Did you update your router for the WPA2/PSK KRACK nonce re-use attack yet?

Note that the fix for Krack was not a fix in the distributions, but a fix to wpa_supplicant, an external program. So the key person who should be notified was the developer of wpa_supplicant. Note that the "zero password" problem, probably the worst of the lot, could have been fixed privately as if it were a minor improvement (eg instead of zeroing the password, it could have been filled with random chaacters and released without inciting much suspicion. Of course making sure that users actually upgraded would have been a challenge without the urgency of it being a major flaw that could be attacked.

Who are you talking about here? There is a big difference between a bug which only annoys and a bug which is a security issue.

In some cases it may be possible to run alternative firmware, such as dd-wrt or tomato, once the appropriate versions for your router have been patched.

--
----------------------------------------------------------------------------- 
  Roger Blake (Posts from Google Groups killfiled due to excess spam.) 

  NSA sedition and treason        -- http://www.DeathToNSAthugs.com 
  Don't talk to cops!             -- http://www.DontTalkToCops.com 
  Badges don't grant extra rights -- http://www.CopBlock.org 
-----------------------------------------------------------------------------

He who is William Unruh said on Wed, 18 Oct 2017 18:25:42 -0000 (UTC):

Thanks William (and Marek), for explaining what the problem is, but what did the researcher propose as the *solution* for open-source code?

Did he propose that OpenBSD *wait* until the announcement 50 days later? How is the researcher going to *enforce* that 50-day waiting period?

I see you answered my fundamental confusion (see below).

1. Researcher finds bug on day 0 & plans to announce it 50 days later.
2. OpenSource community has to *wait* until the announcement to ship fixes.
3. Closed-source community can ship when? (any time or wait the 50 days?)

If that's the rules, it seems like it's going to be difficult to *enforce*.

Thank you for confirming he wanted OpenBSD to sit and wait before releasing the code. I was worried that some *other* researcher ran a diff and had to "sit on his discovery of that diff" which would have revealed to the seconde researcher what the flaw in wpa_supplicant was.

What you're telling me is that nobody did that manual third-party "diff" of the source code so it wasn't revealed in the wild to a third party to our knowledge before the 50-day waiting period was up.

(Note Marek said 60 days but I think the researchers mentioned only 50 days but let's not quibble if either one of us is wrong as it's close enough.)

That would mean *every* open-source vendor would have to "sit on the fix" until the announcement. That's fine if the researcher can enforce that.

I guess that is what the standard *should* be but who decides such things?

Ah. THANK YOU FOR EXPLAINING. I *knew* there was regret, but to me, a "diff" is something a third party does of the open-source code to figure out what's different. The guy who wrote the code doesn't have to run a "diff" because he *knows* what he wrote.

So I thought a third party who accidentally found out about the bug by doing a "diff" on the open-source code had to "sit" on it. But that didn't make sense given the rest of the conversation.

So THANK YOU for explaining that: A. Researcher finds bug on day 0 & gives all vendors 50 days to fix. B. OpenBSD fixes it early and asks for permission to ship the code. C. Researcher provides permission but then regrets that decision.

In the future, I guess, researcher wishes to deny *permission* of open-source code to ship the fix early, which is a moral conundrum indeed.

And how does the researcher *enforce* this denial of permission to ship open-source code?

I see the moral conundrum which pits the visibility of open-source code against the obfuscation of proprietary code for the case of a knowledgeable bad guy...

I. In open-source code, a bad guy can do a *diff* to see what changed. II. If something interesting changed, a bad guy can take advantage of it. III. In effect, they get to have their own personal 0-day vulnerability.

For the price of a "diff", the bad guy gets his own 0-day vulnerability. It's a moral conundrum I had never even thought about until today.

He who is William Unruh said on Wed, 18 Oct 2017 18:41:25 -0000 (UTC):

Thanks William (although I know you're responding to someone else) for bringing up a second moral conundrum, which is how many people to tell.

Since we all know that a secret is no longer a secret if everyone knows about it, do they normally enforce secrecy rules among *all* developers of code?

I can imagine, for example, that the Chinese government has at least one programmer in their employ at *every* software company in the USA (as just one example), where that programmer is a sleeper (which is trivial for them to do so I can't believe they don't do it so I assume that they do).

The moral question that arises is: Do you tell that sleeper if he is NOT the developer of "wpa_supplicant"?

If we couple the fact that the bad guys (e.g., the Russian mafia, the Communist sleepers, the NSA/CIA/FBI/GCHQ, etc.,) have an immense bankroll with the fact that a 'diff' is so obvious to run on open-source code, I don't know what the *standard* solution is when it gets down to details.

William and Marek already said that the researcher wanted OpenBSD to sit on the release of the fix until the vulnerability was announced, which opens up "fork" issues (which may be minor though).

Worse - it opens up the "sleeper" issue depending on what the "need to know" classified level of who gets to know about the bug before it's released.

