The DMZ

Mark Buffalo

13:03

@BarryCarlyon It's Barry! Just in time for his 2nd message of the month.

diagprov

@Adi It seems so.

@AviD It's the deep web because it travels under the sea in cables :)

Mark Buffalo

I visited the Deep Web once... left in horror.

M'vy

I was in a Deep S... oh wait. That's not it.

AviD

@MarkBuffalo I don't always visit the deep web, but when I do...

Rory Alsop

@SEJPM why would you sleep for moths?

SEJPM

13:09

@RoryAlsop cause bears do that

Journeyman Geek

moths?

SEJPM

@JourneymanGeek typo :/

meant to write "months"

Rory Alsop

:-)

BlueBerry - Vignesh4303

Rory Alsop

Bleichenbacher!

BlueBerry - Vignesh4303

13:12

arstechnica.com/security/2016/03/…

SEJPM

@BlueBerry-Vignesh4303 why would anybody use / allow 40-bit encryption!?

Mark Buffalo

@BlueBerry-Vignesh4303 :-D

diagprov

Oh man another TLS bug?

BlueBerry - Vignesh4303

@SEJPM sorry friend,i m little new to encryption,decryption,just observing the discussion of dmz and learning here

SEJPM

@BlueBerry-Vignesh4303 that's a different question. I can probably brute force 40 bit encryption on a cheap EC2 instance. I can't for 128-bit and definitely can't for 256-bit

diagprov

13:15

2

Q: SSL cipher suite: what does "export" mean?

I saw a cipher suite EXP-EDH-RSA-DES-CBC-SHA, EXP stands for export, How do I explain the "export"? What is the different between "EXP-EDH-RSA-DES-CBC-SHA" and "EDH-RSA-DES-CBC-SHA"?

BlueBerry - Vignesh4303

@SEJPM arent u refering to this one?

40-bit encryption

40-bit encryption refers to a key size of forty bits, or five bytes, for symmetric encryption; this represents a relatively low level of security. A forty bit length corresponds to a total of 240 possible keys. Although this is a large number in human terms (about a trillion, nearly two hundred times the world's human population), it is possible to break this degree of encryption using a moderate amount of computing power in a brute-force attack, i.e., trying out each possible key in turn. == Description == A typical home computer in 2004 could brute-force a 40-bit key in a little under two weeks...

Mark Buffalo

@ThomasPornin Save us. arstechnica.com/security/2016/03/…

BlueBerry - Vignesh4303

@diagprov yep it seems to be a tls bug

Simon

Sometimes, I wonder if Markus reads only what he posts, or what is directly said to him.

diagprov

@MarkBuffalo He already has. He's just waiting for us to finish TLS 1.3 then he'll tell us if we got it right.

SEJPM

13:18

@BlueBerry-Vignesh4303 it's an SSLv2 issue

you can recover the key from clever queries :/

apparently many configs didn't disable SSLv2

BlueBerry - Vignesh4303

previous attacks claimed as crime & beast

how it differs from previous one's

SEJPM

CRIME and BEAST are on the TLS compression IIRC

i.e. direct TLS flaws

this is a SSLv2 flaw that can make you learn the static key and break TLS using this key

Mark Buffalo

You know....

Who released this?

SEJPM

it's like CRIME, BEAST and FREAK in the scale of server's affected

@MarkBuffalo couldn't read the whole article yet :/

Mark Buffalo

I'm looking for the researchers

BlueBerry - Vignesh4303

13:21

@MarkBuffalo myself searching whom released it :(,i m sure google guys doesnt released it

Mark Buffalo

This makes me wonder if it's the Berkeley.edu guys

BlueBerry - Vignesh4303

i m looking at their google projectzero no updates :(

Mark Buffalo

They've been scanning the entire internet lately

SEJPM

@MarkBuffalo pretty standard?

BlueBerry - Vignesh4303

drownattack.com

openssl.org/news/newslog.html

diagprov

13:22

@SEJPM Not quite. Beast exploited the fact that the IV of the next TLS record (run of CBC-encrypted data) was initialized from the last block of the previous connection, i.e. it was predictable. Consequently if you can control the first block (browser) you can manipulate the ciphertext and reveal data. Google "here come the xor ninjas".

