What is SRM?
SRM stands for Storage Resource Management. SRM consists of agents running on the storage server which communicate with the application running on management server. The Management server has the database which collects information from the agents in storage server. and also allows for queries,analysis and reporting. Further, a remote administrator can connect to the management server via Ethernet.
Major Communication technologies used in Storage
The two major communication technologies used in storage industry today are Fiber Channel and i-SCSI.
Functionality
SRM products offer various functionality like collection and retention of data about the storage system. They use the information for performace analysis such as pi pointing storage bottlenecks in SANs or reporting when the systems approach designated capacity.
This information is also used to automate certain things which were done manually previously like storage provisioning, forecasting future needs and networkexpansion management.
SRM products can integrate with backup,business continuity and disaster recovery systems.
Some SRM tools like IBM's Total Storage Productivity Centre capture current device configurations for help with recovery if storage system crashes.
They also recognize storage system virtualization and monitor and report on virtualized assets as well as physical ones.
Capacity Utilization
By integrating with filesystems, SRM provides information about file location and data usage.This identifies for ex, little used data that org could archive in lower priority less expensive storage facilities.
Collecting Data
SRM products collect data from storage systems in different ways.They can gather data via systems APIs or command line interfaces or via Secure shell or telnet protocol.
Another new method is the Storage Management Initiative Specification (SMI-S). SMI-S provides common protocols and data models that storage vendors can use. This enables hardware and software interoperability among networked storage products from different vendors.
Coordination with data-center management tools
Many organizations want a single console to manage their data centre not just storage. Thus, some SRM products like:-
1. EMC's Control Center
2. HP's Storage Essentials SRM std Ed
these can communicate with third party system via technologies like Simple Netwrk Management Protocol (SNMP) or commercial integration packages.
Deployment Approaches:
Numerous Products like:-
1. Estorian's Looking glass
2. Monosphere's Storage Horizon
offer SRM feature like cpacity management. They are well suited for limited budget users.
Vendors such as IBM,HP provide broad, integrated SRM functionality in a single product.
System with and without agents
Broad SRM products are typically deployed with agents.The advantage is that these agents analyze each part of the storage system in depth and report back to the management server. The disadvantage is tht these agents must be installed on all storage related equipment and life cycle upgrades applied to each whenever there is an upgrade in the s/w or h/w.
Agentless SRM tools:
Tek tools storage profiler are agentles. This approach is easier to manage because administrators don't have to install and maintain agents on every storage platform.
However, agentless tools workfrom a management interface like API or SMI-S, which don't analyze storage resources in depth and thus don't gather all the information that some administrators require.
They also create communications overhead which increases network traffic and complexity. Agents on other hand, talk directly to SRM systems using their main communications technology such as FC or i-SCSI.
Conclusion
As of now, the major issue is interoperatability between different vendors. Also, once SRM technology matures, many manual processes like reporting,provisioning etc can be automated leading to a roadmap for automation of data centre ops.
Mana
Wednesday, December 16, 2009
Wednesday, October 28, 2009
ROUTING SECURITY
ROUTING SECURITY
BGP uses TCP to connect two routers via port 179. They exchange data about the routers between domains including regularly changing information on the best paths between ASs.
BGPSec
The Directorate for Science & Technology is developing BGPSec and expects to finish within a few years. The technology would add digital signatures and PKI to the BGP process.This is to verify that the routers which send the packets are really the ones that they claim to be.
The technical details of BGPsec haven't been worked out yet.
A PKI enables public key cryptography in which is message is encrypted using a public key which can be opened only by a private key. Therefore,for each public key, a private key is made which makes it impossible for anyone to open the message, but only for the intended router.
senders utilize the public key of the recipient to encrypt the message. the message can only be decrypted by a private key, which only the recipient knows.
How does the sender know that the public key he is using is that of the recipient?Here comes the role of third party certificate authority, which verifies the recipient's ownership of the public key.
An important issue is that the process adds cryptographic overhead to router functionality. This can slow down the infrastructure or require software upgrades which may hinder adoption.
RPKIInternet Engineering Task Force (IETF)'s secure inter domain routing workgroup is working on RPKI (Resource Public Key Infrastructure) to provide specific PKI for routing authentication.
The goal is to create a system that verifies that internet users have been allocated the IP addresses and AS numbers they are working with.The organization like ..ISP or IANA which issues this certificate attests this allocation.
