January 5th, 2009

Steps for fresh Windows XP installation on a Virtual machines.

Preparation
LSI storage controller drivers are not available on the Windows XP installation CD, so complete the following preparation before installation:
1. Download the LSI 53C1030 drivers from the LSI Web site.
2. Using MagicISO or other third-party solutions, create a
.flp image containing the LSI Logic drivers.
3. SCP the floppy image to the virtual machine’s ESX host.
If you are using VirtualCenter 2.5, you can use the VIC to upload the file to the desired datastore.
4. Have a Windows XP CD or ISO image that is accessible form the virtual machine.

Pre-installation Virtual Machine Modifications
Make the following modifications to the virtual machine hardware profile before starting the Windows XP
installation:
1 Using the Virtual Infrastructure Client (VIC), connect to VirtualCenter.
2 Locate the virtual machine that was initially created.
3 Edit the following hardware settings:
a Ensure there is a floppy drive present.
b Ensure the floppy drive is configured to connect at Power On.
c If using a floppy image, ensure the Device Type is set to use a floppy image and is pointing to the LSI
Driver image.
d Ensure the CD/DVD drive is present and configured to connect at power on.
e Ensue the CD/DVD Device Type is configured to point at the Windows XP CD or ISO image.

Installation
Once you complete the pre-installation preparation and modifications, you can install Windows XP:
1 From the Virtual Infrastructure client, connect to VirtualCenter.
2 Power on the virtual machine created earlier.
3 Use the console to view the boot process and to send input to the virtual machine.
4 As the Windows Setup process begins, press F6 to add an additional SCSI driver.
This lets you specify the LSI Logic driver on the floppy image.
The Windows setup process copies all the necessary files to the virtual disk. Complete the setup just as you
would for any normal Windows XP installation. Because this image will be used as a template, however, it is
a good idea to make the configuration as generic as possible. For an instance of customization, see “Creating
a Guest Customization” on page 4).
After completing the Windows setup, perform the following tasks before you finalize the image. Some of these
steps will vary from organization to organization, depending on your Windows imaging standards; some are
optional and noted as such. Many can be managed using a group policy (see “Common GPOs for Managing
Virtual Desktops” on page 5).

December 30th, 2008

Master-Master replication: Two database systems at two different locations can be configured to run in synchronization with each other using Mysql Master-Master replication. Any changes on DB1 at Loc1 can be updated at realtime on DB2 at Loc2 and vice versa using Mysql’s built in replication feature.

Following are steps used to setup a Master-Slave replication :-
1. Assumptions:

1. CentOS 5.2
2. Mysql 5.0
3. • DB1: DbGrmtech
   • Loc1: 10.1.31.9
   • DB2: DbGrmtech
   • Loc2: 10.1.31.6
4. Loc1 is master and Loc2 is slave

2. Create a dump file of DB1 using the command:
mysqldump –port 3306 -u GrmtechUser –password=GrmtechPass -h 10.1.31.9 –databases DbGrmtech –add-drop-table > dbdump.sql

3. Rsync the dbdump file to Loc2 using the command:
rsync -azpSv -e ’ssh -p 99′ dbdump.sql 10.1.31.6:/home/user/domains/dbbackup/grmtech/dbdump.sql

4. Import the dbdump file in Loc2 using the command:
mysql -u GrmtechUser –force DbGrmtech –password=GrmtechPass -h 10.1.31.6 < dbdump.sql

5. Edit Loc1’s /etc/my.cnf file and add the following lines under [mysqld]:
server-id = 1
log-bin = /var/log/mysql/bin.log
binlog-do-db=DbGrmtech

6. Give the command to restart mysql daemon:
service mysqld restart

7. Login to mysql at Loc2 and give the following command:
GRANT REPLICATION SLAVE, REPLICATION CLIENT ON *.* TO ‘crmuser’@'10.1.31.6′ IDENTIFIED BY ‘crmpass’

8. Edit Loc2’s /etc/my.cnf file and add the following lines under [mysqld]:
server-id = 2
master-host = 10.1.31.9
master-port = 3306
master-user = crmuser
master-password = crmpass

replicate-wild-do-table=DbGrmtech.%
replicate-ignore-table = DbGrmtech.tracker

log-bin = /var/log/mysql/bin.log
log-bin-index = /var/log/mysql/log-bin.index
log-error = /var/log/mysql/error.log

relay-log = /var/log/mysql/relay.log
relay-log-info-file = /var/log/mysql/relay-log.info
relay-log-index = /var/log/mysql/relay-log.index

9. Give the command to restart mysql daemon:
service mysqld restart

10. Login to mysql at Loc2 and give the following commands:
Show Slave Status;
If master host setting is not what u set at my.cnf, use this command at mysql prompt:
CHANGE MASTER TO MASTER_HOST=’10.1.31.9′, MASTER_USER=’crmuser’, MASTER_PASSWORD=’crmpass’, MASTER_LOG_FILE=’bin.000001′;

By doing the above settings the changes at DB1@Loc1 will be auto replicated to DB2@Loc2 but not vice versa.