SEJPM

blog.cryptographyengineering.com/2016/03/…

Francesco

hello everybody :)

SEJPM

Nimrod Aviram
1
, Sebastian Schinzel
2
, Juraj Somorovsky
3
, Nadia Heninger
4
, Maik Dankel
2
,
Jens Steube
5
, Luke Valenta
4
, David Adrian
6
, J. Alex Halderman
6
, Viktor Dukhovni
7
,
Emilia Käsper
8
, Shaanan Cohney
4
, Susanne Engels
3
, Christof Paar
3
and Yuval Shavitt
1

Simon

If only people would bother configuration their SSL instance with the Mozilla suggested cipher suites.

There's a damn page dedicated to that.

SEJPM

more than one :(

diagprov

13:23

CRIME is compression, namely that if you can control part of some unknown data, some values will compress better than others and you can guess the unknown data, even if encrypted, by trying various values and observing the length.

Mark Buffalo

@SEJPM Anyone at berkeley.edu?

Simon

@SEJPM True.

BlueBerry - Vignesh4303

To put things succinctly: with access to a whole hell of a lot of computation, an attacker can intercept a TLS connection, then at their leisure make many thousands of queries to the SSLv2-enabled server, and decrypt that connection. The "general DROWN" attack actually requires watching about 1,000 TLS handshakes to find a vulnerable RSA ciphertext, about 40,000 queries to the server, and about 2^50 offline operations.

this makes sense

diagprov

@Francesco Welcome.

SEJPM

@MarkBuffalo don't know

Mark Buffalo

13:24

@schroeder You awake?

SEJPM

@BlueBerry-Vignesh4303 quite a lot but feasible

Mark Buffalo

That's pretty feasible for a large website with lots of customers =\

SEJPM

maybe a few thousand dollars on EC2

and afterwards you can impersonate them and decrypt non forward secrecy traffic :/

Simon

Who's gonna be the first one to post a question about drown?

Mark Buffalo

@SEJPM I googled "drown attack ssl researchers" to try and find the researchers at first glance

Simon

13:26

Should we be typing DROWN?

Mark Buffalo

and google suggested, "drone attack ssl researchers"

SEJPM

@MarkBuffalo drownattack.com/drown-attack-paper.pdf

Mark Buffalo

That site is inaccessible here

SEJPM

@Simon yes

Mark Buffalo

:L

powers up phone

Matthew

13:27

@MarkBuffalo At least it didn't reply "good idea. We've got the ex-Oracle CISO working here now, so it's our new policy!"

SEJPM

@Simon i try to be this time :P

BlueBerry - Vignesh4303

@MarkBuffalo test.drownattack.com

Francesco

Uff I hoped to be able to ask a simple shopping question to you and instead I find out that you are discussing a new urgent threat ...

Rory Alsop

It happens...

Simon

#urgent

Rory Alsop

13:28

@Simon your face is urgent :-)

Francesco

Now as it Always happens with your links I will have to read and try to understand. So long for a peaceful afternoon...

Simon

@RoryAlsop :OOOO

I like it when these new vulnerabilities are released and I don't have to do shit because I did enough research before setting up servers.

Am I a bad person for testing if SSLv2 was deactivated on drownattack.com?

Rory Alsop

Lol

SEJPM

done

AviD

@Simon yes, but nothing to do with SSLv2

Rory Alsop

13:31

;-P

Simon

@AviD ovid pls

AviD

@Simon yessir

BlueBerry - Vignesh4303

blog.qualys.com/ssllabs/2016/03/01/…

any exploits available for testing this drowing attack in my server :(?@MarkBuffalo

Simon

@SEJPM YOU REP WHORE

2

0

Q: What is DROWN and why is it even possible?

There is a new recent attack "on TLS" named "DROWN". I understand that it appears to use bad SSLv2 requests to recover static (certificate) keys. My question is: HOW? How can you recover static encryption or signature keys using SSLv2? Bonus question(s): How can I prevent the attack from app...

tls openssl

SEJPM

@Simon thanks.

BlueBerry - Vignesh4303

13:35

nice one buddy

Francesco

So if the question has been asked, can I proceed with my shopping list recommendation? Pretty please?

SEJPM

@Simon now wait for @ThomasPornin to answer that and get +1000 votes

Thomas Pornin

@Simon I am on it

Rоry McCune

ranum.com/editorials/corporate-songs/Symantec_Revolution.mp3 <-- there's some "quality audio production" in here, I'm sure @RoryAlsop and @kalina could learn a thing or two from these lyrics

Xander

@AviD For myself, I say: Go look at the first version of the question. That was the one I was presented with.

Mark Buffalo

13:37

@SEJPM luls, I am in the process of posting it

But you beat me to it

Simon

@ThomasPornin Mind answering it on my account?