During interent communications, if a router receiving routing data verifies that the sending router's host organization owns the IP address and AS numbers it is using, the recipient would assume that the information it transmits is accurate.
In RPKI, the organizations that allocate IP addresses would act as certificate authorities.
Two other projects which add authentication to routing process are also underway.
S-BGP
BBN has developed Secure BGP (S-BGP) which provides for a specific plan for using PKI and digital certificates to let routers validate other routers, as well as blocks of address space, belong to a specific organization. The PKI lets each org issue certificates to its routers certifying that these routers represent it.
The PKI would let routers verify the owner of one or more address space blocks by binding the owner 's addresses to one of its public keys.
S-BGP would change data packet's structure by adding address-ownership attestations. This implies change in infrastructure software.
SoBGP
SoBGP was developed by Cisco systems which works with a decentralized approach rather than centralized PKI approach that S-BGP uses.
Each AS would operate its own routing authentication system with a database of trustworthy keys and certificates. Over time as more routing data arrives, the database would get bigger.
The system would try to match the stored keys and certificates to arriving routing data to authenticate the information. If incoming routing information doesn't match keys and certificates in the database, the sender would need a digital certificate from a trusted third party to authenticate itself.
VARIOUS TYPES OF ROUTING ATTACKS
SPOOFING
When a device assumes the identity of another router by providing information about itself that actually belongs to another routing domain.
FALSIFICATION
Falsification is sending alse routing data that causes router to send bad information to its peers. Hackers could use it to send flase information about best routes to and from various points into the internet. A BGP peer adds the best route information it receives to its routing tables and advertises the data to other peers. This adds the incorrect information to the internet's global routing tables.
PREFIX HIJACKING
Hackers advertise false best routes throughout the internet and thereby have large amounts of traffic sent through networks they control.
This can result in usurpation, in which an atacker redirects traffic through rogue routers and thereby hijacks a session, gaining control over related routing functions.
InterferenceInterference occurs when an attacker disrupts the communication of legitimate routers.
Overloadoverload takes place when a hacker reroutes large traffic volumes and overwhelms routers.
Man in the middle attacks
Hacker uses false routing information to impersonate two routers and relay messages between them, thereby accessing and controlling hteir communications.
Saturday, August 1, 2009
report on adding Workstations to SAN
The SAN layout has been completely changed. There are four ports on each storage controller and four numbers of controllers. There are six numbers of SAN switches in cluster Group-A. The layout of the controller ports and their conenction to SAN switches are shown in figure A.
Table-1 shows the connection of front end servers (ns1 to ns8).
The workstations were accessing the GPFS file system through Ethernet earlier. There were performance issues while doing graphics apps which queried the storage data through ethernet. It was certain that adding a fibre connectivity to the work station which talks directly to the storage via SAN swich would improve performance to a very high degree.
once the w/s and storage controllers were connected via SAN switch, the workstation HBA's were visible in the Storage Manager interface. Then they were added to each of the four controllers.
in the workstation, RDAC drivers were installed and the LUNs sda,sdb,...sdm were visible in the workstation. GPFS daemon once started, idetified these LUns to be part of file system.
copying a test file showed a throughput improvement of 5 times.
some more insights:-
1.The front end and node servers have two ethernet cards, one with 192 Ip and other with 22
IP.
2. All tape drives are in one SAN switch.
3. There are no ISL between the six SAN switches. each of them is independent.Managed by cisco works using a IP assigned to each switch.
4. The fibre HBA has three LED's indicating the speed. Green, yellow and red for 1,2 & 4 Gb/s.
5. If the Tx-Rx mate is correct while plugging in the LC cable to HBA, only one among the three LED glows. If the cable is improperly connected, the three LEd's blink.
Table-1 shows the connection of front end servers (ns1 to ns8).
The workstations were accessing the GPFS file system through Ethernet earlier. There were performance issues while doing graphics apps which queried the storage data through ethernet. It was certain that adding a fibre connectivity to the work station which talks directly to the storage via SAN swich would improve performance to a very high degree.
once the w/s and storage controllers were connected via SAN switch, the workstation HBA's were visible in the Storage Manager interface. Then they were added to each of the four controllers.
in the workstation, RDAC drivers were installed and the LUNs sda,sdb,...sdm were visible in the workstation. GPFS daemon once started, idetified these LUns to be part of file system.
copying a test file showed a throughput improvement of 5 times.
some more insights:-
1.The front end and node servers have two ethernet cards, one with 192 Ip and other with 22
IP.
