Merit Can Be Bought, Passion Can't

How to compile libhdfs

2012-02-10T00:37:00.000-08:00

gcc sample.c -I${HADOOP_HOME}/src/c++/libhdfs -L${HADOOP_HOME}/c++/Linux-amd64-64/lib -lhdfs -o sample

at your .bashrc

export LD_LIBRARY_PATH=/usr/lib/jvm/java-6-sun/jre/lib/amd64/server:/usr/lib/jvm/java-6-sun/jre/lib/amd64/server/libjvm.so:/usr/local/hadoop/c++/Linux-amd64-64/lib

export CLASSPATH=/usr/local/hadoop/lib:/usr/local/hadoop

Shared Library

2012-02-09T23:55:00.000-08:00

To run a program linked with the shared version of the library the operating system must be able to locate the corresponding `.so` file at runtime. If the library cannot be found, the following error will occur:

     $ ./a.out      ./a.out: error while loading shared libraries:      libgsl.so.0: cannot open shared object file: No such      file or directory

To avoid this error, either modify the system dynamic linker configuration¹ or define the shell variable LD_LIBRARY_PATH to include the directory where the library is installed.

For example, in the Bourne shell (/bin/sh or /bin/bash), the library search path can be set with the following commands:

     $ LD_LIBRARY_PATH=/usr/local/lib      $ export LD_LIBRARY_PATH      $ ./example

In the C-shell (/bin/csh or /bin/tcsh) the equivalent command is,

     % setenv LD_LIBRARY_PATH /usr/local/lib

The standard prompt for the C-shell in the example above is the percent character ‘%’, and should not be typed as part of the command.

To save retyping these commands each session they can be placed in an individual or system-wide login file.

To compile a statically linked version of the program, use the -static flag in gcc,

     $ gcc -static example.o -lgsl -lgslcblas -lm

Dell system won't boot after ubuntu maverick server installation

2012-02-07T01:37:00.000-08:00

The system thrown you out to initramfs, saying there is something wrong with root delay. What you need to do is to go to recovery mode, and edit grub.cfg (/boot/grub/grub.cfg). ADD rootdelay=90.

### BEGIN /etc/grub.d/10_linux ### menuentry 'Ubuntu, with Linux 2.6.32-21-generic' --class ubuntu --class gnu-linux --class gnu --class os {     recordfail     insmod ext2     set root='(hd0,1)'     search --no-floppy --fs-uuid --set c00fadde-f7e8-45e7-a4da-0235c605f756     linux   /boot/vmlinuz-2.6.32-21-generic root=UUID=c00fadde-f7e8-45e7-a4da-0235c605f756 ro rootdelay=90
    initrd  /boot/initrd.img-2.6.32-21-generic }

Set Up NTP Server and Client

2012-02-05T21:06:00.000-08:00

Setting Up NTP Server

Since servers in the data center is not able to synchronize with any external time server, we have to set up our standalone time server.

192.168.37.183 as Time Server

a. Make sure ntpd server is installed.

sudo apt-get install ntp

b. Since we are not going to synchronize with any external server. Edit /etc/ntp.conf:

# You do need to talk to an NTP server or two (or three).

server 127.127.1.0

fudge 127.127.1.0 stratum 10

c. Permit access for other servers in 192.168.37.0 network to synchronize

restrict 192.168.37.0 mask 255.255.255.0 nomodify notrap

broadcast 192.168.37.255

d. Star the ntp:

sudo service ntp start

At Client

a. Get ntpd server installed.

b. Synchronize to 192.168.37.183

#You do need to talk to an NTP server or two (or three).

server 192.168.37.183 iburst

c. Start the ntp:

sudo service ntp start

d. Command to check the ntp activity:

tail –f /var/log/syslog

ntpq –c lpeer

ntpd –q 192.168.37.183

ParaView in Ubuntu

2012-02-01T01:30:00.000-08:00

Before launching paraview, do this:

tux@remote:~$ export LIBGL_ALWAYS_INDIRECT=y

JAVA PATH

2012-01-31T22:35:00.000-08:00

export JAVA_HOME=/usr/lib/jvm/java-6-sun
export PATH=$PATH:/home/ARM/bin:/usr/lib/jvm/java-6-sun/bin

How to know your ubuntu release

2012-01-31T22:14:00.001-08:00

cat /etc/lsb-release
uname -r

How To Install Sun Java 6 JDK on Ubuntu

2012-01-31T22:11:00.000-08:00

- sudo add-apt-repository "deb http://archive.canonical.com/ lucid partner"
- sudo apt-get update
- sudo apt-get install sun-java6-jdk
- sudo update-java-alternatives -s java-6-sun

How to syntax highlighting to uncommon file type in VIM

2012-01-02T17:17:00.000-08:00

In this case, I want to use html syntax highligthing on index.tt, which is actually a file with html format.

I will just issue:
:

au BufReadPost *.tt set syntax=html

How to create CA, Server and Client Certificates

2011-11-27T17:00:00.000-08:00

1. CA
a. openssl req -out CA.pem -new -x509 -days 365
b. openssl pkcs12 -export -out CA.pfx -inkey privkey.pem -in CA.pem

2. Server
a. openssl genrsa -out server.key 1024
b. openssl req -key server.key -new -out server.req
c. openssl x509 -req -in server.req -CA ../CA/CA.pem -CAkey ../CA/privkey.pem -set_serial 01 -out server.pem

3. Client
a. openssl genrsa -out client.key 1024
b. openssl req -key client.key -new -out client.req
c. openssl x509 -req -in client.req -CA ../CA/CA.pem -CAkey ../CA/privkey.pem -set_serial 02 -out client.pem

PS:
a. pay attention to COMMON NAME during configuration. It should match the server name.
b. to skip hostname verification configured by COMMON NAME.
curl_easy_setopt(curl, CURLOPT_SSL_VERIFYHOST, 0L)

how to compile libcurl and libyaml

2011-09-26T18:08:00.000-07:00

gcc -lcurl -lyaml CDMI_Create_Container.c -o CDMI_Create_Container

How to compile cJSON

2011-09-25T22:38:00.001-07:00

gcc -lm cJSON.c yourprogram.c -o yourprogram

Feeling crappy. Where is my coffee?

2011-04-17T06:45:00.000-07:00

Sunday Morning

2011-04-17T06:18:00.000-07:00

setTimeOut Doesn't Work in IE

2011-03-30T18:55:00.000-07:00

setTimeOut() is working fine in firefox, but strangely it is not working in IE. the solution is you have to includ Math.random as one of the parameter in your ajax request. or otherwise IE will not refresh the page, because it stupidly thinks the content is the same. function getContainerList(){ new Ajax.Request('/mongo_list_container', {method: 'get', parameters:{ sid : Math.random()}, onSuccess: function(transport){ var response = transport.responseText; var containerlist = response.evalJSON(); var NumberOfContainer = containerlist.container.length; var newHTML=""; for (i = 0; i < NumberOfContainer; i++){ var withSlash = containerlist.container[i]; var withoutSlash = withSlash.split("/"); newHTML = newHTML + "" + '" + ""; } $("ContainerList").innerHTML = "

" + newHTML + "

Container List

"; }, onFailure: function(){ alert('Something went wrong...'); } }); setTimeout("getContainerList()",5000); }

Perl Dancer Interesting Functions

2011-03-27T17:54:00.000-07:00

To Return Form Parameters

my %allparams = params;
foreach (keys %allparams){
print "key: ", "$_\n";
print "Hash: ", "$allparams{$_}\n";
}

To Return Client Address
my $address = request->address();
print "$address\n";

Send and Download your uploaded file
my $body = request->body();
print "$body\n";

Print Content, Upload to Folder, Filename
my $allupload = request->upload('filename');
my $content = $allupload->content;
print $content; #print content of your uploaded file
$allupload->copyto('/UPLOAD/test.doc');
my $FILENAME = $allupload->filename; #uploaded file name
printe $FILENAME;

How to Upload File using Web Form in Perl Dancer

2011-03-25T00:55:00.000-07:00

First you have to create a form for the user to upload the file. The form will look like this.

form action="" method="post" enctype="multipart/form-data" action="<% upload_file %>"
input type="hidden" value="0" name="createContainer">
input type="hidden" value="1" name="createFile">
table>
tbody>
File To Upload:
input type="file" name="filename">

input type="submit" value="Create">

/form
-->
Then at your route, code something like this

post '/upload_file' =>sub{
session 'logged_in' => true;
my $allupload = request->upload('filename');
$allupload->copy_to('/UPLOAD');
};

Multi Cloud Storage API

2011-03-15T00:04:00.000-07:00

Why do you need Multi Cloud Storage API?
a. Importing and exporting data from different clouds.
b. Syncing clouds automatically
c. Having Cloud Fail-over
d. Having a single view

Choices:
a. Zcloud API (C)
b. JClouds (java)
c. CloudLoop (java)
d. SimpleCloud (PHP)
e. SMEStorage Multi-Cloud API

DBD::mysql installation

2011-03-13T22:34:00.000-07:00

CPAN complaints cannot find mysql_config file during installation DBD::mysql.
You have to install libmysqlclient-dev package using synaptic.

