« first day (816 days earlier)      last day (4064 days later) » 

12:24 AM
@Ladadadada haha yup same guy
and @AviD no I did not give up
 
 
1 hour later…
1:39 AM
@twinborn So help us out.What kind of IT work have you done, what do you consider to be a couple of your best relevant skills, and do you know of any specific IT security areas that interest you?
 
2:31 AM
Morning.
 
 
2 hours later…
4:08 AM
 
 
2 hours later…
5:49 AM
I just pushed Bear #2 over Hendrik on the rep charts. :)
 
 
1 hour later…
7:17 AM
morning
 
@LucasKauffman morning!
 
I'm just reading up on the Diginotar hack
those guys really screwed up badly
 
@LucasKauffman That breach last year?
Yeah that was really screwed.
 
@LucasKauffman Judge not lest ye be judged.
 
@AviD morning.
 
7:21 AM
Makes me wonder what Breach Volleyball would be like...
@TerryChia yes, yes it is. And way too soon, if you ask me.
 
@AviD I'll judge if you are a CA that gets bumfucked because you fail to apply over 30 security patches to your main website on a internet facing production server and furthermore use one password to administer all your machines
 
@LucasKauffman So which is the bigger crime in your eyes? :P
 
@LucasKauffman well yeah, sure. But anyone can, and often does, screw up royally. Even a CA.
 
@TerryChia The fact that their password was Pr0d@dm1n
 
@TerryChia for me, it would be password. not even a single one, but the fact that it IS a password, and not something stronger.
 
7:24 AM
"Sure that password won't be in any dictionaries, it's super secure"
 
@LucasKauffman whoa. that was really their password??
 
@AviD yea
 
@LucasKauffman heh, "we used a @ instead of 'a'!"
 
@LucasKauffman hahahaha. they should have hired me to setup their stuff. even i know better.
 
I mean if you sell flowers ok, but if your product is trust and privacy
 
7:25 AM
@TerryChia looks like they did take some inexperienced teenage college student to do it for them.
No Offense.
@LucasKauffman you don't really understand the CA's business model, then...
It's more like protection money.
 
@AviD None taken. Most of my classmates are pretty shitty at this stuff.
 
Pop quiz - what business is Google in?
 
@AviD Advertising.
 
@AviD datamining and advertising
 
very good boys and girls.
now where is that well known cynical analysis regarding CA's business?
Here's another one: What is Facebook Inc.'s product?
 
7:27 AM
@AviD if you can find it, you can give it to me, I'm making a presentation on cyber warfare
I just hate the fact that I have to use cyber in it
 
@LucasKauffman tolja, its protection money.
@LucasKauffman haha, joys of the big 4?
 
@AviD People.
 
@TerryChia I was gonna go with "you are", but that's good enough. :-)
 
@AviD You should see some emails pass-by, everytime I see cloud and security in one email, a little part of me dies
 
If you aren't the customer, you are the product. ;)
2
 
7:29 AM
@TerryChia eggsactly.
 
@ThomasPornin Thanks! Although I really wanted my highly-voted answers to be on different (More difficult to google) subjects. But I'm aiming for 2k here and 3k in SO, hopefully I'll get them soon.
 
@Adnan but that is contrary to how it works. It's specifically the easy-to-google subjects that bring the crowds, and its the crowds that bring the votes.
 
@Adnan Hell, look at my top 3 answers. I'm not exactly proud of two of them.
90
A: "Username and/or Password Invalid" - Why do websites show this kind of message instead of informing the user which one was wrong?

Terry ChiaIf a malicious user starts attacking a website by guessing common username/password combinations like admin/admin, the attacker would know that the username is valid is it returns a message of "Password invalid" instead of "Username or password invalid". If an attacker knows the username is vali...

90 votes for common sense.
 
@AviD I didn't explain correctly. I meant I want my answers to be new and more interesting than a link to a Google help page.
 
@Adnan but that is what the unwashed masses can understand easily.
 