2. All tape drives are in one SAN switch.
3. There are no ISL between the six SAN switches. each of them is independent.Managed by cisco works using a IP assigned to each switch.
4. The fibre HBA has three LED's indicating the speed. Green, yellow and red for 1,2 & 4 Gb/s.
5. If the Tx-Rx mate is correct while plugging in the LC cable to HBA, only one among the three LED glows. If the cable is improperly connected, the three LEd's blink.
Sunday, April 26, 2009
cat 6 color coding for straight and cross cable
T568A
1. White green
2. Green
3. White orange
4. Blue
5. White blue.
6. Orange
7. White brown
8. Brown
All odd pins are white and even pins are colored. It starts with white orange or green and ends in brown..aid to memory.
T568 B just interchange green and orange in T 568A
1. White orange
2. Orange
3. White green
4. Green
5.white blue
6. Blue
7. White brown
8. Brown
T568A- T 568B is cross cable.
T568A- T 568A is straight.
T568B- T 568B is straight.
RJ 45 connector lock down position in front of you, from left is pin1 and extreme right is pin 8.
Cross Cable: 13-26
that is, after crimping one end staright, the other end would be:-
1. pin1 goes to pin3 (ie,pin3 would be rep by OW), Pin 3 goes to pin 1 (pin1 would be rep by GW).
2.pin2 goes to pin6 (pin6 would be rep by O) and Pin 6 goes to pin 2 (pin 2 would be rep by G).
Wednesday, April 8, 2009
circuit for simple li ion bat charger
.jpg)
a not so recommended charger, can say its a simple dc source utilizing the buck topology. circuit traced from pcb of a chinese mp3 player charger. This charger used to always damage the battery.
WDM PON
PON is point to multipoint optical network that lets a single fiber trunk provide data to many different nodes including individual customers who areserved digitally and interim pedestal facilities that connect electrically to customers.
PONs traditionally used for local rather than long distance transmission thus reduce the amount of fiber that a network requires.
PONs are passive, because they use pricesly engineered optical components that split and route signals via refraction rather than electrical process.
PON implementation
Mid 90s-> FSAN (Full service Access network) std given by consortium of equipment providers and ITU
ATM PON (APON)
Broadband PON (BPON)
Gigabit PON (GPON)
802.3 ah Ethernet PON (EPON)
NTT deployed world's first commercial PON, an APON network in Japan in year 2000.
Currently carriers use either GPON (2.4 Gbps down, 1.2 Gbps up) or EPON (1 GBps both ways).
ITU, FSAN and IEEE are working on faster versions of these approaches.
HOW they Work?EPON and GPON --> TDM (Time division multiplexing)
WPON
PONs traditionally used for local rather than long distance transmission thus reduce the amount of fiber that a network requires.
PONs are passive, because they use pricesly engineered optical components that split and route signals via refraction rather than electrical process.
PON implementation
Mid 90s-> FSAN (Full service Access network) std given by consortium of equipment providers and ITU
ATM PON (APON)
Broadband PON (BPON)
Gigabit PON (GPON)
802.3 ah Ethernet PON (EPON)
NTT deployed world's first commercial PON, an APON network in Japan in year 2000.
Currently carriers use either GPON (2.4 Gbps down, 1.2 Gbps up) or EPON (1 GBps both ways).
ITU, FSAN and IEEE are working on faster versions of these approaches.
HOW they Work?EPON and GPON --> TDM (Time division multiplexing)
WPON
WDM PONS use either AWG (Arrayed waveguide grating) or thin film filters as multiplexers to redirect and route each wavelength of light to appropriate recepient node.
WDM is of two types:-1. Coarse WDM (CWDM)2. Dense WDM (DWDM)
CWDM uses lower cost optical components, offer lower capacity and shorter transmission ranges. CWDM uses thin film filters to multiplex and de-multiplex optical signals.
WDM is of two types:-1. Coarse WDM (CWDM)2. Dense WDM (DWDM)
CWDM uses lower cost optical components, offer lower capacity and shorter transmission ranges. CWDM uses thin film filters to multiplex and de-multiplex optical signals.
Tuesday, April 7, 2009
Network on Chip (Noc)
Microprocessor connects to peripherals using bus. Lately, to speed up the process, multiple topologies and peripherals have been fabricated on same chip which contains the processor. This is called system on Chip or SoC.
