Slashdot Mirror
Security Probes for New Clients?
archaic0 asks: "I've recently acquired a new client (I do on call tech work for several companies where I live) who have requested a security audit. In the past I've hired several friends (self-proclaimed security consultants) in the industry to run various exploits and tests for me, but due to the time involved and the cost, I'd like to find a short introductory type option to start a new client off with. I recently ran across a program called Retina, by eEye, and I'm quite impressed however it comes with a $1400 price tag per use (or $14,000 a year for a bulk license). Can anyone point me to tools they've used to do a pretty well-rounded security scan that can produce detailed reports? I know there is no substitute for a real security professional spending time confirming your network security, but I'd like to have at least one good tool to start a new client off with before throwing a huge security team at them."
My company recently purchased an SSL cert from verisign and recently received an email from http://www.qualys.com (in conj. with our purchase) to perform a web based security scan of internet facing machines, such as web servers. The results and demo reports appeared a bit better than our usual Nesus vulneravility scan, however, Qualsys is not free. Try these tools out, for web servers, they have done quite well for my end.
It's the shit.
Good free ones: nessus, nmap, nikto. Besides Retina, look at Foundstone. There is also Qualsys, nCircle and several others (search for vulnerability assessment tools). Make sure that you understand the network topology, especially if firewalls & routers are involved. There are also host-based scanning tools designed to be run on individual systems, primarily to harden them.
Check out Nessus. Nuff said.
Revolutions are never about freedom or justice. They're about who's going to be top dog. -- Kilgore Trout
They requested a security audit? They.. asked for one?
For most companies security is an afterthought that's not worried about until they they notice their server's been compromised for a couple of months.
Most impressive, congratulations for finding a clueful employer. I hope the audit goes well.
[links not provided: it is assumed you can google]
First you'll want "nessus" -- this scans and attempts to exploit vulnerabilities. Comes complete with up-to-date 'signatures' for attacks to ensure that systems are patched or that firewalls are blocking access.
Second you'll want "GFI Languard" and run that to scan the internal Windows PCs -- it will give a nice report of each machine and patches needed (assuming you've got approval and admin access on the domain). This costs like $1k, but has a 30 day free trial to get the client started. Can also be used to deploy patches.
If you don't want to use Languard, which is really quite a bit better, you should at least use Microsofts "Baseline Security" tool. Again, requires admin access, but gives a nice report for each machine you scan.
nmap is nice to document open ports on machines, particularly so-called DMZ or other firewalled internet-accessible hosts.
dsniff is a good tool to watch for insecure protocols. Always fun to report that everyones pop3 password seems to be the same as their domain login password.
lopht crack is good to give a baseline indication of how secure user passwords are. Run it for a set amount of time -- 1 hour say -- using all of the passwords found by dsniff over a day or two as part of it's dictionary.
There's a lot more to do -- check routers etc. for default passwords, war-dial all phone numbers of the company looking for rogue modems and more default passwords, etc. But the tools above should give a pretty good start.
All of these tools produce reports in some flavor, which you can then combine manually. I assume the client is paying you for the report, so some manual effort is OK.
Make sure to push for a 'follow-up' audit after the client has remediated the problems.
"But actually trying to use m4 as a general-purpose langage would be deeply perverse" --ESR
war-dial all phone numbers of the company looking for rogue modems
Combine this with talking each answering person into giving their authentication information. I understand the easiest way to achieve that is by telling them you are hired by their company to make a security audit and said authentication information is necessary to point out flaws in their IT security. Not like I were experienced in the field but that's what they keep telling 'round the 'net, Mr. Mitnick for instance.
Have fun!
I think, therefore thoughts exist. Ego is just an impression.
A proper security audit shoud include a vuln assessment from the internet, but how about
1. Dial in lines..
2. social engineering - ring someone and say "Hi I'm the new guy in IT and I've been asked to check everyones password, can I have yours". Ring the IT dept, "Hi I''m fred from xyz sales inc. we sell firewalls (or whatever) can I spend a few minutes talking about your network security" amd so on.
3. Do they have a security Policy. How to they enforce the policy.
4. What about disaster recovery?
5. What happens when the senior IT security is on holiday/off sick and you get a reported breach?
6.
For Windoze systems checkout the Microsoft Baseline Security Analyzer although it relies on Windows Update, not a good sign, but it can at least check against MS known vulnerabilities - the client can already download and run but it can be used as a base level of checking to show how good your "industrial strength" tools are.
At the end of the day, its a cost/benefit exercise in trying to balance the clients budget against their paranoia.
-- Sig meltdown immine...
Others already posted links to various tools, so I'm not going to repeat that. However, you should be aware that these tools cover only a very small part of what a "security audit" should look into.
Corporate security is about much more than buffer overflows. Sure, it's worth keeping your PCs patched, but that doesn't mean that you're doing your security right. If I were hiring a contractor to do some sensitive work, I would look very carefully at e.g.