7:35 AM
@TerryChia Thanks, hopefully I'll feel better about it.
 
look, imagine a Venn diagram. One circle is Awesome answers, another circle is High-Voted answers. The intersection is quite small, and mostly populated by @ThomasPornin.
@Adnan same thing with me. And @Polynomial. I'm sure quite a few others.
 
@AviD You see, I'm here for 3 reasons. 1- I really want to make the Internet a better place. 2- I want to show off to current/potential employers. 3- Some sort of disorder that pushes me to help people.
 
@Adnan same here.
 
These answers help me with 1 and 3. But not 2.
 
truth is, you dont really need a lot of high rep answers for that.
you just need a handful of really well-written, non-obvious answers - and collect them in your Careers2.0 profile, point people at that.
Sure, they can read on, and see your other posts, such as your popular-yet-crappy answers, but so what?
A person is not judged on everything he writes (besides the negative things), but on those he finds cause to be proud of.
I actually spent some time on my careers profile, crafting the image that reflects my persona. The answers that I feel really show off either my knowledge, my writing, or my non-obvious stance on security issues.
 
7:40 AM
@Adnan my highest voted answer is how to uninstall bloat ware. The problem is that with more complex answers you won't get as much sympathy as with simple answers
 
@AviD Yeah, I find comfort that I can do that.
@LucasKauffman True. That is actually helping me in my RepTrain quest.
 
8:05 AM
Morning SE. Just wanted to know have anyone worked on Ambush IPS? ambuships.com
Working on a project to detect DLLs that are normally used by exploits. This way we don't need any signature of the exploit, if it is using those DLLs, it will be flagged
 
@void_in DLLs?
 
@void_in wouldnt it be trivial to obfuscate those dlls?
 
You mean, .dll files?
 
@AviD actually exploits call some of the DLLs in the windows. Like DownloadToURL is always used for downloading the second stage. This way if you hook this DLL, you are going to caught it without any signature
I know you will think it has a false positive rate but detecting a few "dangerous" DLL functions in correlation (one after another) is really going to bring down the false positives, or so at least I am hoping
@Adnan Yes. I will hook DLL files and whenever any process call any function/API in those DLLs, I will get an event
 
Would you please give us more insights the environment in which you're trying to implement this?
 
8:16 AM
@void_in Note sure I quite understand. Are you hooking the malicious DLLs, or the System DLLs?
And either way why wont the attacker just obfuscate that?
 
@AviD I am hooking system DLLs
 
@void_in and watching for any use by "unauthorized" programs?
at least according to history, it was possible to bypass that by channeling through an authorized program.
I do like the idea, though.
 
@AviD yes. Unauthorized program most of the time call those DLLs in order to achieve full code execution
@AviD the program is only going to channel through authorized programs if it is in complete control
 
Sounds interesting, I'm guessing the main issue will be managing the false positive rates.
 
@AviD But in order to get to that stage, it has to call a few DLLs
@TerryChia you are right. That is why just hookinga single DLL is not going to cut it. I need to hook a few and then look for calling those in some particular order
 
8:20 AM
Haven't really looked into the topic, but I wouldn't be surprised if some anti-virus vendors have already incorporated it into their heuristic detection methods.
@void_in Do share it on here if the project is successful. :)
 
@TerryChia honestly I have worked a lot on AV bypass and tell you the truth the state of AV is not as good as their sales people want us to believe
 
@TerryChia They have, at least NOD32.
Yeah, and also F-Secure.
 
@TerryChia sure. Right now I am detecting MS 12-020 exploit. Snort detects it over the wire and if during 10 seconds Nagios do not here from the machine, it means the attack was successful.
I just need now which machine was the launch pad. That is where the DLL hooking comes into play. First a certain machine process calling particular DLLs, suddenly snort detecting exploit code on the wire, and in 10 seconds the system for which the exploit code was destined is down. Seems the analysis has to start at the machine where the first event (DLLs) occured
A single technique can bypass MSE, Kaspersky, McAfee, Symentec, and F-Secure. those I have tested myself. MSE was by and large the most difficult
The problem is those AVs just hook DLLs in kernel32.dll (high level), so malware started calling the low level DLLs in user32dll.dll which function the same but bypass all the hooking by AV
The idea of ambush is if you find out a certain malware is calling a particular DLL that the AV or any other control is not detecting, you just hook it yourself.
 