But, Still the fundamental problem of communication between the mup and peripherals still used to cause a bottleneck.
Driving Forces:-Why Noc Approach is needed?
Single Chip embeded systems such as IBM's cell and those used in HDTV are becoming increasingly complex. Such Soc have many cores which perform distinct function, DSP, graphics etc and operate at different clock frequencies. This complicates on chip communication. For example, if no. of cores on a processor increases, then number of potential communication paths between them rises exponentially and with more components, a single bus must address communication over a larger area, which leads to latency.
Typically, chip makers have tried to segment buses for use with different elements on Socs. This process turns buses from being long wires that are globally clocked and stretch the entire length of chip to shorter segments of locally clocked bundled wires connected to bridges at each end.
However, segmenting buses is different for complex Soc and multicore chips. Buses must be manually designed with different segments that suit specific chip architectures. This can be expensive and time consuming process.
Technical Issues:-
As the wires on a chip increases the wires get closer to each other thus contributing to parasitic capacitance. This affects performance.
How it Works:-
Some HPC and data n/w processors could benefit from circuit switching Noc techniques. However, the approach does not provide for routing flexibility and thus does not avoid n/w congestion. Therefore, today's NOCs use packet switching.
Packet Switched Systems:-
Each core has a NIU or n/w interface unit which packetizes the data. The NIUs of each core and that of external peripherals (which are on the Soc itself) are connected by copper paths to the router (residing on Soc itself). As in telecom systems, Noc technology uses routing algorithms and tables to decide the optimal way to send packets.
Networking Techniques:-
On chip communication must be very fast,latency free and flexible. Thus, the n/w techniques must be simple. The trade off is that simple network offers fewer capabilities.
But, Still the fundamental problem of communication between the mup and peripherals still used to cause a bottleneck.
Driving Forces:-Why Noc Approach is needed?
Single Chip embeded systems such as IBM's cell and those used in HDTV are becoming increasingly complex. Such Soc have many cores which perform distinct function, DSP, graphics etc and operate at different clock frequencies. This complicates on chip communication. For example, if no. of cores on a processor increases, then number of potential communication paths between them rises exponentially and with more components, a single bus must address communication over a larger area, which leads to latency.
Typically, chip makers have tried to segment buses for use with different elements on Socs. This process turns buses from being long wires that are globally clocked and stretch the entire length of chip to shorter segments of locally clocked bundled wires connected to bridges at each end.
However, segmenting buses is different for complex Soc and multicore chips. Buses must be manually designed with different segments that suit specific chip architectures. This can be expensive and time consuming process.
Technical Issues:-
As the wires on a chip increases the wires get closer to each other thus contributing to parasitic capacitance. This affects performance.
How it Works:-
Some HPC and data n/w processors could benefit from circuit switching Noc techniques. However, the approach does not provide for routing flexibility and thus does not avoid n/w congestion. Therefore, today's NOCs use packet switching.
Packet Switched Systems:-
Each core has a NIU or n/w interface unit which packetizes the data. The NIUs of each core and that of external peripherals (which are on the Soc itself) are connected by copper paths to the router (residing on Soc itself). As in telecom systems, Noc technology uses routing algorithms and tables to decide the optimal way to send packets.
Networking Techniques:-
On chip communication must be very fast,latency free and flexible. Thus, the n/w techniques must be simple. The trade off is that simple network offers fewer capabilities.
- TCP/IP entails too much latency and wouldn't be helpful.
- Open core protocol international partnership thus developed the OCP standard for on chip communications with which some NOC vendors work.
Architectures and Topologies
It depends and changes from chip to chip.
Mesh NOC topology :- Used where high degree of parallelism required. Well suited for multiprocessor Socs whoose core must run in parallel. Used in Technion univ Israel.
Clustered Mesh:- BONE (Basic on chip network) developed by Korea Advanced institute of Science and Technology.
Flat Tree topology:- SPIN of Piere & Marie Curie university.
Advantages:-
- Processor cores can focus on processing rather than inter-core communication.
- Nocs work with multiple routers or switches connected by shorter wires, so they are more energy efficient.
- Shorter connections reduce complexity of designing wires to yield predictable speed, power,noise and reliability..
Performance:-
Noc Solution makes chips perform three times as fast as they would be using conventional bus system.
Commercial implementations:-
ST microelectronics designs and produces its own system VSTNOC which company uses in its chips including those for HDTV.
