Codedstyle

Windows Integrity Levels

Sun, 06 Dec 2009 17:25:00 +0000

Explains the basis behind UAC (among other things). I’ve heard of ‘integrity levels’ but this is the first meat I’ve seen about them. Plus links to chml and regil, utilities to view and manage integrity levels. Handy. Related posts:How does BizTalk Server 2006R2 or BTS2009 support the HIPAA validations known as WEDI-SNIP Levels 1 through 7?Transparency Models: A Tale of Two LevelsUser Account Control and WMI Related posts:

DeepInTrouble, the new KID on the block

Sun, 06 Dec 2009 17:21:00 +0000

Background:
There are many performance monitoring/analyzing tools for different bottleneck conditions of resources. From my experience, I can’t use these tools always to analyze, because I need a prevailing condition where I can use these tools. But, readymade condition is not available always. So I thought of investing some time to create one WPF application, DeepInTrouble (DIT) which will simulate stress on different local resources. You can easily create one such condition and turn on your favorite tool; investigate the patterns arising from it. Currently DIT addresses four different areas. Processor, Memory, Disk and Thread Contention .
Processor:
This will keep the processor occupied at least for the duration mentioned below. The start button will be disabled until the request is completed. This is to prevent users from submitting repeated request of the same kind to the system.
Input fields
Duration:
Number of seconds/thread to keep the processor busy
No of workers:
Number of threads to be competing for CPU scheduling each to keep processor/core keep 100% busy for the “Duration”.
One Core:
If selected, all the threads will run only on the first core. Rest of the cores will be idle. By default it will occupy all the cores.
Memory:
This will simulate a memory leak into the DIT process space. It can be slow or fast, depending on the value of Alloc Interval.
Input fields
Size:
Value in Kbytes that will determine the size of each allocation unit
Allocations:
Number of allocations to be made. Alloc intervals: It controls the speed of allocation.
UnManaged:
If selected, memory will be occupied only from the unmanaged/native area. By default it will allocate from the managed space.
DISKIO:
This will generate a lot of disk IO. Right now it will generate each file with 97MB of data. Depending on the number of files and thread count the traffic can be controlled
Input fields
Number

Reading Notes: 7-Dec-09

Sun, 06 Dec 2009 15:55:00 +0000

Comments
are in Italics and marked as red.

Multiplication
and registers
Why do we need two registers to store the result
of a multiplication? This is because each register or integer memory cell on
x86 PC can contain a number between -2147483648 and 2147483647.

We can think of EDX:EAX pair as two memory cells
joined together to hold the large multiplication result. (P.32)

At
first, I wrote a simple sample for examination.

int a, b;
__int64 c;

int main(int argc, char*
argv[])
{
           a = 5555555;
           b = 5555555;
           c = a * b;

           return
0;
}

I
was expecting the result of “c” to be 30864191358025 because I use __int64 to
hold the number. Instead, “c” is 556368969.

In
WinDbg, I set the breakpoint at the very beginning and checked the registers
values step by step.

0:000> p
eax=0054c563 ebx=00000000 ecx=00000000 edx=00000000 esi=0182f73a edi=0012ff68
eip=004113a7 esp=0012fe9c ebp=0012ff68
iopl=0         nv up ei pl nz na po nc
cs=001b
ss=0023 ds=0023 es=0023
fs=003b gs=0000             efl=00000202
Test!main+0×37:
004113a7
imul    eax,dword ptr [Test!b
(00417140)]    ds:0023:00417140=0054c563

[a]
has been moved to eax (0×0054c563 = 5555555) and is going to multiply [b] in
the next step.

0:000> p
eax=21298449 ebx=00000000
ecx=00000000 edx=00000000
esi=0182f73a edi=0012ff68
eip=004113ae esp=0012fe9c ebp=0012ff68
iopl=0         ov up ei pl nz na po cy
cs=001b
ss=0023 ds=0023 es=0023
fs=003b gs=0000             efl=00000a03
Test!main+0×3e:
004113ae cdq

The
large number should be divided into EDX:EAX to hold the large number but it’s
not. EAX = 0×21298449 = 556368969 which is exactly what I saw in Visual Studio
2005/2008.