@void_in not really. For example, DDE still works, at least on XP. If it's .NET, you can call that directly. Often standard EXE's will support an API, or at least a commandline.
these are the ways products like this have been bypassed in the past.
 
Every time I look at this "First rule of stack exchange: Be NICE" I remember Tyler Durden!
 
8:32 AM
I think you would need to give a good hard look at this, how you'll block / identify indirect accesses, if you want this product to work well.
 
@AviD And that is where you are going to hook LoadLibrary() as well.
 
but as I said, I'm interested in hearing how it goes :)
@void_in LoadLibrary wouldn't be used for any of those (at least not necessarily), and even if you did hook it you would be swamped, no?
@TerryChia yes, its been done, but not well.
@void_in heh, yeah a little bit of cognitive misdirection.
 
@AviD you know there is a thousand $$ protection that claims to provide a defense against APTs just by hooking the following DLLs
WinExec
CreateProcessW/A
CreateFileW/A
ShellExecute
UrlDownloadToFileW/A
UrlDownloadToCacheFileW/A
 
@void_in I'm not surprised. How well does it work?
 
@ThomasPornin When using PBKDF2 to derive a password, how many bytes should I grab from the algorithm before I base64 it?
 
8:37 AM
Although that was an epic failure blog.trailofbits.com/2012/10/29/… but the idea is it is doable with certain strings attached
 
I bet its worth at least 10.00$.
 
@AviD you are right. At least that much amount. But lets see where my "research" goes. At least you all will know it didn't work :)
 
@void_in cool, my breath is bated :-)
 
9:23 AM
nice answer @TerryChia
I haven't seen @Polynomial in a while on here
 
@LucasKauffman Thanks. :)
@LucasKauffman Probably busy getting drunk before he starts his job as a pen tester.
 
9:43 AM
@AviD @TerryChia another update on those surface Cables. Marion reckons that the surface RT one is propietary and that's what Microsofts docs seem to say although I saw an article that suggested it was Micro-HDMI. The Surface pro one whilst it is displayport based not all cables might fit as port is a bit angled so the housing on some cables gets in the way...
 
@RoryMcCune Thanks. tbh even when I do get around to buying one, I'm not too worried about that - the point of the surface is to be mobile, so I wouldnt be using it tethered to a monitor...
though it is nice to have options
e.g. a projector for lectures and presentations...
ugghh, dammit now I am bothered.
 
yeah that's the main use I see for it, although the missus is thinking about using it as a main PC hooked up to an external monitor...
 
@RoryMcCune Ahh interesting.. Funny how Apple gets all the hate for proprietary connectors.
 
yeah MS seem to be taking a leaf straight from the Apple playbook
 
@RoryMcCune That's the setup I was planning to recommend to a friend. The Surface RT seems nice for some one whose main computing needs revolve around a browser and Office..
 
9:47 AM
well the missuses review of the pro will be up on our blog later I reckon, I'll drop in a link when she finishes it..
 
speaking of which, I'm not sure which would be the best option for me to host VMs, for a home-VDI type setup.
Hosting on Windows 8 Pro (client hyper-v) / Windows Server 2012 (with HyperV) / Standalone Hyper-V server
 
@AviD well we have a Hyper-V setup which is reasonable but managing it is a huge pain
 
any suggestions?
@RoryMcCune how so?
 
it can't sit in the domain 'cause the DC is a guest on it so it's in it's own workgroup and the remote management tools dont have a good way to specify creds
there's docs on MS' site but they're stupidly complex
 
@RoryMcCune ah, right. That's actually something I was just thinking about...
 
9:49 AM
involve creating local accounts on the hyper-v host and then same named account on the guest and changing Dcom options manually
 
That's actually a point in favor of barebones hyper-v standalone, as long as the DC is on the VM.
(which would only be temporarily for me.)
 
you could look at ESXi which (last time I looked ) had a free version...
 
