that we know of and can
> test with opens our options if in the future a situation best suited for a
> tool presents itself."
>
> Every business that delivers security services has a set of tools that
> they use. These tools change from business to business, but common ones are
> nessus, webinspect, CANVAS, Core Impact, Metaspoloit, etc. From the answer
> above, it looks like they like the same tools as most people. That said,
> we've seen no proof of talent from anyone at PlanNetGroup yet. So we're near
> certain that their deliverables ARE the product of automation.
>
> -) Can you provide us with sample deliverables? (sanitized)
>
> "* No, too much time. Even to sanitize creates an opportunity for a
> liability in the event that a customer name is exposed ... accidents do
> happen! I will say that we do not take dumps from applications and
> regurgitations the information on paper. We limit our executive summary to 6
> pages at most and attempt to keep the entire report limited to 25 pages in
> total. Our goal with a deliverable is to get the precise information to the
> key stake holders so that they can make a decision."
>
> Woha, it takes too much time to create a fake deliverable? Well that's one
> way to get out of it, but we don't buy it. Either way, at this point we
> don't feel that a sample report would help this review, we've seen nothing
> impressive yet.
>
> -) Do you offer the option of performing Distributed Metastasis?
>
> "* No, not really. This is my decision as in a previous life I got walked
> out of Bell Atlantic Mobile (Verizon Wireless) using this technique when I
> compromised their Unix infrastructure by compromising the rlogin function
> (on all Unix servers, across all data centers). There is no substitute for
> experience, especially bad ones!"
>
> It sounds like Michael has a difficult time sticking to the scope of work.
> Any time anyone performs Distributed Metastasis it should be built into a
> scope of work first. If it is not, then do not perform the testing because
> it is invasive and will get you into trouble. This is a big negative point
> in our eyes as its critical that providers are able to adhere to the scope
> of work for each specific engagement.
>
> -) What is your background with relation to information security?
>
> "* Too long, too boring. Yeah got the CISSP (nice vocabulary test), but
> had to as I worked for DOD. Got a number of Certifications (I have a stack
> almost an inch thick and only get into them about once a year to throw
> another couple on top of the previous ones - too much alphabet soup for me,
> but bosses and customers like it. Spoke at a number of
> European conferences, but found too many people did not understand a word
> I was talking about, so I got tired of that and quit that scene. My outlook
> on security has changed, to the point that I will advise customers of their
> risk, attempt to make it practical - but if they make a conscious choice not
> to listen - I do not fret over it.?"
>
> It sounds like Michael is a corporate security guy and has no experience
> as a hacker. Certifications hold little to no water when it comes to real IT
> security. What does hold water is experience and from what we can tell,
> Michael has no real hacker experience.
>
> -) Do you resell third party technologies?
>
> "* No, but kind of wished that we would. I think that it would help with
> sales."
>
> We don't think that it is a good idea that Professional IT Security
> Providers sell third party technologies. Specifically because they become
> biased towards a specific technology and push that technology as a method of
> remediation when better methods might already exist.
>
> -) Can you tell me why the EIP is important?
>
> "* The EIP controls an applications execution. If an attacker can modify
> the EIP while it is being pushed on the stack then the attacker *could*
> execute their own code and create a thread (aka. a buffer overflow condition
> exists). I had a good refresher this past year at Blackhat with a course run
> by Saumil Shah - he had an interesting buffer overflow
> for the Linked-In client."
>
> The EIP is the Instruction Pointer for the x86 architecture. The purpose
> of the EIP is to point to the next instruction in a particular code segment.
> If the EIP can be overwritten then the flow of control of an application can
> be changed. In most cases this can lead to the execution of arbitrary code
> on the targeted system. Hackers use this to penetrate vulnerable systems.
>
> -) Can you define a format string exploit?
>
> "* A format string exploit leverages what is considered a programming
> bug. If input is not sanitized, an attacker can perform calls to the
> stack; read, write, etc without knowing details about the EIP."
>
> Unfortunately this answer isn't accurate or detailed enough as almost all
> software vulnerabilities are the result of user input that is not properly
> sanitized or validated. A format string condition occurs when a user inserts
> a format token into a C based application and that input is not properly
> sanitized. Hence why it is called a format string vulnerability. When that
> input hits a function that performs formatting, such as printf() the input
> is interpreted in accordance with the format tokens. Sometimes this can be
> used to write arbitrary data to arbitrary memory locations. The EIP isn't
> the only valuable memory location.
>
>
>
>
> If you've managed to get this far, then you've survived reading Michael's
> answers to our questions. We're not going to spend much more time writing
> this review because by now we've formed our opinion. We did take a quick
> look at the PlanNetGroup's website and as with their people, we were not the
> least bit impressed.
>
> Our opinion of the PlanNetGroup is that they'd have a hard time hacking
> their way out of a wet paper bag. Their security expert is not an expert by
> our standards, as he did not properly answer any of our questions or help to
> define any of their services. We're pretty sure that the PlanNetGroup could
> run nessus and offer basic vulnerability assessment services. We're also
> pretty sure that they could offer IT services at some level. But we'd hardly
> call them subject matter experts and wouldn't recommend their services to
> anyone.
>
> If you are using the PlanNetGroup services and feel that we have not given
> them a fair review then please comment on this post. We will consider your
> comments. We have to say that Jim and Michael were both very polite,
> friendly, and respectful, but we can't let their kind nature impact our
> opinion of their service delivery capabilities. We think that they should
> sit down and try to define their services properly. We also think that they
> should hire an ethical hacker with real world experience if they intend to
> protect anyone.
>
> Score Card (Click to Enlarge)
>
>
> <http://bp2.blogger.com/_VcwqM25xL9M/R5PxN8GqVTI/AAAAAAAAACU/D7T4RSQlSXs/s1600-h/96YV5X.jpeg>
>
> --
> Posted By secreview to Professional IT Security Providers - Exposed<http://secreview.blogspot.com/2008/01/plannetgroup-f.html>at 1/20/2008 04:21:00 PM
> _______________________________________________
> Full-Disclosure - We believe in it.
> Charter: http://lists.grok.org.uk/full-disclosure-charter.html
> Hosted and sponsored by Secunia - http://secunia.com/
>

Content of type "text/html" skipped

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/

Powered by blists - more mailing lists