Tales from the Crypto - All Comments

Should We Worry About the New SSL Attack? - Internet Security, Online Safety, Web Filtering and all the other net nastys no-one talks about - The IT Juggler

Mon, 16 Nov 2009 01:04:50 GMT

Pingback from Should We Worry About the New SSL Attack? - Internet Security, Online Safety, Web Filtering and all the other net nastys no-one talks about - The IT Juggler

Should We Worry About the New SSL Attack? - Internet Security, Online Safety, Web Filtering and all the other net nastys no-one talks about - The IT Juggler

Mon, 16 Nov 2009 01:04:46 GMT

Pingback from Should We Worry About the New SSL Attack? - Internet Security, Online Safety, Web Filtering and all the other net nastys no-one talks about - The IT Juggler

re: DELAY or NODELAY - Riffing on Larry, who's riffing on Raymond...

Ben — Sat, 14 Nov 2009 07:06:51 GMT

How about when sending chunks of data larger than the maximum segment size. If the number of segments the data needs to be broken into is even, Nagle causes decreased throughput. This is a common case on embedded devices, where the MSS may be 256 bytes or less.

re: Sending chunks of data

Yes, that's why you don't do that. You are apparently engaging in one of two things:

Sending a bulk data stream and splitting it badly.
Expecting an acknowledgement from a protocol that isn't sending an acknowledgement.

In the first case, if you're sending a stream of data larger than a segment, only the last segment should be smaller-sized, and there should be an application acknowledgement from the receiver thereafter.

In the latter case, your protocol is badly designed and needs rethinking so that if you need to wait for an acknowledgement, the acknowledgement you wait for should come from the application and not the network stack - because you, the application, don't actually care if the network stack received the data.

So, no, this doesn't invalidate my argument.

Saturday, November 14, 2009 06:27 PM by Alun Jones

re: Removing Apple Mobile Device Support

Smitty — Thu, 12 Nov 2009 10:54:33 GMT

AMDS is giving me a headache. It won't go away, I want to sync the mYpod, but that darned uninstall always wants to re-install itself. I need a drink.

My take on the SSL MitM Attacks – part 2 – clarifications

Tales from the Crypto — Thu, 12 Nov 2009 05:20:09 GMT

Since the last post I made on the topic of SSL renegotiation attacks , I’ve had a few questions in email

My take on the SSL MITM Attacks ??? part 1 ??? the HTTPS attack | Windows 2008 Security

Tue, 10 Nov 2009 17:35:19 GMT

Pingback from My take on the SSL MITM Attacks ??? part 1 ??? the HTTPS attack | Windows 2008 Security

Why changing passwords should be done regularly | Windows 2008 Security

Why changing passwords should be done regularly | Windows 2008 Security — Tue, 03 Nov 2009 18:03:35 GMT

Pingback from Why changing passwords should be done regularly | Windows 2008 Security

FTP and SSL | IIS & Exchange Digest

FTP and SSL | IIS & Exchange Digest — Wed, 28 Oct 2009 10:33:58 GMT

Pingback from FTP and SSL | IIS & Exchange Digest

FTP and SSL | IIS & Exchange Digest

FTP and SSL | IIS & Exchange Digest — Wed, 28 Oct 2009 10:33:52 GMT

Pingback from FTP and SSL | IIS & Exchange Digest

Video | Enjolt.com | Innovate for Success

Video | Enjolt.com | Innovate for Success — Sat, 24 Oct 2009 17:46:11 GMT

Pingback from Video | Enjolt.com | Innovate for Success

re: DELAY or NODELAY - Riffing on Larry, who's riffing on Raymond...