@RoryMcCune nooooo
actually I dont think my HV options are bad. Just not sure which is best.
 
@AviD RLY? I thought it was ok last time I used it...
 
@RoryMcCune well, its been a while for me...
 
9:51 AM
me too, and I didn't use it too heavily. I really like VMWare workstation but I don't use the server too much...
 
win8/client-HV would allow me to use the host as a regular machine, in addition to VMs, much simpler than a VM for simple ussage.
server 2012 would allow me some use, but other HV benefits, such as vGPU. (which I love the idea of - no idea in practice).
standalone HV seems the most efficient, but then there is no direct basic usage.
@RoryMcCune heh, I'm the other way - hate hate HATE the VM workstation, but having used the server only as a client it seemed okay.
 
@AviD The thing is with HV (if it works) is that a lot features are just free, for instance live migration. For which you pay mucho dineros in VMware
 
btw windows licenses are not an issue for me, I have access to most any license.
@LucasKauffman yeah, I saw a comparison. Also much much higher limits with HV (on server at least) than VMWare.
e.g. CPU, memory, sessions, etc...
also I'm assuming that it is possible to install non-MS versions as guests on HV, but if not that would be a substantial drawback.
btw @RoryMcCune I saw somewhere that an easier solution to the DC problem - is to just have one DC on metal. Makes sense, I guess, for larger shops anyway.
 
@AviD yea you can do that without problem, most used ones anyway, I think even Azure supports the penguin these days
 
@LucasKauffman cool
 
9:56 AM
@AviD yeah for big co's it's not a problem, but for smaller shops it's a bit of a pain..
 
@RoryMcCune on standalone HV, can you pull up one of the VMs locally? Or does it only run headless?
 
@AviD Not AFAIK it has a console locally but it's text based (cmd/powershell) so commandline stuff only
 
also a nice thing about HV which I haven't seen in practice yet is that you are supposedly able to outscale towards Azure. For instance lets say you expect 1000 times extra hits on your machines for 1 day, you can just purchase time on azure and your machines will loadbalance between local machines and machines on azure
which I think is a really cool feature if it works properly
 
@RoryMcCune so you cant access the guest from the local host?
interesting. That would mostly rule it out for me, then - I'm VDI'ing on my soon-to-be hulking desktop.
@LucasKauffman yes, very cool.
 
@AviD don't think so, you can use the powershell to start/stop stuff but you need the Hyper-V tools to manage it properly, so to an extent Win Server 2012 is a better option 'cause you can install the tools there...
 
10:00 AM
@RoryMcCune I don't mean for managing, I mean for using the guest.
but I guess there would be no GUI, so no RDP, no interactive sessions...
 
@AviD sorry yeah I was including that. getting a console on the guest uses the management tools. Obv. from another machine you can RDP to guests like normal, but I don't think you can from the Hyper-V host
just checked mstsc isn't installed on hyper-V host.
 
okay, thanks, so we've ruled that right out.
so which will I use as host - W8Pro (w/Client-HV) or Server2013 with full HV.
I think it may be possible nowadays to actually run games well in a VM, if the host is Server2013, thanks to things like vGPU.
But if not, I would want my host to be W8Pro.
I suppose... testing is the only way.
yay, reinstalling windows half a dozen times.... woot.
 
yeah I've not tried the games thing, it could be a possible...
 
@AviD i'm not familiar with vGPU but i've had all kinds of wierd issues with virtual graphics and i'd doubt anything works particularly well playing games. Would be cool to be proved wrong on that though...
I remember finding out that MsRDP switches graphics drivers when you connect which really stuffs things up...
 
@D3C4FF is that recent? Because things have changed a ton in recent versions.
 