File upload to http server

2011-03-09T01:38:00.000-08:00

USE LIBCURL!

create container using curl

2011-03-08T22:41:00.001-08:00

curl -X PUT -L -H "content-type:application/vnd.org.snia.cdmi.container+json" -H "accept:application/vnd.org.snia.cdmi.container+json" -H "HttpHeader.X-CDMISpecification-Version: 1.0" --user-agent "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.2.13) Gecko/20101206 Ubuntu/10.10 (maverick) Firefox/3.6.13" -H "Connection:keep-alive" --data '{"metadata":{},"exports":{"OCCI/NFS":{}}}' http://localhost:8080/cdmi-server//jungle

cdmi server access using curl

2011-03-08T18:45:00.000-08:00

curl -L -H "Accept:text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8" --user-agent "Mozilla/5.0 (X11; U; Lnux i686; en-US; rv:1.9.2.13) Gecko/20101206 Ubuntu/10.10 (maverick) Firefox/3.6.13" -H "Connection:keep-alive" -H "Keep-Alive: 115" -H "Accept-Language en-us,en;q=0.5" -H "Accept-Encoding: gzip,deflate" -H "Accept-Charset: ISO-8859-1,utf-8;q=0.7,*,q=0.7" http://localhost:8080/cdmi-server

What is Cloud Storage?

2011-03-02T18:12:00.000-08:00

Cloud storage is, of course, a storage to store your files. Your files may be documents, pictures, photos, spreadsheets or mp3s. Some people call them, objects or unstructured data. In a LAN environment, you can mount an NFS share or SAMBA share to your system. Latency is not a problem. But over the internet, this methods will not work.

When you are using a cloud storage provided by let say Amazon S3, you are using its API (Amazon S3 API, s3curl, etc) to store and retrieve your files. You are not doing the NFS/SAMBA thing over internet. The API is usually based on REST web service. REST does not require sessions between your system and the cloud. Everything you need to complete the request is there in the request itself.

Using Curl to DELETE (delete entry)

2011-03-01T22:42:00.000-08:00

curl --request DELETE -H "Content-Type:text/html" http://localhost:3000/rest/person/6

Using Curl to POST (create new entry)

2011-03-01T22:28:00.001-08:00

curl --request POST --data '{"firstname":"despicable","lastname":"me"}' -H "Content-Type:text/html" http://localhost:3000/rest/person/1

Using Curl to PUT Rest Web Service

2011-03-01T21:51:00.000-08:00

Use the following command:

curl --request PUT --data '{"firstname":"linus","lastname":"sebastian"}' -H "Content-Type:text/html" http://localhost:3000/rest/person/1

OR

curl -X PUT --data '{"firstname":"donald","lastname":"duck"}' -H "Content-Type:text/html" http://localhost:3000/rest/person/1

Using Curl to Access Rest Web Service

2011-03-01T18:52:00.000-08:00

I have a rest web service the following content type mapping.

__PACKAGE__->config('default' => 'JSON',
'map' => {
'text/html' => 'JSON',
}
);

I can access the web service in shell using the following command:

curl -H "Content-Type:text/html" http://localhost:3000/rest/person/1

How to comment out multiple lines in VIM

2011-02-28T05:41:00.000-08:00

Navigate with the cursor to either the first or last line you want to comment out.
Pess Control and v to select a new visual block.
Move your cursor down (or up respectively) to the last (or first) line you want to comment out. You just formed a visual block of your comment.
Press I to insert text before all selected lines.
Type your comment code (for example // or # or ;).
Press Escape. Your comment code will be inserted before every line of the visual block.

to uncomment..
In Vim, you can replace a selected portion of text across multiple lines with a new piece of text using a Visual-block change. As an example, highlight a portion of text across multiple lines using ctrl-v (blockwise visual mode), now press the "c" key to delete the selected text and enter insert mode. Type in some new text and press the Esc key. The text should be inserted for each row of the selection.

Perl Catalyst Step By Step

2011-02-27T05:37:00.000-08:00

1. Create Application
$ catalyst.pl AddressBook
2. Create and Populate Database
3. Link the database to a Catalyst Model.
$ perl script/addressbook_create.pl
model AddressDB
DBIC::Schema
AddressBook::Schema::AddressDB
create=static
dbi:SQLite=/temp/addressbook
4. lib/AddressBook/Model/AddressDB.pm -->The actual Model that Catalyst uses.
lib/AddressBook/Schema/AddressDB.pm --> This is the schema that the Model
points to.
lib/AddressBook/Schema/AddressDB/Result/People.pm
lib/AddressBook/Schema/AddressDB/Result/Address.pm --> This is where the
real schema data is stored.
5. Create relationship between tables in People.pm and Address.pm
6. Configure TTSite
a. Templates are stored in /root/src
b. Configurations are stored in /root/lib
7. Creating Index Page (root/src/index.tt2) and Not Found page (root/src/not_found.tt2)
8. Create 'Person' Controller
$perl script/addressbook_create.pl controller Person
9. Add 'Person' Controller 'List' Function
sub list :Local{
my ($self, $c) = @_
my $people = $c->model('AddressDB::People');
$c->stash->{people} = $people;
}
10. The 'Person' Controller 'List' Function will be called using http://localhost:3000/person/list
11. Now we have to create the view for it, named it list.tt2, put it in root/src/person/list.tt2
12. Add 'Person' Controller 'Delete' Function
sub delete :Local{
my ($self, $c, $id) = @_;
my $person = $c->model('AddressDB::People')->find({id => $id});
$c->stash->{person} = $person;

if ($person){
$c->stash->{message} = 'Deleted '. $person->name;
$person->delete;
}
else {
$c->response->status(404);
$c->stash->{error} = "No person $id";
}
$c->response->redirect($c->uri_for_action('/person/list'));
}
13. The 'Person' Controller 'Delete' Function will be called using
http://localhost:3000/delete/1, where 1 is the person's ID number in the
database.
14. Add Form to add person into the database. We are going to use FORMBUILDER.
a. Form Definition in /root/forms/person/edit.fb
b. Template where we can render this form in /root/src/person/edit.tt2
15. Add 'Person' Controller 'Edit' Function
sub edit :Local Form {
my ($self, $c, $id) = @_;
my $form = $self->formbuilder;
my $person = $c->model('AddressDB::People')->find_or_new({id=>$id});
if ($form->submitted && $form->validate){
#form was submitted and validated
$person->firstname($form->field('firstname'));
$person->lastname($form->field('lastname'));
$person->update_or_insert;
$c->stash->{message} = ($id > 0 ? 'Updated' : 'Added'). $person->name;
$c->forward('list');
}
else{
if(!$id){
$c->stash->{message} = "Adding a new person";
}
$form->field(name => 'firstname', value=> $person->firstname);
$form->field(name => 'lastname', value=> $person->lastname);
}
}
16. Add 'Person' Controller 'Add' Function
sub add :Local{
my ($self, $c) = @_;
$c->response->redirect($c->uri_for_action('/person/edit'));
}
17.Creating template for address editing and put it in root/src/address/edit.tt2
18.Creating form definition, put it in root/forms/address/edit.fb
19.Creating Address Controller
perl script/addressbook_create.pl controller Address

Cloud Storage

2011-01-23T16:48:00.000-08:00

No. It is not related to weather or cloud system. Cloud storage is a storage for your files, data or documents. But it is not your local hard drive, thumbdrive or portable hardrive. Cloud storage is is an off-site storage system maintained by a third party. So, instead of storing information to y our computer's hard drive or other local storage device, you save it remotely, on the off-site storage system through the connection provided by the Internet.

Now you don't have to carry around your physical storage with you all the times. As long as you have the Internet access, any where in this planet, you will be able to access your files or documents. You can even provide certain level of access, to allow your friends or colleagues to access your data.