Does everyone sign an NDA before they're told about the bug? Who enforces when/if that NDA is broken?

Policies like this can be maintained by requiring recipients of vulnerability predisclosures to agree to maintain confidentiality prior to embargo dates. An organization could choose to break that agreement, but they shouldn?t expect to be trusted with predisclosures in the future if they do so.

You can inspect object code to discover what the bugs were, too. A high-profile example was the 2015 analysis of the Juniper RNG vulnerability, where no source code was required.

--
https://www.greenend.org.uk/rjk/

Yes, wait. Enforcement is a non-issue. He cannot force compliance. He can however let it be known what they did which will then cause others not to tell OpenBSD anything when the next problem arises. The people doing open source distros are responsible people and their intention is not to make it easy for the bad guys to do nasty things.

And that made Theo uncomfortable, so he asked permission to ship early. He did not simply go ahead and do it.

WE will never know for sure of course. But there was apparently no evidence that this crack was actuallyused in the wild. Of course NSA may have discovered it 5 years ago, or perhaps ISIS did.

What is your hang up about "enforce"?

AGain you are hung up on a "enforce"/central authority thing. People can behave well all on their own, without legal sanctions or centralized laws.

The permission was to release a patch for the problem. The bad guys could than run a diff on the released new code to see what was changed. That is where the diff comes in.

The worry was that by releasing a patched system, it made it easier for some bad guy to discover there was a problem and what it was.

And again. Note that thousands of security holes have been found and annnouncent embargoed in the past. So there is a history of the open source community acting responsibly.

Yes, but many many others have thought about it.

They are slow to fix usability issues, but faster to fix security issues.

--
Personal attacks from those who troll show their own insecurity. They cannot 
use reason to show the message to be wrong so they try to feel somehow 
superior by attacking the messenger. 

They cling to their attacks and ignore the message time and time again.

He who is William Unruh said on Thu, 19 Oct 2017 01:09:07 -0000 (UTC):

Thanks for explaining the situation, and the process.

Does anyone know what *version* of wpa_supplicant contains the KRACK fix?

LINUX: x@x:~$ which wpa_supplicant /sbin/wpa_supplicant x@x:~$ !$ -version wpa_supplicant -version wpa_supplicant v2.4 Copyright (c) 2003-2015, Jouni Malinen and contributors

WINDOWS: c:\> wpa_supplicant 'wpa_supplicant' is not recognized as an internal or external command, operable program or batch file.

None. It is at present a bunch of patches. However, most distributions have pushed out a fixed version by now. On Mangeia it is 2.6-1.1 on Mageia 6, 2.6-1 on Mageia 5 and 2.6-3 on Mageai Cauldron. You do not tell us which Linux you are refering to. Note that 2.4 probably does not have the fix. But it all depends on your distro.

Windows does not use wpa_supplicant. That is an open source wpa program. It is used by Android, but again it is probably not a "stock" version, but may well have been altered by the phone manufacturer to include special features of the wireless chipset they use.

He who is William Unruh said on Sat, 21 Oct 2017 00:57:13 -0000 (UTC):

Thanks William for that detailed technical answer.

I apologize for not providing complete information.

$ lsb_release -a No LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu 16.04.2 LTS Release: 16.04 Codename: xenial $ cat /etc/*release | grep VERSION VERSION="16.04.2 LTS (Xenial Xerus)"

Thanks William. It's so refreshing to deal with actual technical answers!

Here is what my Android 4.3 phone just reported:

$ wpa_supplication -version wpa_supplicant v2.1-d15el-4.3 2015-08-13/21:13:54 Copyright (c) 2003-2014, Jouni Malinen and contributors

Debian (on which Ubuntu is based) came out with those patches on Monday I think. Ie, the version on the Ubuntu sight by Wed was probably the patched version. If yours was installed befor Wed it almost certainly is not patched. Definitely if it was before Mon the 16th it is not patched.

It really says nothing. They could have patched that version of 2.1 Or probably not. Although there is a claim in the original that it was wpa_supplicant 2.4 and later that were problematic. And that it is android 6 and later that is problematic. I have no idea whether that is because those are the only ones tested, or because earlier ones are OK.

You need 2.4-0ubuntu6.2.

Use dpkg -l to check the package version.

--
https://www.greenend.org.uk/rjk/

Or apt-cache:

~$ apt-cache policy wpasupplicant wpasupplicant: Installed: 2.4-0ubuntu6.2 Candidate: 2.4-0ubuntu6.2 Version table: *** 2.4-0ubuntu6.2 500 500

xenial-updates/main amd64 Packages 500 xenial-security/main amd64 Packages 100 /var/lib/dpkg/status 2.4-0ubuntu6 500 500 xenial/main amd64 Packages

Also latest version of 16.04 is:

$ lsb_release -d Description: Ubuntu 16.04.3 LTS

sudo apt-get update && sudo apt-get dist-upgrade

--
Take care, 

Jonathan 
------------------- 
LITTLE WORKS STUDIO 
http://www.LittleWorksStudio.com

Patches at the access point end are relatively unimportant.