10:07 AM
Last time i tried on win7 yep
couple months ago tops.
 
not on win8, though.
still, its likely that you're right.
 
sending 25FPS unbuffered (realtime) is quite a feat... i'd be very suprised if you can virtualise that near real-time
 
Do you want this question? Topic wise it seems to fit sec.se, but it probably needs a bit of editing.
-1
Q: Truecrypt compromised?

nitrlIn light of recent developments in computer forensics and "password recovery" utilities, can encryption technologies such as TrueCrypt still be relied on for a high degree of protection? It would seem to me as though the ability to retrieve decryption keys from a device's RAM (even after shutdow...

 
@CodesInChaos I think it's a decent question even as it is, although the asker is quite misguided.
 
@D3C4FF what if its local?
i.e. sitting at the host but working in the guest.
25? I'm sure we can get that to 60.
@CodesInChaos "can we trust XX?" -> not a great question.
Harvey Dent.
Can we trust him?
 
10:18 AM
The core of the question is rather a risk comparison of software vs. hardware disk encryption
 
I agree, think it could be here, but needs some real editing.
@CodesInChaos I agree.
dont have time to work that out now, though - but if anyone else does, I say go for it.
 
@AviD i think there'll still be issues. But i haven't tested. Certainly, there's issues with RDP. I've never gotten smooth framerates on anything, citrix has the same issues, VmWare also shit, i dunno about HV, VNC... maybe
 
@RoryMcCune here's another virtualization-related question - an updated machine today would be recommended to have at least an SSD for a system drive, right? But if the installed OS is just the host - and my "real" OS is running in a VM, does that still make sense?
Assuming the VM (of which there will be multiple) can't all fit on the SSD.
@D3C4FF assuming network isnt the issue (either local network or even on-box), I've had very good results with RDP. Also VNC.
RDP actually also degrades over a bad connection better than most.
 
@AviD Probably not, unless your host uses the SSD as as caching solution.
 
@TerryChia yeah, that's kinda what I was starting to be thinking. Better to use the SSD as a system disk cache, than as a system drive.
There are several technologies that do this now, either by chipset, or motherboard, or SSD tech.
 
10:29 AM
If the VMs have considerable overlap, a filesystem that does blockwise deduplication might allow you to fit all of them onto the SSD
 
@AviD well the SSD is useful for general OS speed but yeah host OS speed is less important. What I do is put a couple of commonly used VMs on my SSD and all the test/dev ones on a spinny disk..
 
@AviD Yeah, the newer intel motherboards has hardware support for it. Not sure how well it works though.
 
@AviD Hmmm, i've had disappointing performance using my desktop (fairly powerful) crossover cabled to my laptop when using RDP.
Also. @AviD I personally use a 512GB SSD for both a boot/OS drive and the drive i run my VM's off. Alternatively, i'll occasionally run VM's off of a USB3.0 drive with no noticeable performance issues.
 
@D3C4FF Holy, a 512GB SSD? Must have cost you a bomb.
 
hmm, yeah... Already ordered a smaller SSD.
Not sure how that will go.
I guess I could always upgrade it later...
but I think it could be possible to keep one or two basic VMs on the SSD. And, if they or the system gets too big, it would be simple to move it.
yay for virtualization!
 
10:34 AM
@AviD Could always grab another SSD and use it in RAID 0.
 
@RoryMcCune seperate partition, or single?
@TerryChia That's silly, for SSD.
If I grabbed another SSD, I'd have one for system, and one for VMs.
Or possibly one of them for system cache.
 
@AviD Just the one, but then I'm just using VMWare workstation, so just Dirs on the drive...
I've got a 256GB so OS files + a couple of VMs seems to fit ok
 
@RoryMcCune sounds good.
 
@TerryChia yeah it did when i bought it at the end of 2011... roughly 700 bucks
 
@D3C4FF daaang. Hope you maxxed out that baby, get your money's worth.
 
10:38 AM
But. I can run 3VMs and the OS all off the same drive without breaking a sweat (although RAM gets eaten up quick)
 
@D3C4FF thats why you need a shedload of RAM.
 
@AviD Hellz yeah ^_^ I'd just started my new job so with my first paycheck i built my 'dream machine'
 
:-)
 
I didn't eat for a while, but you don't need to eat when your online >_>
 
what you get for GPU?
 