A prototypical cloud storage system is made up of a number of x86 servers, each with its own storage, most commonly using four to 16 SATA drives. Users and their applications access the system through standard file access protocols like CIFS and NFS or via object storage and retrieval protocols like SOAP and REST.

The storage nodes in a private cloud are linked together with a layer of smart software, which performs several functions. First, it maintains a global name space that allows all the storage in the cluster to be accessed as a single entity, so that administrators can add storage capacity on the back end without having to tell applications at the front end how to reach it. The software also handles drive failures and keeps data available to applications and end users.

A private cloud storage infrastructure should also be able to scale from hundreds of terabytes to multiple petabytes. That level of scalability is achieved not with a forklift upgrade, but simply by adding more servers as they're needed.

This architecture provides two major benefits. First, storage administrators can configure and provision new storage nodes quickly and inexpensively. Second, administrators can add capacity only as demand requires, instead of purchasing additional disk space to meet anticipated future growth and then having that capacity sit idle in the present.

However, there are also trade-offs. Cloud storage is best suited to unstructured data, such as medical images, engineering drawings, and Office documents. For another, because each x86 server isn't as reliable as a high-end enterprise disk array, a private cloud must store copies of the data on multiple nodes. This requires more raw disk space than an enterprise disk array using a RAID-5 or 6 system. For example, if you set a policy for your private cloud to keep three copies of a 60-GB file for data protection, it would require 180 GB of disk, whereas a 6+2 RAID-6 system would need just 80 GB.

How to Install MongoDB on Ubuntu 8.04

2011-01-19T19:14:00.000-08:00

Edit your sources.list...

sudo nano /etc/apt/sources.list

Add this to the bottom...

deb http://downloads.mongodb.org/distros/ubuntu 9.4 10gen

Then run these commands...

sudo apt-key adv --keyserver keyserver.ubuntu.com --recv 7F0CEB10 sudo apt-get update

Finally install MongoDB...

sudo apt-get install mongodb-stable

Cloud Storage Implementation

2011-01-18T01:39:00.000-08:00

Structured and Unstructured Data, can they be merged?

2011-01-17T21:15:00.000-08:00

So, what is structured data? What is unstructured data?

Structured data is the data that resides in your database. They are "structured" with fixed formats. The data may come in table format. Data that you store in your database server is structured data.

Unstructured data, on the other hand, is "unstructured", of course. It may be in the form of spreadsheet, documents, photos, scanned documents, videos or audio. Unstructured data is "impossible" to be stored and managed in your database server. But, is it right?

Challenges for Storing Unstructured Data

The first challenge to be considered, and perhaps the most obvious, is the storage of large volumes of unstructured data. It is desirable to have a system that gives you the flexibility to meet the specific requirements of storing unstructured data, with minimal cost and management overhead. Key issues to be considered in the planning of the storage of unstructured data include:

The cost of storing unstructured data, including not only the hardware costs, but also management overhead costs.
Physical storage locations for unstructured data, such as file servers and network attached storage (NAS) devices.
Managing retention and archival policies.
Integrating unstructured data files with associated relational data, and maintaining transaction integrity between structured and unstructured data sources.
Minimizing the management overhead associated with maintaining both relational and unstructured data.
Performance and scability.
Securing unstructured data, and ensuring consistent security with associated relational data.
Availability and recoverability of unstructured data.

Challenges for Using Unstructured Data

The development challenge in building applications that use structured and unstructured data, including writing code to create, retrieve, update and delete (CRUD) unstructured data, and maintaining transactional consistency between associated relational and unstructured data sources.
Indexing and searching unstructured data.
Extracting explicitly available metadata (for example from form fields or file attributes) and exposing it to users.
Converting document content into format that can be searched and queried (for example, converting audio files into text that can be searched by a database query or full-text engine).

Storing Unstructured Data in SQL Server 2008

Applications that rely on relational and non-relational data typically use one of three architectures:

Relational data in the database and non-relational Binary Large Object (BLOP) data in file systems and file servers.
Relational data in database and non-relational data in a dedicated BLOP store.
Relational data and non -relational data in the database.

Each approach has its own benefits and drawbacks. For example, storing unstructured data in file servers or dedicated BLOB stores can reduce costs in terms of dollars per gigabyte, but typically adds complexity for managing and building applications, because the application has to maintain integrity between records in the database and the separate system that contains the BLOBs to which those records are associated. On the other hand, storing BLOBs in the database conveniently centralizes data storage, but typically comes at higher cost and can result in reduced performance.

SQL Server 2008 introduces two new capabilities for storing BLOB data:

FILESTREAM: An attribute you can set on a varbinary column so that the data is stored on the file system ( and therefore benefits from its fast streaming capabilities and storage capabilities) but is managed and accessed directly within the context of the database.
Remote BLOB Storage: A client-side application programming interface (API) that reduces the complexity of building applications that rely on an external store for BLOBs and a database for relational data.

SQL Server BLOBs

SQL Server 2005 introduced the varbinary (max) data type, which enables you to store large binary data values of up to 2,147,483,647 bytes in a SQL Server column or variable. When using the max modifier, you can control how the data is physically stored in the data pages of the table by setting the large value types out of row table option. When this option is set to ON, all values are stored on separate linked pages and a 16-byte root pointer to these ages is stored on the data page for the row. When this option is set to OFF, values of up to 8,000 bytes are stored inline in the data page for the row, and larger values are stored on separate linked pages.
Although the newer FILESTREAM and Remote BLOB capabilities are designed to provide better performance and manageability than standard varbinary BLOB columns, there may be some scenarios in which a varbinary column is appropriate (typically when BLOB sizes average less than 250 KB in size.)

The FILESTREAM Attribute

In SQL Server 2008, you can apply the FILESTREAM attribute to a varbinary column, and SQL Server then stores the data for that column on the local NTFS file system. Storing the data on the file system brings two key benefits:
· Performance matches the streaming performance of the file system.
· BLOB size is limited only by the file system volume size.

However, the column can be managed just like any other BLOB column in SQL Server, so administrators can use the manageability and security capabilities of SQL Server to integrate BLOB data management with the rest of the data in the relational database—without needing to manage the file system data separately. Defining the data as a FILESTREAM column in SQL Server also ensures data-level consistency between the relational data in the database and the unstructured data that is physically stored on the file system. A FILESTREAM column behaves exactly the same as a BLOB column, which means full integration of maintenance operations such as backup and restore, complete integration with the SQL Server security model, and full-transaction support.
Application developers can work with FILESTREAM data through one of two programming models; they can use Transact-SQL to access and manipulate the data just like standard BLOB columns, or they can use the Win32 streaming APIs with Transact-SQL transactional semantics to ensure consistency, which means that they can use standard Win32 read/write calls to FILESTREAM BLOBs as they would if interacting with files on the file system.
In SQL Server 2008, FILESTREAM columns can only store data on local disk volumes, and some features such as transparent encryption and table-valued parameters are not supported for FILESTREAM columns. Additionally, you cannot use tables that contain FILESTREAM columns in database snapshots or database mirroring sessions, although log shipping is supported.

The Remote BLOB Store API

Although the FILESTREAM attribute combines the performance and scalability of the file system with the manageability and data consistency of storing BLOBs in the database, there are often scenarios where it is more efficient or cost-effective to store BLOBs in a system like EMC Centera, Fujitsu Nearline, or any other dedicated BLOB store. The Remote BLOB Store API in SQL Server 2008 makes it easier to integrate dedicated, remote BLOB storage solutions with the relational data in your database by offering a provider-based architecture that enables your applications to use any BLOB store without requiring BLOB store-specific functionality or code.
There are two key components to the architecture:

· The client library: A component that enables the use of any provider to work with your BLOB and relational data together. Client applications use the client library and any applicable provider libraries to insert, update, and query data in the BLOB store right alongside data stored in the database.
· The provider library: A component typically provided by BLOB store vendors (although it could be custom-developed). A provider library implements a common interface and exposes a set of standard services that abstract the details of how to perform Create, Fetch, Enumerate, Delete and Garbage Collect operations for that particular store.

The Remote BLOB Store API is ideal when your database BLOBs need to be stored on a different server from the database and interoperability with other systems is required. There are no restrictions on the BLOB store. Any system that offers a Remote BLOB Store provider, or for which one can be built, is supported. Because the details of interacting with the BLOB store are abstracted by the provider library, you can change the store without requiring changes to applications, which offers substantial benefits and reduced complexity to developers and administrators. Microsoft is working with all storage vendors in the industry to ensure that a proper library of providers is available to use with the Remote BLOB Store API.