What you need to patch most are your client devices.

--

Brian Gregory (in the UK). 
To email me please remove all the letter vee from my email address.

LEDE 17.01.4 includes the KRACK fixes, no point not upgrading.

True, but it includes an optional AP-side workaround for devices that haven't been (may never be?) patched ...

How does this colloquial summary for my family look - in case you want to send one to YOUR family? ======== People are asking what to do about the KRACK Attack vulnerability (note the pleonasm), so I figured I'd let everyone know what it is & I figured I'd give folks the opportunity to ask question if they're concerned.

The canonical site for the attack is written by the white hat who found it:

Here's my ad-hoc summary, written with respect to what you and I need to know & do.

1. In May, the white hat notified the government & vendors he found a bug in all WPA WiFi (e.g., WPA2) where someone who is *close* enough to intercept the signals can see everything you do.

1. It affects all WiFi but the worst affected is Android at or over version

6, macOS, Linux, and really fast (i.e., 802.11r fast roaming) routers set up as repeaters (i.e., as a second router).

Far less affected are iPhones, WiFi iPads, WiFi iPods, older Android devices, Windows computers, and normal routers (e.g., 802.11n or 802.11ac), especially if they're set up as the main router (and not as a repeater).

1. There is only one viable solution, which is to *update* your device firmware or software, whether that be a mobile phone, a laptop, a desktop, a router working as a repeater, or the main router.

The order of priority should be: a. If you have Android 6+, then you *should* update soon. b. If you have MacOS or Linux, then you should update soon. c. If you have an 802.11r router, then you should update soon.

You can take your sweet time on everything else, but everything needs to be updated.

1. The problem, of course, is *how* to update each device. a. First look for your device to see if there is an update b. Then try to find the update c. Then update.

What a pain. Let me know if you have questions. ========

No, whose interception point is close enough. Thus your wirelessly connected fridge, which usually has attrocious security, could possibly be used as an interception point for an attacker who is in Mongolia say.

All wifi is susceptible, including Windows. The problem with Android and Linux is they all use wpa_supplicant and it has a problem that, for security, it zeros the password after using it. But that means that when the replay occurs it uses that 0 password. Thus fixing wpa_supplicant fixes the problem, and it has in principle been fixed. Of course one needs to get that fixed version into the devices.

The problem is a client side problem. But all of those devices are susceptible to at least some version of Krack, and should not assumed to be "far less affected". They are all affected. That they do not have the "zero password" problem is irrelevant since they have other attack vectors.

Yes, for all of them, not just Linux and Android.

If you have any wireless device you should update soon. I have no idea why you are categorizing them.

Why in the world would you say you can take your sweet time on them. They are all vulnerable.

So, 4 is the only thing you really need to say. Of course how you are going to update your fridge or your toaster is a bit obscure. Do you really want a "owned" wifi device anywhere on your internal network?

He who is William Unruh said on Sun, 29 Oct 2017 14:47:35 -0000 (UTC):

Interesting point. Thank you for that observation.

I think what you're saying is that if they can get to *any* of your devices, over the Internet, then, *from those devices*, they can intercept your traffic to, for example, your Linux laptop or Android smart phone.

But I'm confused about the risk in that case.

Are they only intercepting from the refrigerator-to-the-client-device? Or are they then able to get from your router-to-your-client-device?

(The latter would be more dangerous.)

Here is a writeup I made for my family that others can use which shows just

*one* example of fixing a WiFI device. In this case, it's a Ubiquti radio set up as an access point and only going about a half kilometer, but it could be set up to go for miles (as some of my other radios are set up).

(0) Log into your radio

(1) Check the firmware version (noting the board revision, e.g., XW)

(2) Hit the "Check Now" button to see if you can update from here

(3) If not, go to the manufacturer's web site to locate the firmware file

(4) You may have to agree to the manufacturer's updated EULA

(5) Download the file to a known location on your computer

(6) Save the file in a logical location on your computer for future use

(7) Then in the radio, press the "Upload Firmware Choose File" button

(8) Wait for the firmware to upload (it may take a minute or two)

(9) Once uploaded, press the "Update" button to update the firmware

(10) Wait for the firmware to be updated (it may take a minute or two)

(11) Do not power down while you are waiting for the firmware to update

(12) When done, the radio will reboot; log back in to check results

(13) You should note that the firmware should now be updated

(14) Doublecheck now that everything is updated that it is working fine

I have over a dozen WiFi devices in my house.... so I'm updating them one by one. I'm more worried about my grandchildren not knowing how to update

*their* devices, and my older siblings, etc.

But I agree, it's a PITA to update *every* WiFi device in the house. I have over a half-dozen access point radios, for example, and a few on the roof, etc., some of which connect by WiFi to homes that are 10 miles away, so it's a pain for any of them.