Some commands handy in troubleshooting the entire setup:
At Mysql prompt
1. SHOW MASTER STATUS
2. SHOW SLAVE STATUS
3. START/STOP MASTER
4. START/STOP SLAVE

At command prompt
1. tail -50 /var/log/mysql/bin.000001: to see latest changes in DB1

Helpful links:
1. http://www.onlamp.com/pub/a/onlamp/2005/06/16/MySQLian.html

December 22nd, 2008

Target : We have seven ISS running on multiple servers to provide a better internet connectivity and an ISS routing interface to manually manage the network traffic load by shifting users from one ISS to another one.This article is published to motivate the concept of automatic Network Load Balancing and Traffic Routing to provide a much more better and advanced Network Architecture.

Network Load Balancing, a clustering technology included in the Microsoft Windows 2000 Advanced Server and Datacenter Server operating systems, enhances the scalability and availability of mission-critical, TCP/IP-based services, such as Web, Terminal Services, virtual private networking, and streaming media servers. This component runs within cluster hosts as part of the Windows 2000 operating system and requires no dedicated hardware support. To scale performance, Network Load Balancing distributes IP traffic across multiple cluster hosts. It also ensures high availability by detecting host failures and automatically redistributing traffic to the surviving hosts. Network Load Balancing provides remote controllability and supports rolling upgrades from the Windows NT 4.0 operating system.

The unique and fully distributed architecture of Network Load Balancing enables it to deliver very high performance and failover protection, especially in comparison with dispatcher-based load balancers.

Advantages

Network Load Balancing is superior to other software solutions such as round robin DNS (RRDNS), which distributes workload among multiple servers but does not provide a mechanism for server availability. If a server within the host fails, RRDNS, unlike Network Load Balancing, will continue to send it work until a network administrator detects the failure and removes the server from the DNS address list. This results in service disruption for clients. Network Load Balancing also has advantages over other load balancing solutions—both hardware- and software-based—that introduce single points of failure or performance bottlenecks by using a centralized dispatcher. Because Network Load Balancing has no proprietary hardware requirements, any industry-standard compatible computer can be used. This provides significant cost savings when compared to proprietary hardware load balancing solutions.

The unique and fully distributed software architecture of Network Load Balancing enables it to deliver the industry’s best load balancing performance and availability. The specific advantages of this architecture are described below in the “Network Load Balancing Architecture” section.

Network Load Balancing Architecture

To maximize throughput and high availability, Network Load Balancing uses a fully distributed software architecture. An identical copy of the Network Load Balancing driver runs in parallel on each cluster host. The drivers arrange for all cluster hosts on a single subnet to concurrently detect incoming network traffic for the cluster’s primary IP address (and for additional IP addresses on multihomed hosts). On each cluster host, the driver acts as a filter between the network adapter’s driver and the TCP/IP stack, allowing a portion of the incoming network traffic to be received by the host. By this means incoming client requests are partitioned and load-balanced among the cluster hosts.

Network Load Balancing runs as a network driver logically situated beneath higher-level application protocols, such as HTTP and FTP. Figure 2 below shows the implementation of Network Load Balancing as an intermediate driver in the Windows 2000 network stack.

Figure: Network Load Balancing runs as an intermediate driver between the TCP/IP protocol and network adapter drivers within the Windows 2000 protocol stack Note that although two network adapters are shown, only one adapter is needed to use Network Load Balancing.
This architecture maximizes throughput by using the broadcast subnet to deliver incoming network traffic to all cluster hosts and by eliminating the need to route incoming packets to individual cluster hosts. Since filtering unwanted packets is faster than routing packets (which involves receiving, examining, rewriting, and resending), Network Load Balancing delivers higher network throughput than dispatcher-based solutions. As network and server speeds grow, its throughput also grows proportionally, thus eliminating any dependency on a particular hardware routing implementation. For example, Network Load Balancing has demonstrated 250 megabits per second (Mbps) throughput on Gigabit networks.

Network Load Balancing Performance

The performance impact of Network Load Balancing can be measured in four key areas:

• CPU overhead on the cluster hosts, which is the CPU percentage required to analyze and filter network packets (lower is better).
• Response time to clients, which increases with the non-overlapped portion of CPU overhead, called latency (lower is better).
• Throughput to clients, which increases with additional client traffic that the cluster can handle prior to saturating the cluster hosts (higher is better).
• Switch occupancy, which increases with additional client traffic (lower is better) and must not adversely affect port bandwidth.