Using the Remote BLOB Store API gives you the flexibility to use different systems to manage relational and non-relational data. This is particularly useful for an application that uses multiple stores via the same API. For instance, applications can effectively exploit different store attributes to enable tiered storage or an HSM-like hierarchy. The Remote BLOB Store API maintains integrity between rows in the database and BLOBs in the external store through link-level consistency. For example, if you delete the BLOB reference from the database through a delete statement, the system removes the BLOB from the store. However, the use of the Remote BLOB Store API does not provide you with the full data-level consistency that comes with storing BLOBs directly in the database, as you would by using FILESTREAM or varbinary. [Note: For write-once BLOBs, the consistency is very close to that provided by FILESTREAM].

Searching Unstructured Data in SQL Server 2008

One of the most common requirements for applications that use non-relational data is to enable full-text search over that data.
Prior versions of SQL Server support full-text search in prior versions. However, it has always relied on an external service, the Windows Search Service, for its full-text search engine. Using an external service created of the following two challenges:

· The reliance on an external service created greater deployment complexity and support difficulty.
· Queries that combined full-text search predicates with standard SQL predicates suffered in performance because the entire result set from the full-text portion of the search had to be spooled into the engine before the standard SQL predicates could be applied. The query optimizer could not create plans that incorporated the full-text search predicates.

In SQL Server 2008, full-text search is fully integrated into the database engine. By integrating the full-text engine directly into the SQL core, SQL Server 2008 resolves the challenges that were presented by previous releases, reduces the cost of managing and deploying SQL Server, while improves the performance of searching unstructured data.
Conclusion
SQL Server 2008 provides a flexible solution for storing and searching unstructured data. FILESTREAM enhances the performance of storing non-relational data like BLOBs in the database alongside relational data, without sacrificing manageability. If you need to use a dedicated BLOB store, then the Remote BLOB Store API helps you to integrate your BLOB data with the relational data in your database and provides a consistent programmatic interface for interacting with this data, independent of the system that is used to store the data. Finally, SQL Server 2008 makes searching non-relational data more efficient and easier to manage when you need to store and use large libraries of documents by using any of the solutions discussed in this whitepaper.

For more information:
Microsoft SQL Server 2008http://www.microsoft.com/sqlserver/2008/en/us/default.aspx

Working with Feeds

2010-06-13T05:38:00.000-07:00

Custom Feeds

Ok, you've developed your package, and now you want to use it via make menuconfig, OR you are developing a package and you want to test it in a build before you try to get it included in OpenWRT.

The solution is a custom feed. You can either create an entirely new feed, or use a modified version of one of the standard ones.

Creating the package directory

Adding or package to an existing feed

For this example we assume that your are in /home/user/openwrt as your base directory.

You create your current project dir project.

Then while in /home/user/openwrt/project

svn co svn://svn.openwrt.org/openwrt/trunk openwrt (for the OpenWRT base system)
svn co svn://svn.openwrt.org/openwrt/packages packages (for the packages feed)

Add your package in the appropriate subdirectory under /home/user/openwrt/project/packages

Creating your own feed

Create your project dir and get trunk, as above
Create your package dir and copy your package into it (e.g. cp packagedir /home/user/openwrt/project/customfeed/), so that your package is under (in this example) /home/user/openwrt/project/customfeed/packagedir

Using the feed

Edit your feeds.conf (e.g. /home/user/openwrt/project/openwrt/feeds.conf)
Add a line to access the feed.
1. In the case of adding to the packages feed also comment out the normal packages feed.
e.g.

#srv-svn packages svn://svn.openwrt.org/openwrt/packages
src-link customfeed /home/user/openwrt/project/packages

or in the case of the second example:

src-link customfeed /home/user/openwrt/project/customfeed

Update the feed: from the openwrt dir (e.g. /home/user/openwrt/project/openwrt), do: ./scripts/feeds/update customfeed
And then install it ./scripts/feeds/install -p customfeed
Now your package(s) should be available when you do make menuconfig.

Add a new application to openWRT

2010-06-13T02:47:00.000-07:00

This application note shows how to add a simple package to OpenWrt. With an example, it describes how to integrate a simple application into the OpenWrt build system. The example is a simple application package called simple. The simple application prints 1 to 99.

Create a New Package

Change to the package directory:

$ cd openwrt/package

Create a new directory named simple:

$ mkdir simple
$mkdir simple/src
$cd simple/src

Add source files

Create main.c in simple/src directory.

#include 
int main(void)
{
 int i=0;
 while (i < 100) {
  printf(" Hello world %d \n",i);
  i++;
 }
 return 0;
}

Create Makefile as shown below to compile main.c in simple/src directory. Note that lines in Makefile must be indented with tabs.

# build executable on typing make
all: simple

%.o: %.c
 $(CC) $(CFLAGS) $(EXTRA_CFLAGS) -c -I. -Iinclude -o $@ $^

simple: main.o
 $(CC) -o $@ $^ -L.

clean:
 rm -f *.o simple

Compile on Local PC

--- Optional step ---

The file main.c can be compiled using a compiler on the host PC to see if there are any errors in the code and if the Makefile is working fine.

$ make

The created application can be run and tested on the host PC to see if it is working fine.

./simple
Hello world 0
Hello world 1
...
Hello world 99

Please note that we have not cross-compiled for the target system yet. So, the application is built using GLIBC instead of libc intended for the target. This means that the created application will run only on your host PC but not on the target. Finally run clean to clear the object files and executables created as they are not cross compiled.

$make clean

Integrate to OpenWrt Build System

Create a top level Makefile in package/simple directory. This Makefile will intergrate the package to the openWRT build system along with the menuconfig.

Change to package/simple directory:

$cd ..

Create the following Makefile:

#
# Top level makefile for simple application
#

include $(TOPDIR)/rules.mk

PKG_NAME:=simple
PKG_VERSION:=1.0.0
PKG_RELEASE:=1

include $(INCLUDE_DIR)/package.mk

define Package/simple
 SECTION:=utils
 CATEGORY:=Utilities
 TITLE:=simple -- prints simple 1 to 99
endef

define Build/Prepare
 mkdir -p $(PKG_BUILD_DIR)
 $(CP) ./src/* $(PKG_BUILD_DIR)
endef

define Build/Configure
endef

TARGET_CFLAGS += $(FPIC)

define Package/simple/install
 $(INSTALL_DIR) $(1)/bin
 $(INSTALL_BIN) $(PKG_BUILD_DIR)/simple $(1)/bin/
endef

$(eval $(call BuildPackage,simple))

Some of you might notice that this Makefile has eliminated several items such as downloading sources from Internet, checking md5sum etc because the application sources are present locally.

Compile

To compile the application, first the application needs to be selected for building. This can be accomplished by running menuconfig at the top level directory of openWRT.

$make menuconfig

Select the simple application under the "Utilites" sub item:

OpenWrt Configuration
  Utilities
      <*> simple

Compile the distribution by typing the following command at the top level directory:

$ make V=99

The above V=99 option gives a verbose output. It allows you to get back a more information about the build and see what happened if the compilation fails with an error.

Verify if the application is created in the build directory and if relevant files are included in the romfs image.

Verify

Load the image to the target. The application will be present in the /bin directory on the target.

Common errors

Some of the common problems are as follows

menuconfig does not show the new package

This problem happens if openwrt/packages/tmp is already present. Delete tmp directory and retry make menuconfig.

Applet not found

This problem typically happens when the application was built for the host PC instead of the target. One way to verify is to go to the openwrt/build-dir/simple-1.0.0 directory and do the following:

ubicom32-elf-objdump -x simple

Check the output and verify that the architecture of the created application is ubicom32vX and that the libc is not the GLIBC of the host PC but that of target.

Arduino IDE in UBUNTU HARDY

2009-10-28T00:20:00.000-07:00

I just bought myself an arduino Duemilanove 328. So I download the IDE, install the IDE,
and the compilers of course using apt-get install (gcc-avr, binutils-avr, avr-libc).
And compiled the sample source code (blink led things). And it won't compile.
After googling for a whili, it turned out that my compiler packages from Ubuntu Hardy won't
support Atmega328. Uninstalled all the three packages above.
Reinstall the followings:
gcc-avr_4.3.2-1_i386.deb
binutils-avr_2.18-4_i386.deb
avr-libc_1.6.2.cvs20080610-2_all.deb

How To Add Default Gateway in Linux

2009-05-25T02:38:00.000-07:00

# route add default gw 192.168.1.254 eth0

How to Create RamDisk in Linux

2009-05-10T22:21:00.000-07:00

mkdir /mnt/ramdisk
mount -t tmpfs none /mnt/ramdisk -o size=256000
chmod 1777 /mnt/ramdisk

Linux Multicast Router

2009-04-17T00:58:00.000-07:00

To enable multicasting in linux router you have to enable (value 1) in
/proc/sys/net/ipv4/conf/all/mc_forwarding.

echo 1 > /proc/sys/net/ipv4/conf/all/mc_forwarding

or

sysctl -w net.ipv4.conf.all.mc_forwarding = 1

but chances are you are not able to do it.