10:40 AM
@D3C4FF You dream machine and you didn't get 64GB of RAM? Shame on you! ;)
 
@AviD GTX480
under water of course :D
 
@TerryChia hey, dream or no, it still makes sense to be cost effective.
 
@TerryChia I got 16 ;__;
 
RAM is dirt cheap
 
likely he wouldnt use the top 48 of those for a couple months, and RAM is one of the easiest upgrades.
assuming you plan ahead.
@LucasKauffman it's going up...
@D3C4FF you mean water cooled?
 
10:41 AM
@AviD Eh, I wouldn't call a 512GB SSD cost effective...
 
Yeah, i'm thinking about upgrading the RAM soon actually, especially the more i do photo editing
 
yea so better get those dimms while you can
 
@AviD Yeah, the GPU and CPU both hooked up to a custom rig.
 
I'm considering watercooling, not sure I'm ready to make that leap yet.
@D3C4FF oh cool, on the GPU too.
what do you mean custom, you build it yourself?
 
@AviD yeah, as much as i could. I don't have a proper metalworking shop, but i built the WC enclosure and wired up some of the custom electrics etc
 
10:43 AM
@D3C4FF wow, that is not something I would try any time soon.
especially not with liquids around my precious electronics...
 
@AviD liquids used to cool systems aren't dangerous for the system
 
@LucasKauffman I "know" that, but I just can't bring myself to "believe" that.
 
@AviD I learnt heaps about everything doing it, metalwork, PC hardware, etc.
 
like I said, I'm not quite ready to jump there yet.
Call me old school.
 
@AviD Liquidcooling is very safe if you go for those off-the-shelve solutions.
 
10:44 AM
technically deminiralized water should not conduct electricity
 
Old fogey.
 
@LucasKauffman not sure what you mean aren't dangerous... I use water >_>
 
@D3C4FF if you use deminilirized water normally the water should not conduct electricity anymore
 
Not as fun as doing everything yourself, but much more convenient.
 
if you get cooling liquids from most PC shops, they are often not water but some non-conducting liquid
 
10:46 AM
one of the things important to me in this build is upgradability.
 
oh. @LucasKauffman deminiralized/de-ionized water doesnt TECHNICALLY conduct electricity, but there's still other substances in the loop that eventually ionize the water over time and when i top up i add copper sulfate (i think) to kill any waterborn bacteria/germs that could build up and hinder flow
 
in the past I did not really take care of that, so other than perhaps some memory or hdd, I've not done a real upgrade on my machines.
 
@AviD using custom waterblocks for specific GPUs or CPU sockets limits that, but things like the corsair H80/H100 give good results and you can use in upgrades.
 
That, and I've been on laptop only for the past... hmm... 8 years???
 
reminds me of a guy I knew who used Aceton to cool his machine, he wouldn't understand why that was a very bad idea
 
10:47 AM
@D3C4FF yeah, was looking at those.
 
@AviD If that's the case, focus on getting a good motherboard and CPU first. Those are a real bitch to replace. Almost anything else is pretty simple to swap out nowadays.
 
probably wont get it, maybe next time.
@TerryChia yup.
AND case.
 
Even after I explained that it conducts electricity, that it's highly volatile and that its highly inflamable
 
@AviD They work well, just make sure they get seated right and that you've got one that isn't making much noise, a few friends reported an odd grinding sound...
 
the right PSU makes things easier too.
 
10:48 AM
@AviD Ah yes, case. Plus don't cheap out on the PSU either.
Lian Li makes some really awesome looking cases.
 
@TerryChia for the most part I'm following the school of Jeff on hardware.
He may be a lot of things, but one thing he isnt not is a hardware freak.
If he likes it, it's probably good enough for me - if I can afford it.
 
@LucasKauffman have you seen the custom phase change cooling systems? A couple hundred PSI of toxic gas flowing through your CPU block. Sets the CPU down to -50degC
 
Also TechReport have some great reccomendations.
 
@D3C4FF yea the ones with Nitro?
 