In addition, Network Load Balancing’s scalability determines how its performance improves as hosts are added to the cluster. Scalable performance requires that CPU overhead and latency not grow faster than the number of hosts.

Options are available for Linux based Networks too. We just need to explore them.

This was just a small overview. For the complete technical details, please click on the link below-

http://technet.microsoft.com/en-us/library/bb742455.aspx

December 17th, 2008

Video monitoring  system at ao.grmtech

Grmtech is equipped with the latest Video Monitoring Sytem that can record activities and sound in real time. Our infrastructure enables us to make use of remote online surveillance for raising the performance of the company and make the best use of it’s resources.

December 12th, 2008

Here is a VMware virtual appliance for DansGuardian on CentOS 5.2

The virtual appliance is composed of:
* CentOS 5.2
* VMware Tools
* Squid 3
* DansGuardian

Click on below links to download the virtual appliance.
1. centos-dansguardian-virtualappliance.vmx
2. centos-dansguardian-virtualappliance.vmdk

You can use this virtual appliance to do things like:
1. Do phrase matching, PICS filtering and URL filtering.
2. Restrict list of sites.

Username: root
Password: dans12

Steps to configure the virtual appliance:-
1. yum install squid*

2. /etc/rc.d/init.d/squid restart

3. Enable packet forwarding in /etc/sysctl.conf. Set net.ipv4.ip_forward = 1.

4. Make changes to /etc/squid/squid.conf. A copy of the original files are kept with the extension .orig *

5. Yum install dansguardian or do the ./configure if you want the 2.10.0 version

6. Make changes in /etc/dansguardian/dansguardian.conf. A copy of the original files are kept with the extension .orig *

7. Enable A. squid B. iptables C. dansguardian using ntsysv

8. Configure iptables to redirect requests from port 80 to port 8080 using following commands**:-
a. iptables -t nat -A PREROUTING -i eth1 -p tcp -m tcp -d ! <ip-address-of-i-interface> –dport 80 -j REDIRECT –to-ports 8080
b. iptables-save > /etc/sysconfig/iptables
c. service iptables restart

Improving the performance of squid:
Edit squid.conf with these settings:-
maximum_object_size_in_memory 8 MB earlier it was 8KB
cache_mem 800 MB earlier it was 8 MB

Improving the performance of dansguardian:
Edit dansguardian.conf with these settings:-
urlcachenumber = 5000  (default is 1000)
urlcacheage = 9000 (default is 900)

* The difference between original and copied file can be seen using “diff orig_file new_file”
** This will make sure that requests for web sites hosted on this server will not get redirected to port 8080

December 10th, 2008

Google has the maximum search market share accounting for over 69% (according to 2008 stats) from other competitors. This is because Google has never compromised with quality while presenting search results to their visitors. An effort has always been there from Google’s side to find out the best possible search for any given topic. Nowadays, Google can even crawl and index flash presentations.

Black hat seo techniques to become even more malicious: As a smart SEO one has to decide to his risk tolerance while promoting a website. To sustain in the long run in search results, the best way is to be honest and focused in extending services to the visitors. These days SEs have become smarter in dealing with those who do ‘illegal things’  . Google is not going to tolerate the black hat seo techniques like 300 ease in one slide, dns sub domain hijacking, cross site scripting attacks, installation of Malware  in others machine etc.

Emphasis on personalized search: Although Google does produce search results on the basis of IP address of the visitor, but sometimes results appear from other countries too. It’s indeed more impressive if a visitor sees a list of websites that cover service in his / her country or even area. For example, say a visitor of Virginia in USA is searching for home loan in Google. If s/he finds a list of websites (best available websites) that provide home loan in Virginia, nothing can be better than that from SE’s point of view. So presentation of result keeping in mind about personalization as well as localization has become the need of the hour. On the other hand, the websites that facilitates various languages to see its websites are going to get some extra value for Google in the coming days.

Focusing more in User’s satisfaction level rather than ranking: Few SEOs jump on their toes by seeing their website at no.1 in ranking, but the concept has changed today. What is the use of it if your website is getting less number of conversions and moreover, visitors are just leaving your website after few seconds? Smart SEOs are highly concerned about conversions, server logs, content updates etc. If we can focus on how we can satisfy a visitor more by increasing his time of stay in our website, then only the ranking becomes worthy.  So the challenge is not to pay attention to your ranking, it’s to pay attention to the conversion, content and visitors satisfaction level. Google also appreciates if we can extend services by creating helpful videos, e-books etc.