Mark Buffalo

grats on the rep train

Rоry McCune

@Simon pls we've been through this, "person of negotiable reputation"

Simon

@RоryMcCune Yes, yes, sorry, that's the polite name.

Ohnana

@AviD when i read the question in it's original, sillier form, i understood it as "i want to program a thing to defend my application" which is "programming" which is SO's domain

AviD

13:39

@Xander hmm. Its still not anything to do with coding, but it is sufficiently confused enough that I will excuse you from my mockery.

Ohnana

it was quite unclear, and it appears that many people read it different ways

Simon

@Ohnana You're quite unclear.

Ohnana

@Simon occasionally

Simon

me2

AviD

@Ohnana not really. "how to defend my application" is totally sec.se. "This is what I want to implement, what code should I write" is SO.

@Ohnana that is agreed.

Ohnana

13:40

@AviD yea, and i read it as the second one

minus the explicit "plz gib codez"

AviD

like I said to @xander, it is sufficiently unclear that total mental confusion is possible, despite there being no indication whatsoever to coding in the original version.

Xander

@AviD All I got out of it was that he wanted to add a button to an application to test a printer. There was some other drivel, but I couldn't make heads or tails of it.

AviD

yeah at that point people tended to get lost

Mark Buffalo

Yeah, anyone with moderate coding experience in their language will know exactly how to do that

AviD

still, I found that migration to be surprising.

Mark Buffalo

13:42

It's not hard to add a button to an application, and have an event handler receive the click

AviD

besides, if its so confusing, it shouldnt be migrated anyway

Mark Buffalo

The harder part for him is opening a connection to the printer... he should've asked how to do that

AviD

remember the #1 rule of migrating: Don't migrate crap.

Xander

@AviD Yeah, in hindsight, I agree.

AviD

@MarkBuffalo so, you've limited it to like 20% of programmers then.

Rоry McCune

13:43

@AviD I thought the number one rule of migrating was "Chuck it all, #YOLO"

Xander

@AviD I was biased by the title.

Mark Buffalo

@AviD I dunno... when learning programming, creating a button on my application was the first thing I learned to do :/

Simon

@RоryMcCune It's like the product question rule: it changes constantly.

Xander

On its own, it seemed clear enough.

AviD

@Xander fairynuff.

@MarkBuffalo exactly.

Rоry McCune

13:44

@Simon ooh dem's fightin' words, I hope you have your tigger proof suit on

Simon

@RоryMcCune Yes, I have the "I'm not getting into a fight with ovid today" state of mind.

Rоry McCune

@Simon BTW theregister.co.uk/2016/02/29/… big article on iOS malware here... oh wait it's not , it's about Android malware :op

woopwoopwoop

Simon

o ffs roro

Rоry McCune

@Simon I'm helping you out, if you can avoid rising to the bait of an android troll, you can more easily resist @AviD when he trolls you

Simon

Very true.

AviD

13:46

@RоryMcCune noone can resist me.

Mark Buffalo

@RоryMcCune Heh... another android crapfest

Ohnana

@RоryMcCune malware that you have to intentionally download

Francesco

So I'd like to tap your knowledge with regard to password managers. I know enough about security to know that I am unable to appraise if they are doing proper things or not. I have found out 1password, keepass and password safe (which is by bruce schneier who is Always very interesting to read. But for some reason password safe appears to be not much famous)

AviD

@RоryMcCune I dont get it.

Francesco

The last 2 are OSS which seems a good feature. Could you please help me to assess if any of these have glaring problems (I suppose not) or if one of those is much above the Others?

Mark Buffalo

13:47

@Ohnana Which a lot of people will :(

Ohnana

@Francesco love me some keepass2

Rоry McCune

@Ohnana sure like a lot of malware... it tricks you.. tricksy malware providers...

Ohnana

keepass is nice because it's a stand-alone database

Rоry McCune

@AviD @Simon is suggesting that you are inconsistent in your application of the site's rules

SEJPM

@Francesco personally I use KeePass

AviD

13:48

@RоryMcCune wat?? noe.

Ohnana

you can email it, IM it, save on your desktop, keep it on a usb key... and keepass also has a portable windows client

Rоry McCune

I was suggesting that this was (to use the vernacular) "spoiling for a rumble"

AviD

he is inconsistent in understanding the concept of subtle context.

Simon

Maybe I should have said it in Egyptian.

Ohnana

bird bird squiily line feather man facing left

Simon

13:48

I need to find some hieroglyph unicodes.