@TerryChia see, thats the thing - the best cases spend a lot on looking good. I don't really give much of a damn about how it looks - ideally it would be under the desk, or in the cabinet, or up on a shelf.
 
10:51 AM
or the ones with liquid helium?
 
and even if not - so what?
 
or the ones using CO2 ice cubes?
 
and not just the cases - the RAM, the GPU, etc... who cares?
@D3C4FF whoa.
have link?
 
@AviD Heh. If you don't care about looks, I can recommend Xigmatek cases. Inexpensive and works well, although it looks kinda plain. Using one right now.
 
@LucasKauffman liquid gases like freon, but in essence, any gas that can be compressed. I've seen dudes hook up actual aircon units and use them directly. Its intense
@AviD link to what?
 
10:53 AM
@TerryChia looked at those, didnt care how it looks esthetically, but I care about functionally.
I'm looking at a Corsair box...
 
last time I went with Antec, which was great, but Corsair really upped the box game.
 
imageshack.us/photo/my-images/844/20110307105123.jpg <- this is what my build looked like when it was first built, i've changed it a bit since but the idea stands.
 
Yeah, you can't go wrong with any of the big name brands tbh. Just make sure the case can fit everything you want into it and most will work fine.
 
@D3C4FF oooh I was thinking of doing that myself using an old refrigerator cooling, but then decided it would probably be a bad idea as I probably would kill myself in the process
 
10:54 AM
@D3C4FF I guess there's no OTS product yet...
 
cpu at -153 degrees celcius
 
@TerryChia I disagree. HUGE difference between cases that just "work fine" and "make it a pleasure to work with, options for upgrade, and keeps it cool and silent".
 
@AviD actually there are a few OTS solutions.
 
@D3C4FF very cool. ubergeeky.
 
There are premade enclosures which have to be humidity sealed
 
10:56 AM
@LucasKauffman "champion overclocker"?? There are competitions?
 
@AviD yep.
 
@LucasKauffman "... and a propane torch". Well, that's out for me, my wife would never agree to that.
She doesnt even let me use my saw very often.
Can't say I blame her.
 
I have just arrived at work and unlocked my screen. This is what I saw
 
@AviD Most of the cases I looked into from the big name brands are good to work with. Not really expensive either. The expensive ones are the ones that look good and are made of aluminium etc.
 
Since Friday night, 106 messages directed at me. Yaaay! :D
 
10:59 AM
@Adnan Do
@Adnan you
@Adnan want
@Adnan more?
 
We talk a lot!
 
@Adnan heya
@Adnan howarya?
@Adnan thats a lot
 
You have to wait until I leave work
 
@Adnan I'm doing what @Terry did
@Adnan I mean @TerryC
@Adnan I mean @TerryChia
 
:|
 
11:00 AM
hehe
 
Let's read the transcript..
Anything interesting?
 
@avid Close this?
-1
Q: Python open source Exploitation and Post-exploitation Tools

Soceritsi am working on an automated penetration testing system, i have selected openvas as my vulnerability scanner and i will use the python module openvas.oaplib after i will identify the vulnerabilities, i will exploit the victim one by one, i have searched for an exploiting tool to use, but i can't ...

 
done.
 
> With such tools, you are not targeting "non-cryptographers who should know better than doing homemade schemes", but rather "chimpanzees who should not be allowed to touch a keyboard".
Brilliant.
 
@TerryChia bear?
 
11:08 AM
@AviD Yes. It's his trademark style.
 
of course.
context link?
 
1
A: What statistics can be used to identify pseudorandom data?

Thomas PorninDuring the AES competition, the organizing body (the NIST) got to the trouble of running extensive statistical testings on the output of the 15 submitted block ciphers, with what was believed to be the gold standard of such tests, the Diehard tests. They found nothing, of course. Comments from cr...

 
@TerryChia Usually PBKDF2 is to derive a key from a password, not the other way round. But a Base64-encoded key can serve as a password (as long as you don't have to type it often).
Passwords are as good as they cannot be guessed, if only by exhaustive search, and 100 bits from a good KDF are sufficient to reach that level.
 