Brad — Sat, 24 Oct 2009 15:54:34 GMT

ok here is the thing guys

for most everyday situations you guys are right

but!!!!!

for any mmpog players out there know with nagle and tcp ack/delay diasabled games run faster.. that depend on literly half sec loading time of skills you can not wait for more packet,s to accumlate befor sending information and that bandwidth hogging consant sendign and recviening of packets befor checking them is needed to have that split second reaction....

for anything other then split second reaction times on games then no its not something ppl should just use to think it will speed up there net.. for bowsing and some otherthings and can make it slower and allow bad packets to be sent

Mmpog major fix for reaction time with server

other then that leave it alone

re: Still not a good argument

Sorry to tell you this, but the only thing you've demonstrated is that many MMORPGs are written poorly as far as network performance is concerned.

Once again, remember that you have a choice - data is either time-sensitive, in which case it can be discarded, lost, ignored and updated later, or it is sequence-sensitive, in which case the data must get through even if it takes longer.

The data you are talking about is clearly time-sensitive, not sequence-sensitive. As such, it should be communicated using UDP, not TCP. And UDP has no Nagle or delayed-ACK interactions.

You have simply made an argument that MMORPG writers should employ experienced and skilled network developers, rather than the hacks they currently employ.

Saturday, October 24, 2009 09:27 AM by Alun Jones

re: Windows 7 – what it’s missing

bradley — Fri, 23 Oct 2009 06:20:35 GMT

sbs.seandaniel.com/.../how-to-get-live-messenger-back-in.html

Check that out.

SAL-like code annotations for Java | Windows 2008 Security

SAL-like code annotations for Java | Windows 2008 Security — Wed, 21 Oct 2009 16:52:28 GMT

Pingback from SAL-like code annotations for Java | Windows 2008 Security

re: SAL-like code annotations for Java

Fred — Wed, 21 Oct 2009 07:00:32 GMT

So are they coming to C#?

I want as many runtime errors as possible to be compile time errors.

re: Annotations for C#

Peculiarly enough, I received the following email from Michael Howard:

"Cool! The concept is not new – have you seen the contract stuff in .NET 4.0?

So I went looking - here's what I've found: http://msdn.microsoft.com/en-us/library/dd264808(VS.100).aspx

Wednesday, October 29, 2009 06:27 PM by Alun Jones

SAL-like code annotations for Java

Tales from the Crypto — Wed, 21 Oct 2009 03:44:43 GMT

types.cs.washington.edu/jsr308 seems to be talking about a set of type annotations for Java that

SAL-like code annotations for Java

Tales from the Crypto — Wed, 21 Oct 2009 03:44:41 GMT

types.cs.washington.edu/jsr308 seems to be talking about a set of type annotations for Java that

SAL-like code annotations for Java

Tales from the Crypto — Wed, 21 Oct 2009 03:44:38 GMT

types.cs.washington.edu/jsr308 seems to be talking about a set of type annotations for Java that

re: Error 0x80005000 and DirectoryEntry in .NET

Greg — Mon, 19 Oct 2009 20:20:46 GMT

I just spent about an hour trying to figure this out myself. This was what my problem was. Case sensitivity...meh!

re: Removing Apple Mobile Device Support

wendell Clark — Sat, 17 Oct 2009 02:53:48 GMT

Allen HI im a 25 Year Hardware Tech. I can tell you ALL how to Absolutly uninstall AND remove ALL traces og left over Registry keys and any other files associated with the program. Just download a FREE program call "<name deleted>", I think at <name deleted>. and it will get rid of it in about 2 Minutes. Thanks, Clark

re: Registry cleaners: generally bad news

I am really not going to recommend any registry cleaners up here, sorry. There are so many bogus registry cleaners that install malware, I'm not going to take that risk. Even the bona-fide registry cleaners are so often associated with killing systems that I'm going to recommend that instead of using registry cleaners, you should always work with the application support team for whatever application you're trying to clean up.

Wednesday, October 29, 2009 06:27 AM by Alun Jones

re: Wireless PC Lock - part 2

hamish — Tue, 13 Oct 2009 23:57:22 GMT

heya. Brilliant software! Any chance you can distribute an compiled copy that does auto un-lock upon return? (even with the risks)