The work around is to install pimd. It solves the problem. From there you can try xorp.

about int main (int ac, char* av[]){}

2009-04-09T02:16:00.000-07:00

program1.c

int main (int ac, char* av[])
{
}

after compilation, if you run it by using the following command,

./program1 test1 test 2

ac = 3 ./program1 --> ac =1
test1 --> ac = 2
test2 --> ac = 3

*(av+0) = *av = ./program1
*(av+1) = test1
*(av+2) = test2

Learning to write your own WHO command under linux (UBUNTU hardy)

2009-03-30T00:56:00.000-07:00

OK. What is WHO? Who is a command to display information about the users who are currently logged in.

First we are going to read the manual of WHO

$man who

who [OPTION]... [ FILE | ARG1 ARG2 ]
If FILE is not specified, use /var/run/utmp. /var/log/wtmp as FILE is common.

It says WHO will examine /var/run/utmp file to obtain its information.
To find out more about utmp, type

$man -k utmp fyi, (-k -->option for searching)
utmp (5) - login records

$man 5 utmp
UTMP(5) Linux Programmer's Manual UTMP(5)

NAME
utmp, wtmp - login records

SYNOPSIS
#include

DESCRIPTION
The utmp file allows one to discover information about who is currently using the system. There may be more users currently using the system,
because not all programs use utmp logging.

The file is a sequence of entries with the following structure declared in the include file (note that this is only one of several definitions
around; details depend on the version of libc):

#define UT_UNKNOWN 0
#define RUN_LVL 1
#define BOOT_TIME 2
#define NEW_TIME 3
#define OLD_TIME 4
#define INIT_PROCESS 5
#define LOGIN_PROCESS 6
#define USER_PROCESS 7
#define DEAD_PROCESS 8
#define ACCOUNTING 9

#define UT_LINESIZE 12
#define UT_NAMESIZE 32
#define UT_HOSTSIZE 256

struct exit_status {
short int e_termination; /* process termination status */
short int e_exit; /* process exit status */
};

struct utmp {
short ut_type; /* type of login */
pid_t ut_pid; /* PID of login process */
char ut_line[UT_LINESIZE]; /* device name of tty - "/dev/" */
char ut_id[4]; /* init id or abbrev. ttyname */
char ut_user[UT_NAMESIZE]; /* user name */
char ut_host[UT_HOSTSIZE]; /* hostname for remote login */
struct exit_status ut_exit; /* The exit status of a process
marked as DEAD_PROCESS */

/* The ut_session and ut_tv fields must be the same size when
compiled 32- and 64-bit. This allows data files and shared
memory to be shared between 32- and 64-bit applications */
#if __WORDSIZE == 64 && defined __WORDSIZE_COMPAT32
int32_t ut_session; /* Session ID, used for windowing */
struct {
int32_t tv_sec; /* Seconds */
int32_t tv_usec; /* Microseconds */
} ut_tv; /* Time entry was made */
#else
long int ut_session; /* Session ID, used for windowing */
struct timeval ut_tv; /* Time entry was made */
#endif

int32_t ut_addr_v6[4]; /* IP address of remote host */
char __unused[20]; /* Reserved for future use */
};

/* Backwards compatibility hacks. */
#define ut_name ut_user
#ifndef _NO_UT_TIME
#define ut_time ut_tv.tv_sec
#endif
#define ut_xtime ut_tv.tv_sec
#define ut_addr ut_addr_v6[0]

We are closing in quickly on how who works. The manual said utmp file is a sequence of utmp entries as defined in struct utmp in the utmp.h file. So, where is utmp.h?

$locate utmp.h
/usr/include/utmp.h
/usr/include/bits/utmp.h

After a careful examination on both files, the struct utmp is actually in /usr/include/bits/utmp.h
Up to this point, we have collected the following facts:
1. WHO gets its information from /var/run/utmp
2. The information in /var/run/utmp should be read using a structure called struct utmp which is defined in /usr/include/bits/utmp.h

Our task is to write WHO. We are going to use system calls open(), read() and close().

open()
purpose: creates a connection to a file
include: #include
usage: int fd = open (char* name, int how)
args: name -> name of file
how -> O_RDONLY, O_WRONLY, or O_RDWR
returns: -1 on error
int on success

read()
purpose: transfer up to qty bytes from fd to buf
include: #include
usage: ssize_t numread = read(int fd, void* buf, size_t qty)
args: fd-> source of data
buf->destination for data
qty->number of bytes to transfer
returns: -1 on error
numread on success

close()
purpose: closes a file
include: #include
usage: int result = close(int fd)
args: fd file descriptor
returns: -1 on error
0 on success

who version 1

/* -*- Mode: C; indent-tabs-mode: t; c-basic-offset: 4; tab-width: 4 -*- */
/*
* main.c - Auto-generated by Anjuta's Makefile project wizard
*
*/

#include
#include
#include
#include
#define SHOWHOST
void show_info(struct utmp*);
int main()
{
struct utmp current_record; /*read info into this structure*/
int utmpfd; /*the file handler*/
int reclen;

reclen = sizeof(current_record); /*reclen is 384 bytes*/
utmpfd = open (UTMP_FILE,O_RDONLY); /*UTMP_FILE is define in utmp.h, to be
precise in paths.h*/
while (read(utmpfd, &current_record,reclen)==reclen)
show_info(&current_record);
}

void show_info(struct utmp* utbufp)
{
printf("%-8.8s", (*utbufp).ut_name);
printf(" ");
printf("%-8.8s", (*utbufp).ut_line);
printf(" ");
printf("%10li", (*utbufp).ut_tv.tv_sec);
printf(" ");
#ifdef SHOWHOST
printf("(%s)",(*utbufp).ut_host);
#endif
printf("\n");

}

Output who version 1

reboot ~ 1238374877 (2.6.24-23-generic)
runlevel ~ 1238374877 (2.6.24-23-generic)
LOGIN tty4 1238374877 ()
LOGIN tty3 1238374877 ()
LOGIN tty5 1238374877 ()
LOGIN tty2 1238374877 ()
LOGIN tty6 1238374877 ()
LOGIN tty1 1238374896 ()
juniarto tty7 1238374912 (:0)
juniarto pts/0 1238377168 (:0.0)
juniarto pts/1 1238377727 (:0.0)
juniarto pts/2 1238389211 (:0.0)

The output seems promising. But there are two issues have to solved:
1. Suppress the blank records
reboot, runlevel,LOGIN...
2. Get the log-in times correct

Examining the /usr/include/bits/utmp.h file, we find the following:

/* Values for the `ut_type' field of a `struct utmp'. */
#define EMPTY 0 /* No valid user accounting information. */

#define RUN_LVL 1 /* The system's runlevel. */
#define BOOT_TIME 2 /* Time of system boot. */
#define NEW_TIME 3 /* Time after system clock changed. */
#define OLD_TIME 4 /* Time when system clock changed. */

#define INIT_PROCESS 5 /* Process spawned by the init process. */
#define LOGIN_PROCESS 6 /* Session leader of a logged in user. */
#define USER_PROCESS 7 /* Normal process. */
#define DEAD_PROCESS 8 /* Terminated process. */

#define ACCOUNTING 9

The list looks helpful. Each record has a member called ut_type. Type 7 looks like our lucky number. If we make the following minor change to show_info(), those blank records should vanish:

void show_info(struct utmp* utbufp)
{
if ((*utbufp).ut_type != USER_PROCESS)
return;
printf("%-8.8s", (*utbufp).ut_name);
printf(" ");
printf("%-8.8s", (*utbufp).ut_line);
printf(" ");
printf("%101d", (*utbufp).ut_tv);
printf(" ");
#ifdef SHOWHOST
printf("(%s)",(*utbufp).ut_host);
#endif
printf("\n");
}

OK, now the time (ut_tv)
UNIX stores times as the number of seconds since midnight, Jan 1, 1970, GMT. The ut_tv data type is an integer that stores a number of seconds. To convert a number of seconds since the start of UNIX time into a sensible format is ctime.