About 300 million Socs that use sonics Inc SMART interconnects NOC technology appear in laptops, PC's, HDTVs, Smart Phones, Gaming consoles etc.
Companies also provide tools that manufacturers can use to incorporate and customize NOCs within their chip designs. For Ex: Alteris provides assemblers, compilers, component libraries, traffic analyzers and simulators. Texas instruments is using them in its OMAP4 family of mobile processors.
FUTURE:
During next five years, NOC adoption will expand to application such as Symmetric Multi Processing (SMP), ASIC, FPGA and internet routers.
Sunday, February 8, 2009
making an active x control work in vista
1.made a simple active x control with a combo box and a public function. compiled and made the ocx
2. in client machine, unregistered the previous version using regsvr32 /u {file_name} and registered the new control.
3. in client browser, added the intranet site to trusted sites and set the security to low and to allow unsigned active x.
4. This arrangement worked and the active x was able to execute inside the browser.
2. in client machine, unregistered the previous version using regsvr32 /u {file_name} and registered the new control.
3. in client browser, added the intranet site to trusted sites and set the security to low and to allow unsigned active x.
4. This arrangement worked and the active x was able to execute inside the browser.
Monday, February 2, 2009
installing windows xp by shrinking vista
Procedure for Installing XP on the HP compaq 6710b Laptop without uninstalling Vista and retaining your preinstalled applications like Office 2007 & Norton AV
First get a computer or a laptop with a working and fast internet connection before doing anything. Then if you feel convenient working with the system, follow the steps below or get some help of a friend who can do it for you. Do this only if you feel you can do these steps, else, you can mess up things bad and prevent your laptop from booting up. Also, installing XP may void the warranty..im not sure about this. But, if you feel that Xp is absolutely necessary, then go ahead…
First create a partition in Vista for installing XP (make it a convenient size. Anything from 2 GB upwards will do). Follow the instructions here:-
http://www.howtogeek.com/howto/windows-vista/install-windows-xp-on-your-pre-installed-windows-vista-computer/
Name this partition as XP, and assign drive letter as D.(you may have to rename the CD drive to some other alphabet)
Insert XP cd in the drive and reboot the laptop. Installation of Xp would begin, choose it to install it in D:\. If an error like “setup did not find any hard disk drives installed” appears then you can prepare your custom XP setup disk with the SATA support . Two driver files (intel AHCI SATA support)( iaahci.inf, iastor.inf) for SATA support have to be downloaded from the site given in S.No. 3 below. Run this setup file (sp37005.exe) and Save these files in a convenient location and .Click the link below to do this:-
http://www.howtogeek.com/howto/windows/resolving-setup-did-not-find-any-hard-disk-drives-during-windows-xp-installation/
After XP is installed, you have to make sure that all drivers for the laptop are installed. The driver download page for this model is located here:-
http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareIndex.jsp?lang=en&cc=us&prodNameId=3356622&prodTypeId=321957&prodSeriesId=3356620&swLang=8&taskId=135&swEnvOID=1093
Now, when u restart the laptop, u will get only XP, on completing this step would bring about a boot menu, which would give you the choice of selecting between vista & XP. For the time being boot in XP and download the following tools(u can look up on google):-
(a) Microsoft .Net Framework 2.0 (about 23 MB)
(b) Vista boot pro 3
After installing vista boot pro, follow the instructions given in the link in 1 above. At this stage I had a problem, that vista was booting properly, but Xp wasn’t. I selected ‘Add New OS Entry’ in vista boot pro 3. I chose legacy version of Windows and named it XP. Then I rebooted, and found that XP & vista were working fine.