@ThomasPornin Oh, I think I didn't phrase my question very well. I'm using PBKDF2 to hash a password. How many bits should I grab from the PBKDF2 function before doing a base64 and storing it in my database?
The implementation I am looking at uses 64 bits. I'm not sure what the reasoning is behind that particular number.
 
@TerryChia well, 100 bits again are enough, but go to 128 bits because people just love powers of two; that's like the red paint on a Ferrari.
If people use 64 bytes then they are American or want to be American.
 
11:13 AM
Blah, bits. Mistyped.
@ThomasPornin Alright, thanks.
 
> you are not targeting "non-cryptographers who should know better than doing homemade schemes", but rather "chimpanzees who should not be allowed to touch a keyboard".
Brilliant.
 
@Ninefingers Look up in the chat. ;)
 
Ok, 64 bits, that might be considered a bit low, but not serious. It means that the effort for finding a matching password is at most 2^64 guesses -- but this matters only if you can get your users to choose passwords with 64 bits of entropy, which would be quite a feat.
 
@TerryChia Haha, we had the exact same reaction and quoted the exact same phrase. What are the chances.
 
@Ninefingers after hanging out in this room as much as you guys do? Pretty good chances.
 
11:19 AM
@Ninefingers The chances are rather high if one of you is the sockpuppet of the other (or both are sockpuppets of @Rory)
 
@ThomasPornin hey hey hey, I thought I was @Rory's sockpuppet.
now @TerryChia being @Ninefingers' sockpuppet, that makes sense.
Since @TerryChia is only about as old as this site, anyway.
 
Oy you two no ambiguous @Rory 'ing I take it you're referring to @RoryAlsop as I would never do such a thing :op
 
hehe
 
@RoryMcCune Ambiguous Rorying? Is there such a thing?
 
11:22 AM
@Ninefingers I think all Rorying is unambiguous.
 
@TerryChia ahh but she's been verified by @Polynomial , so your sockpuppetry slander falls flat ;oP
 
You should know when you've gotten roried.
 
@AviD @TerryChia is not my sockpuppet, and I am not @TerryChia's sockpuppet. I can neither confirm nor deny that I am in fact (or am not in fact) @RoryAlsop's sockpuppet.
 
@AviD well @Rory is ambiguous Rory'ing to an extent... but most rorying is, I think fairly unambiguous
 
@RoryMcCune Rory again?
 
11:29 AM
@TerryChia At most the natural hash size. At least your desired security level. 64 bit is a bit on the low side, but the average password is still weaker, even with strengthening. Personally I'd go with 128 bits.
It's generally possible to brute-force crypto with a 64 bit level, but it's a bit expensive
A state level attacker can probably push to around 80 bits
128 bits is the typical "don't make me think" level.
@ThomasPornin Do you know if RC4 passes die-harder?
 
@CodesInChaos Diehard will detect the bias on the second byte if you take many very small RC4 output streams and concatenate them together.
As far as I know, Diehard does not detect the other known biases of RC4
 
11:44 AM
Dave's hash is getting too much crap IMO. It isn't weaker than other salted single-iteration/fast schemes.
 
@Adnan I can't keep up with all the stuff from our most vocal user... I tried reading up last night, but didn't finish the chat and haven't nearly begun the manifesto we got...
 
Manifesto?
 
@JeffFerland do you mean @LucasKauffman?
 
@AviD I mean I'm not awake. Yesterday was rough. I saw an A at the beginning of the name of the post I replied to and it thought it was you.
 
@JeffFerland I meant which most vocal user?
 
11:58 AM
@AviD I think he meant He Who Just Got Suspended.
 
oy. I KNOW WHO HE MEANT. I was just banging on @LucasKauffman.
cmon, this isnt new ground here.
 
Ahh ok. Brain not working well.
 
maybe I should just stick to using the same victims. But I really like spreading the vile around.
hehe
@TerryChia no, if two of you missed my point, it must have been too subtle. Mea culpa.
 

« first day (816 days earlier)      last day (4064 days later) »