🍩🍩🍩🍩🍩🍩🍩🍩🍩🍩🍩🍩🍩🍩🍩🍩🍩

SEJPM

KeePass has HW accelerated pw key derivation (making attacks harder) has nice extensibility, good defensive strategies and some nice official recommendations (inclouding by the german BSI)

Simon

There.

Francesco

@Ohnana @SEJPM thanks, +2 for keepass. I like it being open source, and by the way in the feature list I see that they mention bruce schneier "KeePass can import TXT files created by Bruce Schneier's Password Safe v2" (from keepass.info/features.html)

Ohnana

@Francesco there you go

keepass2 has a native linux client as well if you're interested in that

Simon

yh it does

Ohnana

13:51

and it auto-generates passwords + allows you to store other secrets easily

SEJPM

and a nice android client

Ohnana

o rly??

link me plz

Simon

@SEJPM By nice you mean super not user friendly?

Francesco

Do you use it also on the smartphone?

Mark Buffalo

@SEJPM Does the android client require full permissions? haha

Francesco

13:51

@SEJPM ah great

SEJPM

@Ohnana play.google.com/store/apps/…

@Simon I mean "nice". Not great but nice.

Simon

@SEJPM Hum that's better than the one I'm using.

SEJPM

It has pre compiled support for yubico OTPs, for many storage providers for key files and all that good stuff

and it has actual defenses against attackers by bringing in its own keyboard

Mark Buffalo

> Keepass2Android ist ein open-source Passwort-Manager für Android.

Ohnana

@SEJPM YOOOOO

@MarkBuffalo change the de url parameter to "en"

Simon

13:53

@Ohnana That's literally the first thing I did.

I feel like Markus is a scam.

SEJPM

sorry for the german version :(

Mark Buffalo

Yeah, I know... I actually read that in English without realizing it was German

Simon

Also, the sentence he has linked is perfectly understandable if I don't speak a single word of German.

Mark Buffalo

lol

SEJPM

@MarkBuffalo storage and contacts

it needs storage for the key files I guess...

Mark Buffalo

13:55

Yeah, but contacts?

Ohnana

@SEJPM hey hey this is the internet. everyone should be speaking american.

Simon

@Ohnana pls.

SEJPM

@MarkBuffalo to connect to your accounts (i.e. google drive)

Francesco

From the download page, the precise definition of what one appreciates in a password manager "If you get a warning that the signature of a downloaded file is corrupt/invalid: this is a false alarm and can be ignored. " :-D

Mark Buffalo

@SEJPM Wouldn't that be accounts, not contacts?

Simon

13:55

@SEJPM Are you from Berlin?

Ohnana

@Francesco er what

Francesco

@Ohnana keepass.info/download.html

SEJPM

@Simon no

Francesco

@Ohnana keepass.info/help/kb/faq.html#siginv

Simon

@SEJPM You suck, why aren't you from Berlin?

Ohnana

13:56

@Francesco yeah, read that

Simon

Who's gonna bring me on a urban exploration trip in Berlin now?

Ohnana

goofy MS browsers

SEJPM

@Simon I know that I suck, but Berlin sucks more

Simon

@SEJPM !!!!!!!!!!!!!!

DUDE THE CLUBS

Mark Buffalo

> This is a false alarm by Microsoft browsers (Edge and Internet Explorer). The digital signatures of the file are neither corrupt nor invalid. So, the warning can be ignored.

Francesco

13:57

@SEJPM hey when I was in Berlin I was amazed by the city. It was Amazing, much Beyond what I expected. On the other hand food was .. ok food was like I imagined it :-)

Mark Buffalo

@Francesco Good or bad?

Francesco

@MarkBuffalo I know, but it is amusing to read such a warning label on the download page of a password manager

SEJPM

@MarkBuffalo idk about android but if you don't want the app to have that right use the offline version play.google.com/store/apps/…

Simon

@Francesco You fancy little Italian pony :p

Francesco

@MarkBuffalo maybe I was simply too old. If I had gone when I was younger eating wurstel fries and cabbage for 3 weeks straight would have been much easier

Mark Buffalo

13:59

@Francesco I would rather lose my accounts than use a password manager

AviD

@Simon be nice to the newcomers.

Mark Buffalo

I don't store anything useful on any of my passworded accounts anyway, so I don't care

Simon

@AviD .........................

Francesco

@Simon really the worst was carbonated water :-)

Mark Buffalo

@Francesco Maybe you ate at the wrong place(s)?

AviD

13:59

@MarkBuffalo you're wrong

Simon

I'm willing to argue that calling someone a pony is more offensive than "prick".

Transcript for