If you get /ntldr error while booting in Xp, then copy two files namely, ntldr and ntdetect from your windows XP CD (I386 folder) to D:\ drive, and reboot. This time it should boot allright
First get a computer or a laptop with a working and fast internet connection before doing anything. Then if you feel convenient working with the system, follow the steps below or get some help of a friend who can do it for you. Do this only if you feel you can do these steps, else, you can mess up things bad and prevent your laptop from booting up. Also, installing XP may void the warranty..im not sure about this. But, if you feel that Xp is absolutely necessary, then go ahead…
First create a partition in Vista for installing XP (make it a convenient size. Anything from 2 GB upwards will do). Follow the instructions here:-
http://www.howtogeek.com/howto/windows-vista/install-windows-xp-on-your-pre-installed-windows-vista-computer/
Name this partition as XP, and assign drive letter as D.(you may have to rename the CD drive to some other alphabet)
Insert XP cd in the drive and reboot the laptop. Installation of Xp would begin, choose it to install it in D:\. If an error like “setup did not find any hard disk drives installed” appears then you can prepare your custom XP setup disk with the SATA support . Two driver files (intel AHCI SATA support)( iaahci.inf, iastor.inf) for SATA support have to be downloaded from the site given in S.No. 3 below. Run this setup file (sp37005.exe) and Save these files in a convenient location and .Click the link below to do this:-
http://www.howtogeek.com/howto/windows/resolving-setup-did-not-find-any-hard-disk-drives-during-windows-xp-installation/
After XP is installed, you have to make sure that all drivers for the laptop are installed. The driver download page for this model is located here:-
http://h20000.www2.hp.com/bizsupport/TechSupport/SoftwareIndex.jsp?lang=en&cc=us&prodNameId=3356622&prodTypeId=321957&prodSeriesId=3356620&swLang=8&taskId=135&swEnvOID=1093
Now, when u restart the laptop, u will get only XP, on completing this step would bring about a boot menu, which would give you the choice of selecting between vista & XP. For the time being boot in XP and download the following tools(u can look up on google):-
(a) Microsoft .Net Framework 2.0 (about 23 MB)
(b) Vista boot pro 3
After installing vista boot pro, follow the instructions given in the link in 1 above. At this stage I had a problem, that vista was booting properly, but Xp wasn’t. I selected ‘Add New OS Entry’ in vista boot pro 3. I chose legacy version of Windows and named it XP. Then I rebooted, and found that XP & vista were working fine.
If you get /ntldr error while booting in Xp, then copy two files namely, ntldr and ntdetect from your windows XP CD (I386 folder) to D:\ drive, and reboot. This time it should boot allright
Sunday, January 18, 2009
my sql
Creating a table
mysql> create table guests (autoID,int unsigned,auto_increment)
mysql> insert into guests VALUES (NULL, 'ádam','pop',46) same order as the fields created.
date is typed as "1990-03-05" year,month,day
to delete a table
mysql> drop table {table_nm};
to display the table:-
mysql> select * from guests;
mysql> update guests SET last_name="xyz123" WHERE autoID=1;
mysql> create table guests (autoID,int unsigned,auto_increment)
mysql> insert into guests VALUES (NULL, 'ádam','pop',46) same order as the fields created.
date is typed as "1990-03-05" year,month,day
to delete a table
mysql> drop table {table_nm};
to display the table:-
mysql> select * from guests;
mysql> update guests SET last_name="xyz123" WHERE autoID=1;
MYSQL ADMIN tasks
monitoring status of database server, uptime etc
mysql>status;
Show the databases mounted:-
mysql>show databases;
Create a database
mysql>create database {name};
use a database i.e., use the databse to create tables in it:-
mysql> use {database_name};
Delete a database:-
mysql> drop database {database_name};
Show tables in a database:-
mysql>show tables;
Show the fields in a table:-
mysql> describe {table_name};
Saturday, January 17, 2009
working with debian distro
1. TO LIST ALL DRIVES INSTALLED IN THE SYSTEM
Type the following at prompt:-
/sbin/fdisk -l
The fdisk file is in sbin directory of root.
2. Mounting NTFS & FAT file systems
Mounting is a process where a filesystem is mounted to a particular folder in the system.
(a) NTFS
1. Create a folder /windows/c in media folder in root directory (sudo mkdir /media/windows/c)
2. mount /dev/ windows/c -t ntfs -o nls=utf8,umask=0222
(b) FAT32
1. Create a folder /windows/d in media folder in root directory (sudo mkdir /media/windows/d)
2. mount /dev/ windows/d
Type the following at prompt:-
/sbin/fdisk -l
The fdisk file is in sbin directory of root.
2. Mounting NTFS & FAT file systems
Mounting is a process where a filesystem is mounted to a particular folder in the system.
(a) NTFS
1. Create a folder /windows/c in media folder in root directory (sudo mkdir /media/windows/c)
2. mount /dev/
(b) FAT32
1. Create a folder /windows/d in media folder in root directory (sudo mkdir /media/windows/d)
2. mount /dev/
Estimating Number of cylinders for a partition based on size estimate
Disk Basics
1.Track: The circular track on a disk
2.Geometric Sector: the Sector made by two radii intersecting all the tracks in the disk from the centre of the disk.