#include
char* ctime(const time_t* timep)
The ctime() function converts the calendar time timep into a string of the form
"Wed Jun 30 21:49:08 1993\n"

WHO version 2

/* -*- Mode: C; indent-tabs-mode: t; c-basic-offset: 4; tab-width: 4 -*- */
/*
* main.c - Auto-generated by Anjuta's Makefile project wizard
*
*/

#include
#include
#include
#include
#include
#define SHOWHOST

void show_info(struct utmp*);
void show_time(long);

int main()
{
struct utmp current_record; /*read info into this structure*/
int utmpfd; /*the file handler*/
int reclen;

reclen = sizeof(current_record); /*reclen is 384 bytes*/
utmpfd = open (UTMP_FILE,O_RDONLY); /*UTMP_FILE is define in utmp.h, to be
precise in paths.h*/
while (read(utmpfd, &current_record,reclen)==reclen)
show_info(&current_record);
}

void show_info(struct utmp* utbufp)
{
if ((*utbufp).ut_type != USER_PROCESS)
return;
printf("%-8.8s", (*utbufp).ut_name);
printf(" ");
printf("%-8.8s", (*utbufp).ut_line);
printf(" ");
show_time((*utbufp).ut_tv.tv_sec);

#ifdef SHOWHOST
printf("(%s)",(*utbufp).ut_host);
#endif
printf("\n");

}

void show_time(long timeval)
{
char* cp; /*to hold address of time in string*/
cp = ctime(&timeval);
printf("%12.12s",cp+4);

}

Output WHO version 2

juniarto tty7 Mar 30 09:01(:0)
juniarto pts/0 Mar 30 09:39(:0.0)
juniarto pts/1 Mar 30 09:48(:0.0)
juniarto pts/2 Mar 30 13:00(:0.0)

Linux Socket Programming 2 (UDP Client)

2009-03-26T00:11:00.001-07:00

Linux Socket Programming 2 (UDP Client)

The client

Code	Explanation
/* -- Mode: C; indent-tabs-mode: t; c-basic-offset: 4; tab-width: 4 -- / / * main.c - Auto-generated by Anjuta's Makefile project wizard * */ #include #include #include #include #include #define MAXBUF 1024	I am using Anjuta IDE to code. Some header-files need to be included. MAXBUF is defined.
int main(int argc, char* argv[]) { int udpSocket; int returnStatus; int addrlen; struct sockaddr_in udpClient, udpServer; char buf[MAXBUF]; if (argc <3) { fprintf(stderr, "Usage: %s \n", argv[0]); exit(1); }	Usage: ./udpclient
/create a socket / udpSocket = socket(AF_INET, SOCK_DGRAM,0); if (udpSocket == -1) { fprintf(stderr, "Could not create a socket!\n"); exit(1); } else { printf("Socket created.\n"); }	The first step is to create socket. The function synopsis for socket function is provided in #include #include int socket (int family, int type, int protocol) family: you can use AF_INET for TCP/IP, or AF_UNIX for internal. In this case we use AF_INET. type: SOCK_DGRAM (for UDP), SOCK_STREAM(for TCP) protocol: typically a zero (0) to use the default for a given family an type.
/client address / /use INADDR_ANY to use all local addresses/ udpClient.sin_family = AF_INET; udpClient.sin_addr.s_addr = INADDR_ANY; udpClient.sin_port = 0;	it is based on address structure struct sockaddr_in { short sin_family; u_short sin_port; /protocol port number / /network byte order/ u_long sin_addr; /net address for the remote host/ /network byte ordered/ char sin_zero[8]; /unused, set to zero/ }
returnStatus = bind(udpSocket, (struct sockaddr_in *)&udpClient, sizeof(udpClient)); if (returnStatus==0){ fprintf(stderr, "Bind completed!\n"); } else{ fprintf(stderr, "Could not bind to address!\n"); close(udpSocket); exit(1); }	We need to bind the address to our socket. int bind(int sockfd, (struct sockaddr) my_addr, socklen_t addrlen); socfd:socket descriptor. in our case is the socket udpSocket (struct sockaddr) my_addr, written as (struct sockaddr*)&udpServer addrlen, the length of of the sockaddr structure, written as sizeof(udpServer)
/client address / /use INADDR_ANY to use all local addresses/ udpClient.sin_family = AF_INET; udpClient.sin_addr.s_addr = INADDR_ANY; udpClient.sin_port = 0;	it is based on address structure struct sockaddr_in { short sin_family; u_short sin_port; /protocol port number / /network byte order/ u_long sin_addr; /net address for the remote host/ /network byte ordered/ char sin_zero[8]; /unused, set to zero/ }
returnStatus = bind(udpSocket, (struct sockaddr_in )&udpClient, sizeof(udpClient)); if (returnStatus==0){ fprintf(stderr, "Bind completed!\n"); } else{ fprintf(stderr, "Could not bind to address!\n"); close(udpSocket); exit(1); } /set up the message to be sent to the server */ strcpy(buf, "For Professionals, By Professionals.\n");	We need to bind the address to our socket. int bind(int sockfd, (struct sockaddr) my_addr, socklen_t addrlen); socfd:socket descriptor. in our case is the socket udpSocket (struct sockaddr) my_addr, written as (struct sockaddr*)&udpClient addrlen, the length of of the sockaddr structure, written as sizeof(udpClient) prepare the message.
/server address / /use the command-line arguments / udpServer.sin_family = AF_INET; udpServer.sin_addr.s_addr = inet_addr(argv[1]); udpServer.sin_port = htons(atoi(argv[2]));	address structure for server
/send message to server/ returnStatus = sendto(udpSocket, buf, strlen(buf)+1, 0, (struct sockaddr_in *)&udpServer, sizeof(udpServer)); if (returnStatus == -1){ fprintf(stderr, "Could not send message!\n"); } else{ printf("Message sent.\n");	int sendto(int s, const void* msg, int len, unsigned flags, const struct sockaddr* to,int tolen); s: the socket to send the datagram to msg: a pointer to the buffer holding the datagram message that you wish to send. len: the length, in bytes, of the datagram that starts at the pointer given my msg flags: use 0. to: a pointer to a generic socket address that you have established. This is the address of the recipient of the datagram. tolen: the length of the address argument to.
/message sent: look for confirmation / addrlen = sizeof(udpServer); returnStatus = recvfrom(udpSocket, buf, MAXBUF, 0, (struct sockaddr_in )&udpServer, &addrlen); if (returnStatus == -1){ fprintf(stderr, "Did not receive confirmation!\n"); } else{ buf[returnStatus] = 0; printf("Received: %s\n", buf); } } / cleanup */ close(udpSocket); return 0; }	int recvfrom(int s, void* buf, int len, unsigned flags, struct sockaddr* from, int* fromlen); s: the socket to receive the datagram from buf: the buffer pointer to start receiving the datagram into len: the maximum length in bytes of the receiving buffer buf flags: option flag. use 0. from: &udpClient (client address structure) fromlen: maximum length in bytes->length of udpClient. Note that the integer that this pointer points to must be initialized to the maximum size of the receiving address structure from, prior to calling the function.

2009-03-25T23:21:00.000-07:00

Linux Socket Programming 1 (UDP Server)

In this article, we are going to explore how a server and a client communicate using UDP. The client will send a text to the server,
then the server will print out the text, and send the client a message that it has received the text.