After
several hours I found that I made an extremely stupid mistake. I didn’t cast
the int to __int64 immediately after multiplication. That’s why even I used __int64
to hold the number, it was still an integer. It ignored the High part of the
result and only kept the low part in eax.

So
here comes the second version.

int
a, b;
__int64
c;

int
main(int argc, char*
argv[])
{
           a
= 5555555;
           b
= 5555555;
           c
= (__int64)a * b;

           return 0;
}

After
multiplication, you can see the result is separated into 2 parts in WinDbg.

0:000> p
eax=21298449 ebx=7ffde000
ecx=0054c563
edx=00001c12 esi=00000000 edi=0012ff68
eip=004113eb esp=0012fe9c ebp=0012ff68
iopl=0         ov up ei pl zr na pe cy
cs=001b
ss=0023 ds=0023 es=0023
fs=003b gs=0000             efl=00000a47
Mul!main+0×4b:
004113eb mov     dword ptr [Mul!c (00417148)],eax

Anyone Interested In Buying A Geo-Aware Social IM Service?

Sun, 06 Dec 2009 15:06:08 +0000

RadiusIM, the New York-based startup behind the eponymous location-aware instant messaging service, is actively trying to find a buyer, we’ve confirmed with multiple sources. We first caught wind of the company back in August 2006, and dubbed it ‘another proximity-based IM service’.
The company is not too shy about e-mailing all the usual suspects to see if they’d be interested in acquiring them outright, so either they wanted this to become public knowledge and we’re helping them, or they’re so desperate to find someone to take the service off their hands that they really had no other option left but to contact anyone who could be potentially interested by e-mail. Or both, of course.
RadiusIM bills itself as a social IM service. It’s a web-based application that enables people to communicate with their contacts on instant messaging networks such as Yahoo Messenger, Windows Live Messenger, AIM, Google Talk and Facebook Chat.
The startup saw early on that there would be demand for location-awareness features in communication services and made it possible for users to see where their friends were hanging out and meet new people based on their own geographical location.
Alas, apparently there wasn’t enough traction or enough interested advertisers to make the venture viable, so the company is resorting to e-mailing potential buyers to take over the service and its existing user base (they claim to have over 1 million registered users who have, according to their website, sent almost 2 billion messages to each other since its inception).
We’ve contacted the company for confirmation and/or more information, and will update when we hear back. But radiusIM is now officially on deadpool watch.
Crunch Network: CrunchGear drool over the sexiest new gadgets and hardware.

Troubleshooting Data Diagnostic Adapters

Sun, 06 Dec 2009 14:05:29 +0000

Note: Cross posted from Rubel’s Blog.
Permalink
As I had mentioned in my earlier posts that you can enable a number of Diagnostic Data Adapters in test setting so as to collect various types of data while running tests in Test Runner. Sometimes it is confusing if you don’t get any log from a diagnostic data adapter though you’ve enabled it in test setting. There can be many reasons for this behavior like data collector is not expected to generate log in that particular scenario, data collector pre-reqs are not met successfully, that particular data collector is not expected to generate log at all e.g Network Emulation. This post intends to explain various reasons for log generation failure from each data diagnostic adapter. This is NOT an advanced guide for troubleshooting all data diagnostic adapters’ warnings / errors but you can quickly refer to it if you are not seeing logs from a data diagnostic adapter.
IntelliTrace and Test Impact
Both IntelliTrace and Trace Impact data collectors have a number of pre-reqs:
1) They work only for applications written with managed code
2) AuT process need to be started after launching the Test Runner as they needs to instrument modules, binaries or executables that are loaded into a process.
3) ASP .NET Client Proxy for IntelliTrace and Test Impact needs to be enabled in test setting for client role while collecting test data from ASP. NET applications hosted on remote boxes.
4) Computer’s IP address / hostname need to be used instead of “127.0.0.1” “local host” while collecting data from locally hosted ASP.NET apps e.g. please use http://microsof-7f734b/webapplication1/default.aspx instead of http://localhost/webapplication1/default.aspx
5) AuT Process / required modules are included for data collection in configuration settings of collectors. Please check articles mentioned below for details on configuring IntelliTrace and Test Impact:
a. Configuring IntelliTrace
b. Configuring Test Impact
So if