3.Track Sector: The intersection of two geometric radii on the same track which is a part of the geometirc sector is called a track sector.
4. Size of sector:- The size of sector for magnetic disks is 512 bytes and for optical media is normally 2048 bytes.
5. Clusters :- A collection of geometric sectors make a cluster for a filesystem. The cluster may not be comprised of continous sectors. The cluster size defines the least amount of size defined by the filesystem while creating the file.
The basic info while starting fdisk to partition a hard drive may be given in the manner shown below:-
64 heads,32 sectors
Which means there are 32 platters and each platter has a head on top and bottom,therefore, 64 heads and each platter(single side) contributes 32 sectors to a cylinder.
A cylinder is formed by taking one sector on each platter (above and below).
Total no of sectors in a cylinder= No of sectors contributing to a cylinder in the platter * No of platters
32 * 32 * 2= 2048 sectors
We have multiplied by 2 as platter has sectors both above and below.
Another statement in Fdisk is “capacity is cylinders of 2048 X 512 bytes
The above statement indicates that each sector can store 512 bytes. This can also be called as block size and each cylinder contains 2048 sectors. Therefore, each cylinder can store 2048* 512=1048576 bytes
Partition of the drive is done in cylinders . Fdisk also reports that the total no. of cylinders available is 8682. Therefore, if we want swap space of 128 MB, the number of cylinders required would be :-
128 * 1024*1024/1048576=128 cylinders.
Out of 8682 cylinders, 128 would be kept for swap, therefore, for root we have 8682-128=8554 cylinders
Therefore, for a disk with SGI label, The picture before root and swap partitions are created would be :-
Partition # Start End Description
9 0 1 SGI volhdr
11 0 8682 SGI volume
Now, we create partition # 1 (root) and Partition No 2 (swap) using 8554 cylinders and 128 cylinders respectively
Partition # Start End Description
9 0 1 SGI volhdr
11 0 8682 SGI volume
1 2 8554 Linux
2 8555 8682 Swap
1.Track: The circular track on a disk
2.Geometric Sector: the Sector made by two radii intersecting all the tracks in the disk from the centre of the disk.
3.Track Sector: The intersection of two geometric radii on the same track which is a part of the geometirc sector is called a track sector.
4. Size of sector:- The size of sector for magnetic disks is 512 bytes and for optical media is normally 2048 bytes.
5. Clusters :- A collection of geometric sectors make a cluster for a filesystem. The cluster may not be comprised of continous sectors. The cluster size defines the least amount of size defined by the filesystem while creating the file.
The basic info while starting fdisk to partition a hard drive may be given in the manner shown below:-
64 heads,32 sectors
Which means there are 32 platters and each platter has a head on top and bottom,therefore, 64 heads and each platter(single side) contributes 32 sectors to a cylinder.
A cylinder is formed by taking one sector on each platter (above and below).
Total no of sectors in a cylinder= No of sectors contributing to a cylinder in the platter * No of platters
32 * 32 * 2= 2048 sectors
We have multiplied by 2 as platter has sectors both above and below.
Another statement in Fdisk is “capacity is cylinders of 2048 X 512 bytes
The above statement indicates that each sector can store 512 bytes. This can also be called as block size and each cylinder contains 2048 sectors. Therefore, each cylinder can store 2048* 512=1048576 bytes
Partition of the drive is done in cylinders . Fdisk also reports that the total no. of cylinders available is 8682. Therefore, if we want swap space of 128 MB, the number of cylinders required would be :-
128 * 1024*1024/1048576=128 cylinders.
Out of 8682 cylinders, 128 would be kept for swap, therefore, for root we have 8682-128=8554 cylinders
Therefore, for a disk with SGI label, The picture before root and swap partitions are created would be :-
Partition # Start End Description
9 0 1 SGI volhdr
11 0 8682 SGI volume
Now, we create partition # 1 (root) and Partition No 2 (swap) using 8554 cylinders and 128 cylinders respectively
Partition # Start End Description
9 0 1 SGI volhdr
11 0 8682 SGI volume
1 2 8554 Linux
2 8555 8682 Swap
TV Tuner on windows choice for capture and viewing
Video for windows (VFW) may be accessed from two drivers:-
1. Traditional (for viewing) This driver delivers good picture quality but has problems while doing capture.
2. New (for capture) :- This driver is primarily for capture. capture is smooth and I have captured hours of tv program using this.