Frequent changes in Meta tags, keyword density etc.: Sometimes we become impatient if we don’t see our website ranking among the top 10 result. We start changing on-page things like page title, meta keywords and descriptions, keyword density etc. Google make a note of such unnecessary optimization stuffs and sometimes throw out the site from its consideration. In stead of doing that, if one can work on adding more services onto the page that will always give a better impression. Say for an insurance website, if the webmaster adds an insurance calculator or similar services that will not only add value but also help in ranking.

December 2nd, 2008

We have a world class datacenter build on latest hardware and software technologies. Some of the latest technologies that we use in our datacenter are IP-KVM, IP Camera, Smooth Wall Firewall.

Our datacenter is already equipped with:-

1. HVAC
2. Fire Detection and Fire Suppression

Servers

IP KVM and USB KVM

SWITCHES

Power Management

Are you wondering how we manage all these servers ?

We use a single display station connected to the IPKVM which is in turn connected to the USB KVM. Here are some pictures explaining how this all works.

We use a single display station connected to the IPKVM which is in turn connected to the USB KVM. Given below is a graphical representation of how it works with explanation.

The connectivity of bus wires starts from A and follows this path-

A -> B -> C -> D -> E -> F -> H & G -> I

A. One end of the VGA Cable is connected to multiple CPUs.
B. Other end of VGA cable is connected with a USB KVM.
C. A USB KVM (Keyboard, Video, Mouse) switch may be referred to as a hardware device that permits a user to control many computers with the help of a single KVM (Keyboard, video and mouse).
D. One end of USB KVM is connected with IP KVM using VGA cable.
E. IP KVM make use of dedicated micro controller as well as video capture hardware which is potentially specialized.This arrangement mainly serves as a means to capture keyboard, video and mouse signals.They are compressed and converted into packets and sent over an Ethernet link to remote console application thereby unpacking and reconstituting dynamic graphical images.

F. One end of VGA cable is connected with an IP KVM which is directly connected to monitor, mouse and a keyboard.

H. Monitor displays the output. With the help of mouse and keyboard we can control the server.
G. It is a network cable (LAN and WAN) connected with switch.
I. Output gets displayed on the VNC using IP access.
Are you wondering how we manage all these cables ?

We use different color network cables to recognize the task that they do. Here are some pictures with the explanation.

We use 5 different colors to distinguish the cables according to their uses.

Red: The Red cable is used to connect WAN i.e. the connection between modem and Internet Sharing Server (ISS).

Green: The Green cable is used to connect LAN i.e. the connection between Internet Sharing Server (ISS) and network switch.

Yellow: The Yellow cable is used to connect the Gateway Computer to network switch.

White: The White cable is used to connect the network switch and the local physical server which runs important virtual server like Samba, Alpha, T2-backup, etc.

Blue: The Blue cable is used to connect the VOIP phone and the Network I/O or network switch.

Entry to the NOC Room is restricted.

Some improvements in the pipeline for the data center are:

Radio Frequency Indentification

December 1st, 2008

The 29th of November 2008, started with a beautiful morning.

All the Grmtechians set out to Narendrapur, at Select House for a great picnic!!! The farm house was awesome, small but done very well. It had a cozy ambiance and a lot of greenery around.

The day was packed with fun, masti and enthusiasm. We had a picnic basket filled with snacks like chips, mint, sandwiches, desserts, candy, fruits and some soft drinks.

There was music in the air. We played a lot of outdoor games. It was an exciting day and would hold some of the greatest memories of being together……………………….

.

November 24th, 2008

rel is a  link attribute and setting its value to no follow tells search engines that certain links on your webpage should not be followed or more importantly should not pass your web page’s hard earned reputation.

<a href=”www.test.com” rel=”nofollow”>test</a>

This can generally be very useful when you have a blog or other areas where you have user generated content like forums, guest books etc.

Suppose you have a blog with public commenting turned on, now links within those comment may pass some your pages reputation to other pages that you may not be willing to endorse. Specially blog commenting is very susceptible to such kind of spammy links. And linking to sites that google considers spammy can affect your website big time. However you can keep a check on such kind of links by CAPTCHA and turning on comment moderation.

Another advantage adding no follow links is suppose you are writing a content and want to add some kind of reference to a particular website but at the same time do not want to pass on your pages reputation you can use this property. For Example: Suppose you are writing a blog post of spammers and want to refer to his website and warn fellow bloggers by putting his links on your post, but you certainly don’t want to share your page’s value. So just by adding rel=”nofolow” to the links will solve your problem.

Also if you want that none of the links on your webpage should be followed you can add that in your robots meta tag in the head section of your website.

<meta name=”robots” content=”nofollow”>

This will nofollow all the link on that webpage

Note: some of the points written above are taken from Google’s Search Engine Optimization Guide