NSFW: Sleepless in London. It’s scary outside the bubble

Sun, 06 Dec 2009 13:14:16 +0000

I’m tired. Very tired. It’s a little after 4am San Francisco time – noon GMT – and I’m sitting in the arrivals lounge Heathrow airport, thanking the lord for Boingo hotspots and trying to commit these few hundred words to cyberspace before the daylight finally penetrates my brain and my whole body goes into jet-lag meltdown.
And to think I was so organised 24 hours ago. My column was written – 1000 words on a big subject of the week; a big subject that I now can’t talk about, for reasons I also can’t talk about. Don’t ask.
Still, I’m a professional and there’s no use crying over spilt milk – I’ve spent five pounds on a coffee, opened a fresh Google Document and am all set to write am alternative column on how happy I am to be back in London, and how excited I am for the opportunity to catch up with all the amazing and inspiring start-ups my erstwhile home has to offer.
But therein lies the problem. While I’m certainly happy to be here – it’s my 30th birthday tomorrow, and there is a party planned – the truth is, I’m just not all that excited about London’s current crop of dot com hopefuls.
When I moved to San Francisco at the start of the year, I promised myself I’d head back to the old country twice a year – mainly to keep my cynicism topped up and to make sure I didn’t lose the accent that your American women find so endearing. But also for a third, more serious reason: I don’t want to forget my roots. The London technology scene is where I cut my columnising teeth, and it’s Brit entrepreneurs that first inspired me to try – and fail – my hand at building a start-up. Whereas

Video Professor To Leverage “Strong Brand Equity” To Raise $10 Million

Sun, 06 Dec 2009 13:11:33 +0000

Video Professor is trying to raise $10 million in convertible debt, according to an investor pitch document that was forwarded to us. The company “incurred substantial losses which depleted its cash reserves” in 2008, says the document, and is looking to use the new money to “retire its line of credit, purchase media, build strategic alliances, finalize the technical development and launch of the e-commerce platform and accelerate growth.”
The company’s product offering was one of the scams we called on in our ScamVille posts. The company lures in potential customers by offering free learning CDs. But they are then billed up to $290 for products they never intended to buy. We outlined how the scam works here, and also point to a number of other sites with thousands of consumer complaints.
According to a revenue chart, revenues for the company peaked in 2006 at around $135 million, but dropped to under $80 million in 2007 and were just $40 million in 2008. Projected 2009 revenues are nearly $100 million.

“VPI’s competitive advantage is the superior quality of its learning programs, its use of subject matter experts, and the trusted relationship that the brand has earned with consumers,” says the document. It fails to point out the hugely negative reviews and complaints in this article as well as Amazon and epinions.
The document also boasts:
STRONG BRAND EQUITY
Having maintained the longest running direct response marketing campaign in the history of television, the VIDEO PROFESSOR™ brand is highly recognized in both consumer and commercial markets with millions of satisfied customers. VPI has supplied computer learning to over 20 million unique customers, many of whom have
purchased multiple learning titles.

And to all those 20 million “customers” who’ve already been scammed one time by Video Professor: look out, they’re coming back for

Viva el debate, vivan los matices

Sun, 06 Dec 2009 11:50:29 +0000

El tema alrededor del manifiesto se va desarrollando, hay mucha más gente que ha escrito a raíz de los sucesos acaecidos la semana pasada, y parece que va habiendo gente que de manera razonable aporta puntos de vista distintos.
De cualquier manera, a mi me va pareciendo que el tema de fondo que lo va empapando todo es el debate sobre la propiedad intelectual y sobre las normas de juego alrededor de los derechos de uso y de copia.
Voy enlazando aquí algunos, que me parecen en general informados, suficientemente desapasionados y bastante razonables. Cada una de su padre y de su madre, eso sí:
Empezamos con un poco de humor gráfico: Eduangi.com
Vida de un consultor: Manifiestos, churras y merinas
Barcepundit
Materias grises: paranoia, piratería y libertades
Geografía subjetiva: Sobre el manifiesto
Industria Audiovisual: Revueltas en el aire digital, una sociedad perpleja a la que se le ha hurtado el verdadero debate
Medios en la red: Razones por las que este blog no se ha adherido al manifiesto
Eudald Domenech: ¿Pirata yo?
Más:
Jesteban.net: Porqué NO apoyo “El Manifiesto por los derechos de Internet”