- physical security (office access controls, guards, cameras)
- personnel (qualifications, turnover, hiring practices, background checks)
- policies about acceptable behavior and whether they are followed (e.g. are you allowed to take your work home? is hard disk encryption mandatory for all laptops? can you give "guest accounts" for your friends or ex-employees?)
- continuity (offsite backups? redundant machines? ability to continue if a key person leaves?)
A security standard such as BS7799 should give you a more complete list of what matters.
Security is a process, not a product (no, I didn't make that up - check Bruce Schneier's company).
Security is a fairly wideranging topic, and involves at least half a dozen different, highly specialized disciplines. You may not need to be particularly thorough in all of them, but if you follow the great advice to use Nessus for network scanning, you may not realize that your client has left a gaping big hole in their ASP code which will allow arbitrary database requests to be executed against your client's database.
Or, you could have tightened down your network and website, but have no protection against viruses or worms on the desktop. Or there may be a wifi point allowing access to all and sundry. Or the server room may be accessible from the kitchen where many casual staff work. Or your client's CEO's daughter's boyfriend might have access to his PC with a VPN connection that automatically starts without prompting for a password....
So, yes, it's a good idea to use automated tools to do a basic audit. Nessus is good. You could do worse than read "Hacking Exposed" - it mentions a lot of good tools, both free and commercial, as well as the basic process for conducting a security audit.
However, make sure your client realizes that a clean bill of health (or fixing the issues your tools reported) does not mean they are "safe", (nor that they can sue you for any breaches that might occur), but rather that their organisation is not vulnerable to the attacks you tested for. If you didn't "test" hiring practices, they have no idea whether they are protected against employee fraud (which is still by far the most common form of computer crime). If you didn't "test" their virus protection policy, they have no idea of how exposed they are to the next email worm.
And of course, you are never "safe" - new threats emerge every day, and a server that was as safe as Fort Knox yesterday might be more like a crackhouse when the latest spl0it is released. So it's an ongoing process - assess, evaluate, repair, repeat & rinse.
Now, if your client is a small local firm with family members as employees, who use computers only for non-critical tasks, the "we'll run Nessus once a month" approach might be okay. If they are - oh, say, Microsoft...- that approach is clearly not sufficient.
Think about the interests of your client - not just in terms of saving them money, but protecting them from risk.
It's all very well in practice, but it will never work in theory.
These guys do inexpensive automated scans for a living. They run all the tools you know and love (nessus, nmap, etc.), and can be set to scan on a schedule, or you can do one-offs.
This is a plug (they're friends), but check it out: It seems to be what you're looking for.
7. Profit!
If you are really wanting to do a thorough job on the cheap, there are various places on the net were you can get a team of experts on it for no charge, just by posting their IP addresses, etc.
Reporting is a problem though.
-- MarkusQ
P.S. Hint for the humour-challenged: this is the kind of post that comes with a "hint for the humour-challenged" attached.
All the comments I've read here talk about penetration testing, etc. None of these provide a true "security audit," if that's what your client is requesting. Although it's important to look at technology tools when doing security audit, it can be more important to look at your processes and approaches to doing work, too.
Slightly off topic, but I've done work in vulnerability assesment, forensics, and security testing. The first lesson anyone going in this realm should know is that if you claim that a network is secure and it gets hacked, your credibility goes right into the toilet.
Make sure you stress heavily that the only secure machine is an unplugged machine and all you can do is look for existing security holes, like missed security updates and firmware or poorly set up computers. Make sure your client understands that most security breaches come from a company's own employees. I've worked on projects that found a company's own network was secure, but their ISP had a security hole that allowed us to completely bypass all their security. I've seen post-its on monitors with username/password written on them. One time we had a guy walk into a bank, claim to be a new employee, and get set up on a terminal with an account. I've seen entire IT departments escorted out of a building by security while the Cisco vans pulled up out front to fix a down network because a router was missing a 6 month old firmware update and some skript kiddie took it down. There's nothing like wiping the grin off of a smug IT Admin's face, but it's a scary business if you don't practice a lot of C.Y.A. or try to claim that someone's network is totally secure.
Run a firewall, antivirus, and keep software and firmware updated and you won't have to worry about outside attacks so much. No software can find post-its with account info stuck to a monitor.
I would break the ice with a new client by giving the CEO a printout of his or her organization's userids and passwords (with some of the characters in the passwords obscured with *s). This usually gets their attention.
Does anyone have any experience with: http://www.scanalert.com/, They apparently check your system on a monthly basis for security holes, etc... I have no idea how thorough they are, but at less than $200/month, the price is right.
These are some of the best security audit tools I know of. Using any of them without written permission, or without giving a good explanation of what they do and what impact they'll have on their network, could subject you to lawsuits or prison.
nessus will scan for known vulnerabilities. I've heard it's the best, but haven't tried it myself. Be aware that running it will most likely crash some servers.
nmap will tell you all the open ports on all the systems on the network, and attempt to identify them.
ethereal will spy on network traffic. Look for suspicious traffic and cleartext passwords that shouldn't be cleartext.