Notice the difference between the dialog boxes for choosing the various options, like say the video source etc between these two drivers.
The video for windows vfw drivers can be changed from the WinVDR options menu.
For viewing the video, you can use either the preview or overlay option. The overlay option uses much less CPU than the preview. So, overlay is the better choice.
1. Traditional (for viewing) This driver delivers good picture quality but has problems while doing capture.
2. New (for capture) :- This driver is primarily for capture. capture is smooth and I have captured hours of tv program using this.
Notice the difference between the dialog boxes for choosing the various options, like say the video source etc between these two drivers.
The video for windows vfw drivers can be changed from the WinVDR options menu.
For viewing the video, you can use either the preview or overlay option. The overlay option uses much less CPU than the preview. So, overlay is the better choice.
Friday, January 9, 2009
installing linux on existing NTFS windows XP
I had a Windows PC with two partitions:-
1. 40 GB NTFS (pri )
2. 40 GB FAT (logical drive on ext )
I am currently using this PC to watch TV. The specs aren't too great, Celeron 1 Ghz Coppermine, 256 MB SDRAM (133 Mhz ) Vintron Intel 810e board, and 80 GB IDE HDD.
Windows XP used to run quite slow on the PC, it was managable, if no multiple windows were open. The only problem in mind was a doubt as to whether the tv tuner card would work in linux. This was the reason I had postponed installation.
However, one day while browsing the net I found that the chipset used in TV tuner card Bt 878 was supported in debain, I decided to upgrade to linux. Now, I had another dilemma..how to keep the existing windows and make a dual boot system.
1. 40 GB NTFS (pri )
2. 40 GB FAT (logical drive on ext )
I am currently using this PC to watch TV. The specs aren't too great, Celeron 1 Ghz Coppermine, 256 MB SDRAM (133 Mhz ) Vintron Intel 810e board, and 80 GB IDE HDD.
Windows XP used to run quite slow on the PC, it was managable, if no multiple windows were open. The only problem in mind was a doubt as to whether the tv tuner card would work in linux. This was the reason I had postponed installation.
However, one day while browsing the net I found that the chipset used in TV tuner card Bt 878 was supported in debain, I decided to upgrade to linux. Now, I had another dilemma..how to keep the existing windows and make a dual boot system.
Dual Boot to Windows
I googled and found this page http://www.nishants.net/articles/ntfsresize.htm . He has neatly mentioned the process. I must say google is a real help. hats off to the people who perfected the search algorithm! My Plan was to install Ubuntu Linux which is a debain based distro. The idea as described in nishant's site was to first 'shrink' the existing windows NTFS partition and then carve out free space for the linux installation. To shrink the partition, and carve out maximum space, i literally emptied my C:\ drive and degragmented it. The idea was to use UBUNTU live CD to do this task. However, ubuntu does not have support for NTFS filesystem. So, two debain packages called libntfs5_1.9.0-1_i386.deb and ntfstools_1.9.0-1_i386.deb were required, once again I googled and found the files, these two files, I copied to the C:\ drive.
Next, I booted using the UBuntu live CD and followed the instructions as given in nishant's site. I could shrink my windows installation to 25 GB as I had desired. This had earmarked around 17 GB of free space for Linux install.
Realizing the Free Space
One more step was necessary however, before this free space was realized. For this I reguired a partition tool. I choose sfdisk. downloaded it (again using google) and made a bootable CD. I booted using this CD and the partition manager launched automatically. The parition manager warned me that there was some error regarding the number of sectors and the space allocated to the volume and whether I wanted to correct it. I clicked yes and immediately, the partition manager showed the free space along with the resized primary partition. The existing logical drive in extended FAT32 partition was also shown. I saved changes, inserted the Ubuntu setup CD and rebooted.
Setting Up UBuntu
I followed the instructions contained in http://ca.geocities.com/zachandloricox@rogers.com/ubuntu/windowsxp.html . Another nice site with all screenshots. The main steps involved were:-
1. select the free space to install ubuntu
2. Select the Automatically Partition the space option
3. Install grub boot loader ( this will show up the boot menu during boot to select between windows and ubuntu OS)
As the installation progressed, I was asked whether I wanted to download the latest packages from ubuntu, i did that, and my installation was being updated online. This is a cool feature.
Now, my next step would be to make my tv tuner card work. This would be dealt in a seperate post! I welcome myself to the Linux World!
Subscribe to:
Posts (Atom)