The server

The Code	Explanation
/* -- Mode: C; indent-tabs-mode: t; c-basic-offset: 4; tab-width: 4 -- / / * main.c - Auto-generated by Anjuta's Makefile project wizard * */ #include #include #include #include #include #include #define MAXBUF 1024	I am using Anjuta IDE to code. Some header-files need to be included. MAXBUF is defined.
int main(int argc, char* argv[]) { int udpSocket; int returnStatus = 0; int addrlen = 0; struct sockaddr_in udpServer, udpClient; char buf[MAXBUF]; /check for the right number of arguments / if (argc <>\n",argv[0]); exit(1); }	The usage of our server is ./udpserver 89 where 89 is the port no to be used for communication.
/create a socket / udpSocket = socket(AF_INET, SOCK_DGRAM, 0); if (udpSocket == -1) { fprintf(stderr,"Could not create a socket!\n"); exit(1); } else{ printf("Socket created.\n"); }	The first step is to create socket. The function synopsis for socket function is provided in #include #include int socket (int family, int type, int protocol) family: you can use AF_INET for TCP/IP, or AF_UNIX for internal. In this case we use AF_INET. type: SOCK_DGRAM (for UDP), SOCK_STREAM(for TCP) protocol: typically a zero (0) to use the default for a given family an type.
/set up the server address and port / /use INADDR_ANY to bind to all local addresses / udpServer.sin_family = AF_INET; udpServer.sin_addr.s_addr = htonl(INADDR_ANY); /use the port passed as argument / udpServer.sin_port = htons(atoi(argv[1]));	it is based on address structure struct sockaddr_in { short sin_family; u_short sin_port; /protocol port number / /network byte order/ u_long sin_addr; /net address for the remote host/ /network byte ordered/ char sin_zero[8]; /unused, set to zero/ }
/bind to the socket/ returnStatus = bind(udpSocket, (struct sockaddr_in *)&udpServer,sizeof(udpServer)); if (returnStatus == 0) { fprintf(stderr, "Bind completed!\n"); } else { fprintf(stderr, "Could not bind to address!\n"); close(udpSocket); exit(1); }	We need to bind the address to our socket. int bind(int sockfd, (struct sockaddr) my_addr, socklen_t addrlen); socfd:socket descriptor. in our case is the socket udpSocket (struct sockaddr) my_addr, written as (struct sockaddr*)&udpServer addrlen, the length of of the sockaddr structure, written as sizeof(udpServer)
while (1) { addrlen = sizeof(udpClient); returnStatus=recvfrom(udpSocket, buf, MAXBUF,0,(struct sockaddr_in )&udpClient,&addrlen); if (returnStatus == -1) { fprintf(stderr, "Could not receive message!\n"); } else{ printf("Received: %s\n",buf); /a message was received so send a confirmation / strcpy(buf,"OK"); returnStatus = sendto(udpSocket, buf, strlen(buf)+1,0,(struct sockaddr_in)&udpClient,sizeof(udpClient)); if (returnStatus == -1){ fprintf(stderr, "Could not send confirmation!\n"); } else { printf("Confirmation sent.\n"); } } } /cleanup/ close(udpSocket); return 0; }	int recvfrom(int s, void* buf, int len, unsigned flags, struct sockaddr* from, int* fromlen); s: the socket to receive the datagram from buf: the buffer pointer to start receiving the datagram into len: the maximum length in bytes of the receiving buffer buf flags: option flag. use 0. from: &udpClient (client address structure) fromlen: maximum length in bytes->length of udpClient. Note that the integer that this pointer points to must be initialized to the maximum size of the receiving address structure from, prior to calling the function. int sendto(int s, const void* msg, int len, unsigned flags, const struct sockaddr* to,int tolen); s: the socket to send the datagram to msg: a pointer to the buffer holding the datagram message that you wish to send. len: the length, in bytes, of the datagram that starts at the pointer given my msg flags: use 0. to: a pointer to a generic socket address that you have established. This is the address of the recipient of the datagram. tolen: the length of the address argument to.

TFTPD

2009-03-10T22:32:00.000-07:00

At last, my tftpd is working. I am using Ubuntu Hardy, and has spent hours to get tftpd working either using inetd.d or xinetd.
The solution is to forget using xinetd or inetd.d. Comment out tftp in /etc/inetd.conf. Delete tftp in /etc/xinetd.d. So that neither inetd nor xinetd are initiating your tftpd.

I installed tftpd-hpa. Edit /etc/default/tftpd-hpa

RUN_DAEMON="yes"
OPTIONS="-l -s /tftpboot"

start it using sudo /etc/init.d/tftpd-hpa restart

that's it.

Can't wait!

2008-11-04T20:01:00.000-08:00

Birthday's Blue

2008-10-04T22:33:00.000-07:00

Who's laughing

2008-10-01T01:55:00.000-07:00

Internet Explorer Under Linux

2008-09-07T00:24:00.000-07:00

But why? You may ask. Well, I have an IP webcam (dlink DCS-950G), and its web interface can only be accessed using IE, because it is using dlink active-x. The only way to access it is using IE running on WINE. You can download the packages from the following site.

http://www.tatanka.com.br/ies4linux/page/Main_Page

Web 2.0

2008-09-02T06:22:00.000-07:00

Web 2.0. It's revolution in the way we develop web sites. Web 2.0 involves many things, javascript, AJAX, Toolkits (DOMO, YUI,etc), CSS,XML, SOAP, Web Services, Mashups, to name a few. It's a very confusing. So...let's dig deeper. Start with 'philosophical' approach.

Mashups

Web Services

How To Add Permanent Routes in Linux and Windows

2008-08-22T20:52:00.000-07:00

If your machine comes with multiple network cards, you may have to play around with your routing table.

For Windows

route add -p 101.101.101.0 mask 255.255.255.0 101.101.101.254 metric 3
The command will route network address of 101.101.101.0 to gateway 101.101.101.254. The integer that comes after 'metric' denotes the level of priority.

For Linux

Create a file in /etc/network/if-up.d, named it as static-routes. The file should be executable (chmod 755). The content of the file:

#!/bin/bash
#set static routes
#
/sbin/route add -net 101.101.101.0 netmask 255.255.255.0 gw 101.101.101.254

Ashokan Farewell

2008-08-19T21:46:00.000-07:00

My version of Jay Ungar's Ashokan Farewell.

Code	Explanation
/* -- Mode: C; indent-tabs-mode: t; c-basic-offset: 4; tab-width: 4 -- / / * main.c - Auto-generated by Anjuta's Makefile project wizard * */ #include #include #include #include #include #define MAXBUF 1024	I am using Anjuta IDE to code. Some header-files need to be included. MAXBUF is defined.
int main(int argc, char* argv[]) { int udpSocket; int returnStatus; int addrlen; struct sockaddr_in udpClient, udpServer; char buf[MAXBUF]; if (argc <3) { fprintf(stderr, "Usage: %s \n", argv[0]); exit(1); }	Usage: ./udpclient
/create a socket / udpSocket = socket(AF_INET, SOCK_DGRAM,0); if (udpSocket == -1) { fprintf(stderr, "Could not create a socket!\n"); exit(1); } else { printf("Socket created.\n"); }	The first step is to create socket. The function synopsis for socket function is provided in #include #include int socket (int family, int type, int protocol) family: you can use AF_INET for TCP/IP, or AF_UNIX for internal. In this case we use AF_INET. type: SOCK_DGRAM (for UDP), SOCK_STREAM(for TCP) protocol: typically a zero (0) to use the default for a given family an type.
/client address / /use INADDR_ANY to use all local addresses/ udpClient.sin_family = AF_INET; udpClient.sin_addr.s_addr = INADDR_ANY; udpClient.sin_port = 0;	it is based on address structure struct sockaddr_in { short sin_family; u_short sin_port; /protocol port number / /network byte order/ u_long sin_addr; /net address for the remote host/ /network byte ordered/ char sin_zero[8]; /unused, set to zero/ }
returnStatus = bind(udpSocket, (struct sockaddr_in *)&udpClient, sizeof(udpClient)); if (returnStatus==0){ fprintf(stderr, "Bind completed!\n"); } else{ fprintf(stderr, "Could not bind to address!\n"); close(udpSocket); exit(1); }	We need to bind the address to our socket. int bind(int sockfd, (struct sockaddr) my_addr, socklen_t addrlen); socfd:socket descriptor. in our case is the socket udpSocket (struct sockaddr) my_addr, written as (struct sockaddr*)&udpServer addrlen, the length of of the sockaddr structure, written as sizeof(udpServer)
/client address / /use INADDR_ANY to use all local addresses/ udpClient.sin_family = AF_INET; udpClient.sin_addr.s_addr = INADDR_ANY; udpClient.sin_port = 0;	it is based on address structure struct sockaddr_in { short sin_family; u_short sin_port; /protocol port number / /network byte order/ u_long sin_addr; /net address for the remote host/ /network byte ordered/ char sin_zero[8]; /unused, set to zero/ }
returnStatus = bind(udpSocket, (struct sockaddr_in )&udpClient, sizeof(udpClient)); if (returnStatus==0){ fprintf(stderr, "Bind completed!\n"); } else{ fprintf(stderr, "Could not bind to address!\n"); close(udpSocket); exit(1); } /set up the message to be sent to the server */ strcpy(buf, "For Professionals, By Professionals.\n");	We need to bind the address to our socket. int bind(int sockfd, (struct sockaddr) my_addr, socklen_t addrlen); socfd:socket descriptor. in our case is the socket udpSocket (struct sockaddr) my_addr, written as (struct sockaddr*)&udpClient addrlen, the length of of the sockaddr structure, written as sizeof(udpClient) prepare the message.
/server address / /use the command-line arguments / udpServer.sin_family = AF_INET; udpServer.sin_addr.s_addr = inet_addr(argv[1]); udpServer.sin_port = htons(atoi(argv[2]));	address structure for server
/send message to server/ returnStatus = sendto(udpSocket, buf, strlen(buf)+1, 0, (struct sockaddr_in *)&udpServer, sizeof(udpServer)); if (returnStatus == -1){ fprintf(stderr, "Could not send message!\n"); } else{ printf("Message sent.\n");	int sendto(int s, const void* msg, int len, unsigned flags, const struct sockaddr* to,int tolen); s: the socket to send the datagram to msg: a pointer to the buffer holding the datagram message that you wish to send. len: the length, in bytes, of the datagram that starts at the pointer given my msg flags: use 0. to: a pointer to a generic socket address that you have established. This is the address of the recipient of the datagram. tolen: the length of the address argument to.
/message sent: look for confirmation / addrlen = sizeof(udpServer); returnStatus = recvfrom(udpSocket, buf, MAXBUF, 0, (struct sockaddr_in )&udpServer, &addrlen); if (returnStatus == -1){ fprintf(stderr, "Did not receive confirmation!\n"); } else{ buf[returnStatus] = 0; printf("Received: %s\n", buf); } } / cleanup */ close(udpSocket); return 0; }	int recvfrom(int s, void* buf, int len, unsigned flags, struct sockaddr* from, int* fromlen); s: the socket to receive the datagram from buf: the buffer pointer to start receiving the datagram into len: the maximum length in bytes of the receiving buffer buf flags: option flag. use 0. from: &udpClient (client address structure) fromlen: maximum length in bytes->length of udpClient. Note that the integer that this pointer points to must be initialized to the maximum size of the receiving address structure from, prior to calling the function.