The Microsoft Baseline Security Analyzer will identify missing patches and weak passwords. Though in my opinion simply running it requires you to be insecure, because it depends on "hidden" administrative shares to access the hard drives of all the systems on the network, which you may wish to disable.
l0phtcrack and Hydra are popular password crackers, used to detect accounts with weak passwords.
And like always (assuming they run Windows):
Check the firewall logs.
Make sure all security updates are installed.
Run the IIS lockdown tool on servers running IIS.
Make sure workstations are free of spyware/adware and other unwanted startup programs.
Look into the Windows gold standard and other popular security templates intended for locking down workstations and servers.
Make sure your wireless routers use adequate encryption. WEP is encrypted but uses weak keys.
Etc. Can go for hours.
If you don't have the budget for Retina, try Nessus. Even Eeye reccommends it, in this post on bugtraq.
It's easily the best product out there with the largest database of detections and reliable ones at that. Nessus is free and maybe has 2/3's of the database that Qualys has. Everyone else is a distant 3rd, with maybe 1/3 of Qualys' database.
For a free one off scan I'd suggest you use Nessus because they cost nothing to setup - just find a spare machine and install linux, and you can throw away the host after you've finished with it. One major thing to watch out for with Vulnerability scanners is that you make sure the host they're installed on is properly secured, I heard abotu a company that installed Foundstone's application, which needed an Microsoft SQL database to support the app - guess how many vulnerabilities adding that support machine added to their network? Qualys of course doesn't have any setup worries - either they run the scan from their remote servers, or you get one of their cute little 1u boxes, plug it in, give it an IP and it's done.
The other downside to the Nessus solution is that the presentation and management of the results isn't particulalry good, again that's one thing you see in the enterprise solutions, work flow management for rememdiation, as well as a lot of nice looking reports and summaries. If you're scanning your own network the Qualys scanner is a fabulous choice, I think qualys used to offer a pay-per scan service, so maybe you could get a deal for a one time scan. But if it finds any problems with you client then you're going to need to stump up more when the vulnerabilities are supposedly fixed.
So... maybe setup a nessus box, and maybe take advantage of Qualys free demo scans.
And make sure you get permission.
And of course turn off all the nessus tests which crash things.
And of course turn off all the nessus tests which crash things.
I'd say clone your production server if you can't afford for it to be down, but DO run the tests that crash things. You do want to know if some bored script-kiddie can take your site down with a trivial syn-flood or ping-of-death.
455fe10422ca29c4933f95052b792ab2
Foundstone (http://www.foundstone.com), ISS Internet Scanner (http://www.iss.net).
Also try Nessus (http://www.nessus.org) on the free side of things.
-Jack Ash
Frankly, if you have to ask these questions, you should shy away from offering security consulting. Pay someone that lives, eats, sleeps and breathes IT security and you'll serve your customer better. I do IT security work (and only IT security work) for a living. I don't know how many times we've gone into a company that paid someone to do a security assessment, asked to see the previous report and been handed the stock report that NessusWX generates. Invariably when we do our work and write our report detailing the risks the customer feels their previous 'security consultants' cheated them. Often we find massive security issues that for one reason or another the automated scanners don't pick up. It won't do your reputation any good to do a poor job. The ability to do proper analysis is not a black art, but it becomes easier with experience and study.
...with a trivial syn-flood or ping-of-death.
1996 called, they want their 1337 haXXs back.
What you'll find is people really defend to the death what they think is security based on how much they *really* know which you'll find is usually about the level of what they read in Computerworld magazine. Even the self-proclaimed experts. You are best off doing what you can, even if it's just scanners, but realize it's not exactly a definitive or even realistic test of security. But it is something and worth doing. Remember to keep it practical and most of all, make sure you can measure it. One problem with a lot of these scanners and semi tests is that they give you some arbitrary high-medium-low talk. Try to put real numbers in there so you can actually measure risk. For more details on practical testing with risk measurement is in the Open Source Security Testing Methodology Manual and a lot of information over at the Institute for Security and Open Methodologies. Inform yourself better and with ISECOM you can at least know it's a lot of information from many many security people (800+) giving peer review.
We use it at my shop, and all it does is provide a GUI to the clueless 'policy enforcers', i.e. the InfoSec guys. They scan you, Qualys says 'X is bad', and its a Federal Case. Of course, if you mention how X is not really true, because you're running in a chroot jail, for instance, it makes no difference. The Computer has spoken, and produces a Shiny HTML Report to prove that you have a Vulnerability. Qualys is just a pretty front and and control GUI for Nessus.
Much better to roll your own, because in the process, you'll have people actually LEARN security.
P.S. My solution? IPFilter on all my Solaris and HP boxen. The idiots only scan from 1 IP address, so its easy enough to send back TCP RSTs to that IP address.
I want to delete my account but Slashdot doesn't allow it.