Транзакционная память – вторая часть

Sun, 06 Dec 2009 10:56:00 +0000

В прошлом посте мы рассмотрели концепцию транзакционной памяти – средство высокоуровневой организации работы параллельно выполняющихся задач. Рассмотрев основы операционной семантики, во многом совпадающие с семантикой транзакций, мы не упомянули несколько интересных расширений, предложенных в работе Composable Memory Transactions и воплощенных в языке Concurrent Haskell. О них и пойдет речь сегодня.
Оператор retry
Цель введения транзакционной памяти – предоставить высокоуровневую замену блокировкам, но согласование обращений к разделяемой памяти отнюдь не исчерпывает все использования блокировок. Еще один важный аспект – объекты синхронизации, которые используются для оповещения о событиях и блокирующего ожидания таких событий. Чтобы быть последовательной, методология транзакционной памяти должна предоставлять аналогичную высокоуровневую концепцию, которую можно использовать для ожидания событий.
Одним из первых предложений в этом направлении было введение условных критических областей (conditional critical regions, CCR) – аннотация транзакций некоторым условием, без правдивости которого транзакция не начинает выполнение. Этот подход работает, но не очень изящен, ибо требует создания новой транзакции даже тогда, когда транзакция уже запущена – то есть, только ради того, чтобы выразить семантику условного ожидания.
Современный взгляд на решение этой проблемы заключается в введении оператора retry, который сигнализирует о необходимости перезапуска транзакции – комбинация этого оператора с условным оператором предоставляет конструкт ожидания некоторого события и во многих случаях делает явную сигнализацию события избыточной. Ниже приведено сравнение реализаций блокирующей коллекции ограниченной емкости, аналогичной коллекции BlockingCollection из .NET Framework 4.0 (реализация, основанная на примитивах синхронизации намеренно упрощена в целях краткости).

Блокирующая коллекция
(реализована при помощи примитивов синхронизации)

Блокирующая коллекция
(реализована при помощи оператора retry)

public class BlockingCollection
{
private Object _lock = new Object();

public void Put(T item)
{
while (Monitor.Wait(_lock, Timeout.Inifinite))
{
try

Unable to edit files uploaded to document library where url length is more than 260 characters

Sun, 06 Dec 2009 10:22:00 +0000

As we all know there is a limitation in URL length as 260 characters for many reason and there are many articles from different specialties talks about it.

Recently I have come across a scenario where users are able to upload documents to the document library and sub folders, if we calculate the total URL length of the uploaded document, it is more than 260 characters, SharePoint didn’t throw any error or warning message while uploading documents to the folders.
When users are trying to edit the same document by clicking on document link or edit using Microsoft Office Word in dropdown menu, we get below warning messages.

              "The URL for this file is too long for the application. A temporary copy of this file will be opened on
               your computer. You must save this copy as a new file"

The document will be opened in Read Only mode and when we try to save the document back to same location we get
             "This file is read-only."
             "The document could not be opened for editing. A Windows SharePoint Services compatible
              application could not be found to edit the document."

Q1. Does SharePoint has 260 characters limit?
         Answer: Yes, it has the limit.

Q2. If SharePoint has 260 characters limit then does it throw any error/warning while uploading documents?
         Answer: Yes, it does throw warning message and terminate the upload process.

Q3. As mentioned SharePoint has 260 limit and it does not allow users to upload, then in what situation we get into the above issue (able to upload but not able to edit).

         Answer: As I have mentioned SharePoint has 260 characters restriction for both operations upload and edit. Sometimes users are experiencing issue only edit not while uploading.
Here is the explanation how restriction applied while uploading and editing.
For example my