The Code	Explanation
/* -- Mode: C; indent-tabs-mode: t; c-basic-offset: 4; tab-width: 4 -- / / * main.c - Auto-generated by Anjuta's Makefile project wizard * */ #include #include #include #include #include #include #define MAXBUF 1024	I am using Anjuta IDE to code. Some header-files need to be included. MAXBUF is defined.
int main(int argc, char* argv[]) { int udpSocket; int returnStatus = 0; int addrlen = 0; struct sockaddr_in udpServer, udpClient; char buf[MAXBUF]; /check for the right number of arguments / if (argc <>\n",argv[0]); exit(1); }	The usage of our server is ./udpserver 89 where 89 is the port no to be used for communication.
/create a socket / udpSocket = socket(AF_INET, SOCK_DGRAM, 0); if (udpSocket == -1) { fprintf(stderr,"Could not create a socket!\n"); exit(1); } else{ printf("Socket created.\n"); }	The first step is to create socket. The function synopsis for socket function is provided in #include #include int socket (int family, int type, int protocol) family: you can use AF_INET for TCP/IP, or AF_UNIX for internal. In this case we use AF_INET. type: SOCK_DGRAM (for UDP), SOCK_STREAM(for TCP) protocol: typically a zero (0) to use the default for a given family an type.
/set up the server address and port / /use INADDR_ANY to bind to all local addresses / udpServer.sin_family = AF_INET; udpServer.sin_addr.s_addr = htonl(INADDR_ANY); /use the port passed as argument / udpServer.sin_port = htons(atoi(argv[1]));	it is based on address structure struct sockaddr_in { short sin_family; u_short sin_port; /protocol port number / /network byte order/ u_long sin_addr; /net address for the remote host/ /network byte ordered/ char sin_zero[8]; /unused, set to zero/ }
/bind to the socket/ returnStatus = bind(udpSocket, (struct sockaddr_in *)&udpServer,sizeof(udpServer)); if (returnStatus == 0) { fprintf(stderr, "Bind completed!\n"); } else { fprintf(stderr, "Could not bind to address!\n"); close(udpSocket); exit(1); }	We need to bind the address to our socket. int bind(int sockfd, (struct sockaddr) my_addr, socklen_t addrlen); socfd:socket descriptor. in our case is the socket udpSocket (struct sockaddr) my_addr, written as (struct sockaddr*)&udpServer addrlen, the length of of the sockaddr structure, written as sizeof(udpServer)
while (1) { addrlen = sizeof(udpClient); returnStatus=recvfrom(udpSocket, buf, MAXBUF,0,(struct sockaddr_in )&udpClient,&addrlen); if (returnStatus == -1) { fprintf(stderr, "Could not receive message!\n"); } else{ printf("Received: %s\n",buf); /a message was received so send a confirmation / strcpy(buf,"OK"); returnStatus = sendto(udpSocket, buf, strlen(buf)+1,0,(struct sockaddr_in)&udpClient,sizeof(udpClient)); if (returnStatus == -1){ fprintf(stderr, "Could not send confirmation!\n"); } else { printf("Confirmation sent.\n"); } } } /cleanup/ close(udpSocket); return 0; }	int recvfrom(int s, void* buf, int len, unsigned flags, struct sockaddr* from, int* fromlen); s: the socket to receive the datagram from buf: the buffer pointer to start receiving the datagram into len: the maximum length in bytes of the receiving buffer buf flags: option flag. use 0. from: &udpClient (client address structure) fromlen: maximum length in bytes->length of udpClient. Note that the integer that this pointer points to must be initialized to the maximum size of the receiving address structure from, prior to calling the function. int sendto(int s, const void* msg, int len, unsigned flags, const struct sockaddr* to,int tolen); s: the socket to send the datagram to msg: a pointer to the buffer holding the datagram message that you wish to send. len: the length, in bytes, of the datagram that starts at the pointer given my msg flags: use 0. to: a pointer to a generic socket address that you have established. This is the address of the recipient of the datagram. tolen: the length of the address argument to.

Merit Can Be Bought, Passion Can't

How to compile libhdfs

Shared Library

To run a program linked with the shared version of the library the operating system must be able to locate the corresponding .so file at runtime. If the library cannot be found, the following error will occur:

Dell system won't boot after ubuntu maverick server installation

Set Up NTP Server and Client

Setting Up NTP Server

ParaView in Ubuntu

JAVA PATH

How to know your ubuntu release

How To Install Sun Java 6 JDK on Ubuntu

How to syntax highlighting to uncommon file type in VIM

How to create CA, Server and Client Certificates

how to compile libcurl and libyaml

How to compile cJSON

Feeling crappy. Where is my coffee?

Sunday Morning

setTimeOut Doesn't Work in IE

Perl Dancer Interesting Functions

How to Upload File using Web Form in Perl Dancer

Multi Cloud Storage API

DBD::mysql installation

File upload to http server

create container using curl

cdmi server access using curl

What is Cloud Storage?

Using Curl to DELETE (delete entry)

Using Curl to POST (create new entry)

Using Curl to PUT Rest Web Service

Using Curl to Access Rest Web Service

How to comment out multiple lines in VIM

Perl Catalyst Step By Step

Cloud Storage

How to Install MongoDB on Ubuntu 8.04

Cloud Storage Implementation

Structured and Unstructured Data, can they be merged?

Working with Feeds

Custom Feeds

Creating the package directory

Adding or package to an existing feed

Creating your own feed

Using the feed

Add a new application to openWRT

Create a New Package

Add source files

Compile on Local PC

Integrate to OpenWrt Build System

Compile

Verify

Common errors

menuconfig does not show the new package

Applet not found

Arduino IDE in UBUNTU HARDY

How To Add Default Gateway in Linux

How to Create RamDisk in Linux

Linux Multicast Router

about int main (int ac, char* av[]){}

Learning to write your own WHO command under linux (UBUNTU hardy)

OK. What is WHO? Who is a command to display information about the users who are currently logged in.

who version 1

Output who version 1

WHO version 2

Output WHO version 2

Linux Socket Programming 2 (UDP Client)

Linux Socket Programming 2 (UDP Client)

Linux Socket Programming 1 (UDP Server)

TFTPD

Can't wait!

Birthday's Blue

Who's laughing

Internet Explorer Under Linux

Web 2.0

How To Add Permanent Routes in Linux and Windows

Ashokan Farewell

To run a program linked with the shared version of the library the operating system must be able to locate the corresponding `.so` file at runtime. If the library cannot be found, the following error will occur: