Showing posts sorted by relevance for query linux. Sort by date Show all posts
Showing posts sorted by relevance for query linux. Sort by date Show all posts

Apr 9, 2013

Demystifying UEFI, the long-overdue BIOS replacement

After more than 30 years of unerring and yet surprising supremacy, BIOS — the IBM PC’s Basic Input Output System — is taking its final bows and shuffling into the theater’s wings. Taking its place in the limelight is UEFI, a specification that begun its life as the Intel Boot Initiative way back in 1998 when BIOS’s antiquated limitations were hampering systems built with Intel’s Itanium processors. Later, the Initiative became EFI, and in 2005 Intel donated EFI to the newly-formed UEFI Forum, a consortium made up of the usual suspects: AMD, Apple, IBM, Intel, Microsoft, and so on.


UEFI, or Unified Extensible Firmware Interface, is a complete re-imagining of a computer boot environment, and as such it has almost no similarities to the PC BIOS that it replaces. While BIOS is fundamentally a solid piece of firmware, UEFI is a programmable software interface that sits on top a computer’s hardware and firmware (and indeed UEFI can and does sit on top of BIOS). Rather than all of the boot code being stored in the motherboard’s BIOS, UEFI sits in the/EFI/ directory in some non-volatile memory; either in NAND on the motherboard, on your hard drive, or on a network share (more on that later).

UEFI stackAs a result, UEFI almost resembles a light-weight operating system. A computer boots into UEFI, an arbitrary set of actions are carried out, and then it triggers the loading of an operating system. Further reinforcing its OSness, the UEFI spec defines boot and runtime services, protocols for communication between services, device drivers (UEFI is designed to work across all platforms), extensions, and even an EFI shell, where you can run EFI applications. On top of all this is the boot loader, which executes an operating system’s boot loader.

UEFI, being a pseudo-operating system, can access all of the hardware on the computer — you can surf the internet from the UEFI interface, or backup your hard drives — and it even has a full, mouse-driven GUI (below right). The fact that all of this boot data is stored on NAND flash or on a hard drive means that there’s a lot more space for things like language localization, boot-time diagnostics (begone meaningless POST beeps!), utilities (backup, restore, malware scanners), and so on.

As a corollary, the fact that UEFI is entirely software-based is what makes it unified. So far UEFI has been used by almost every combination of 32- and 64-bit ARM, Intel, and AMD chips, and in each case the boot code just had to be compiled for the target platform. Every major desktop (OS X, Windows) and server OS (Linux) supports UEFI boot today — and Windows 8, when it rolls out, will have features that only work with UEFI (though it will still run on conventional, BIOS-booted computers).

Asus EFI BIOSUnderneath this crazy, extensible, software-driven interface, UEFI also specifies a few standard features that must be implemented. Windows 8′s ability to detect rootkit and malware infections (and rogue Linux installations), for example, relies on UEFI’s secure boot functionality. Low-level cryptography, network authentication, universal graphics drivers, and more, are all provided as standard. Update: Microsoft now has an excellent article about UEFI, Windows 8, and secure boot (Linux will be able to run just fine!)

Finally, it’s worth noting that UEFI is still incredibly young, and very few operating systems actually take advantage of any of the features listed above. Linux certainly supports UEFI, but no Linux distro really utilizes it. Mac OS X makes slightly better use of UEFI with the Bootcamp boot manager. Windows 8, when it launches in 2012, will probably be the first major OS to take extensive advantage of UEFI, with Restore, Refresh, secure boot, and possibly more.

Read more about UEFI on Wikipedia, UEFI Forum website, or How-To Geek

If you know how to pronounce UEFI, do let us know in the comments. We want to pronounce it “you-eff-eye,” but it could also be “you-fee,” or even “oo-fee”…

Jul 22, 2019

Getting the PPP Username and Password for CenturyLink Zyxel C1000Z Modem

My first DSL modem in 1999 required Telnetting in via serial port to USB. I had to call a network technician at Qwest, and followed by typing in what seemed like arcane commands. I had no idea what I was doing. Things have changed for the better, but most DSL modems still have the ability to log into them directly through command line interfaces. The C1000Z runs BusyBox Linux which comes loaded with your usual base Linux utilities, so if you can wield Bash, you can hack your modem.

Grabbing your PPP username

I was looking to enabled the Transparent Bridge mode for my new Netgear R6050 after a friend managed to break the internal antenna on my Zyxel C1000Z, I wasn’t home so I don’t know the physics involved. Rather than pay $99 to CenturyLink for a new modem/router I decided to buy a new WAP/Router.

Having a little network administration under my belt, I figured I could grab the PPP Password.

The following guide was indispensable and got me 95% of the way there so I suggest checking it out first and/or following it along with my more “For Dummies” guide:

How to Find Your CenturyLink PPP Password on a Zyxel C1000Z Modem

You’ll want a basic understanding of SSH and/or Telnet. OS X regardless of version come with SSH and Telnet as does (almost) every flavor of Linux. Windows users will need Putty.

You’ll want a basic understanding of SSH and/or Telnet. OS X regardless of version come with SSH and Telnet as does (almost) every flavor of Linux. Windows users will need Putty.

Step 1:

First you’ll need to enable telnet in your Router, and you’ll need PPPoe enabled (Under WAN settings), these can easily be done through the Modem’s GUI


Step 2:

Fire up your terminal (Windows users will have to use Putty, and translate the instruction) and type:

telnet YOUR-IP-ADRRESS

In this example, my router’s IP address is 192.168.0.1, this is the default address so I would type:
telnet 192.168.0.1
It make take a moment for the router to respond, once it does, respond something like “BCM963268 Broadband Router” and it should ask for your username, type in the username you entered hit return and it should then ask for your password, enter the password you typed in, hit return.

Step 3:

Using the terminal we can call all the active tasks running on the modem, to do so type:
ps
Geek stuff: Users can use sh to access the BusyBox linux Bash shell and run task monitoring software like top. If you’re feeling adventurous, type sh and poke around using commands like ls and top. You can grab the process ID using top just like we do in step 4.

Step 4:

You should see a long list of responses, that read:
PID USER       VSZ STAT COMMAND  
1 admin     1556 S    init  
2 admin        0 SW<  [kthreadd]    3
 admin        0 SW<  [migration/0]  
4 admin        0 SW   [sirq-high/0]
and so on... We’re only interested in one entry, the one that’s running the pppd (or ppp*) command. it’ll probably be at the bottom. It should read something like:
3494 admin     1808 S    pppd -c ppp0.1 -i ptm0.1 -u myusername@qwest.net -p **
The myusername@qwest.net is your username.

Step 5:
cat proc/3494/cmdline 

Next you’ll need to analyze the process ID further, take special note of the preceding number, in this example its 3494. Type in the console:
pppd-cppp0.1-iptm0.1-umyusername@qwest.net-pjlFrVNtRMtU=-f0-D0-n1-L0-X120 >
The password portion of this is encoded, the tricky part here is identifying it. We know the that this is a concatenated line by gauging from the previous line. The password portion should be between -p  and -.  In this example, the encoded password is:
jlFrVNtRMtU=
Step 6:

This password is encoded in base64, thanks to the leg work Make a new tab or new terminal window, and type:
echo "jlFrVNtRMtU=" | base64 --decode
It should spit back something like:

ac7gkDnUmac-pro:~ user$

The ac7gkDnU will be your PPP password. Congrats! You’re now ready to enable transparent bridge mode on your router.

Article posted by blog@greggant.com

Dec 12, 2014

ATA-over-Ethernet vs iSCSI

Every so often someone voices interest in ATAoE support for Solaris or tries to engage in an ATAoE versus iSCSI discussion. There isn't much out there in the way of information on the topic so I'll add some to the pot...

If you look just at the names of these two technologies you can easily start to equate them in your mind and start a running mental dialog reguarding which is better. But, most folks make a very common mistake.. ATA-over-Ethernet is exactly that, over ethernet. Whereas iSCSI is Internet SCSI, or as some people prefer to think SCSI over IP. So we've got two differentiators just given the names of these technologies alone: ATA vs SCSI command set, and Ethernet vs IP stack. The interesting thing is the latter discussion.

There is a natural give and take here. The advantage of ATAoE is that you don't have the overhead of translating ATA to SCSI then back to ATA if your using ATA drives, so there is a performance pickup there. Furthermore, because we don't have the girth fo the TCP/IP stack underneight we don't burden the system with all that processing, which adds even more performance. In this sense, ATAoE strips away all the stuff that gets in the way of fast storage over ethernet. But, naturally, there is a catch. You can't route ethernet, thats what TCP/IP is for. That means that with ATAoE your going to be building very small and localized storage networks on a single segment. Think of a boot server which operates without TCP/IP, you've got to have one per subnet so that it see's the requests.

iSCSI on the otherhand might be burdened by the bulk of the TCP/IP stack, however it has the ability to span the internet because of it. You can have an iSCSI target (server) in New York and an iSCSI initiator (client) in London connected across a VPN and its not a problem. Plus, iSCSI is an open and accepted standard. ATAoE on the otherhand is open but it was created and developed by Coraid who also happens to be the only supplier of ATAoE enclosures. That may change, but we'll see how well it catches on.

ATAoE promises to be smaller and faster than the industry standard iSCSI, and it is, but unless you are using a very local application your going to be in trouble. Not to mention the lack of enclosure and driver support for non-Linux systems.

The question then becomes: Should OpenSolaris support ATAoE? Personally, I don't think we should ever be against the idea of anything new, if someone wants to do it, we should all get behind it. But looking at Solaris I doubt the idea would stick. First and foremost Solaris is an OS that adheres to the standards and plays by the rules, even when its painful. Linux doesn't always play by those rules and often it gains from breaking them. Linux is a great experimental platform, no doubt, but I just don't think the ideals of ATAoE mesh well with the goals of Solaris. Furthermore, ATAoE doesn't offer the level of scalability, flexablilty, and managability that we get with iSCSI. The performance hit of TCP/IP is definately a downside, but the advantages it brings to the table far out weight the downsides I think.

Here are some links to help you explore the subject more on your own:

Apr 9, 2013

Linux developers working on a way to bypass Windows 8 secure boot

Microsoft has been getting a lot of heat in recent months for its decision to require UEFI secure boot on upcoming Windows 8 PCs. UEFI secure boot is a method of locking down a system’s bootloader so that no unsigned code can be installed. That means pre-boot malware is dead in the water, but it also makes it difficult to install alternative operating systems like Linux.


Well, the Linux community is banding together and might finally have the tools needed to bypass that restriction.

Microsoft originally wanted to require hardware vendors to make secure boot mandatory with no option to disable. Eventually Redmond compromised; x86 systems will have a toggle hidden deep down in the UEFI settings, but WinRT device will have secure boot locked in. A few weeks ago, RedHat announced that it was going to be paying for a Verisign bootloader through the Microsoft Sysdev portal, allowing its software to be installed with secure boot.

Not all Linux devs are content to take this route, so James Bottomley, chair at Linux Foundation’s Technical Advisory Board, is taking action. Bottomley has publicly posted a new Intel Tianocore UEFI boot image to help developers out. Tianocore is Intel’s open source UEFI image, which has Microsoft’s Authenticode built in.

Developers will now be able to run a virtual boot platform with secure boot for testing. This is important because it’s currently very difficult to get your hands on UEFI secure boot hardware in advance of Windows 8. It is hoped that this image will help the community figure out a secure boot bypass.

While it’s really cool to see the community rally around this cause, a bypass to secure boot will negate any security benefits it might have had. Still, if the good guys don’t find it, someone else will.

Jan 3, 2015

UNetbootin : Universal Netboot Installer

UNetbootin is a Universal Netboot Installer that allows you to create bootable Live USB drives for Windows and other Linux distributions without burning a CD.


It runs on both Windows and Linux. You can either let UNetbootin download one of the many distributions supported out-of-the-box for you, or supply your own Windows or Linux .iso file if you’ve already downloaded one or your preferred distribution isn’t on the list.

You can download UNetbootin from here.

Advantages of bootable USB sticks with UNetbootin.

There are some clear advantages of bootable USB flash drives over CDs/DVDs: Your notebook doesn’t have an optical drive or it’s broken… You can carry around a fully functional operating system on your keyring… Data can be stored on the flash drive while on Live CDs you can’t…

Most likely you will want to install a Linux distribution, or a Windows distribution you have the .iso file. A bootable USB drive can also come incredibly handy when you have to backup your broken computer, and there are a lot more scenarios. UNetbootin has built-in support for numerous distributions, including Ubuntu, Fedora, openSUSE, PCLinuxOS, Linux Mint, ArchLinux, Debian, CentOS, Frugalware, FreeBSD, NetBSD, and Parted Magic, though many other distributions can also be installed via an existing ISO or floppy/hard disk image file.

Visit UNetbootin at http://unetbootin.net/

Mar 13, 2013

How-To Install pfSense on ALIX2 Series

So, what is pfSense, and more importantly what the heck is an alix2c1? pfSense is a complete purpose-built firewall software that can be installed on a PC, as well as embedded platforms such as PC Engines ALIX boards. The focus of this recipe is to build an embedded firewall appliance.


Why would I do this when a cheap firewall can be purchased off of the shelf? Simple, the cheaper firewalls don’t come with very many features (VPN built in, RRD network graphs, 3rd LAN port for “orange” DMZ setups, etc.). To get these features in a commercial firewall appliance means spending in the neighborhood of $300 as well as being locked into a certain vendor for VPN, etc. Making my own costs about $150 and give me a lot of flexibility.

So, this is a story of how I spent the past two nights sitting cross-legged on the floor with a laptop connected to a tiny motherboard that would become a firewall/router.


The instructions that I found on both the pfSense website and the m0n0wall website were for installing on the End-of-life WRAP product, but I figured it had to be similar, right? Well, sort of :-) After following the instructions on the pfSense website and then the m0n0wall website, things started to go downhill. The device would boot, pfsense would go through its loading process, but then just die on the loading of the DHCP server.

Never got an IP address on any of the NICs – but each NIC module got its link light when I plugged in the LAN cable so that was a sign that at least the NIC modules were working on some level. So I figured this is a BETA version of pfSense, so maybe I need to install m0n0wall. m0n0wall would boot up just fine and even made it to its menu, but I could never get the NICs to come up. None of them. Could I have a bit o’ bad hardware? So I hit the web again and soon found out from the forums over at pfsense that a BIOS upgrade might be in order.


The following steps are how I finally got this project off of the ground. I used Linux to do my bidding, but you can certainly do this from a Windows box (some of the steps are different, though…)

Hardware you will need (I purchased from Netgate and it came fast and furious, but there is a list of vendors on PC Engines’ website):
  • ALIX2c1 or another embedded platform from PC Engines.
  • Aluminum enclosure (optional if you’re going to install your ALIX in another enclosure).
  • Power supply to power the ALIX (optional if you’re going to use Power over Ethernet, or PoE).
  • A Compact Flash (CF) card (has to be at least 128MB according to the pfSense website).
  • A Compact Flash reader connected to your PC. You’ll need this to write the image to the CF card.
  • A computer with a serial port, preferably one with a true serial port and not a USB-to-serial adapter. It might work just fine with the USB/serial but if it doesn’t then you’ll have one more thing to troubleshoot.
  • A DB9 null-modem cable with a DB9 female on both ends (or some gender changers to get you there)
When you get the hardware – DO NOT mount the ALIX board into the enclosure until you get it working. You might be removing/inserting the CF card a lot and the enclosure doesn’t give you any room to remove the CF card. OK, now that you’ve got all of the necessary hardware, here is the recipe for cooking up your own firewall:

1.] Download pfSense for embedded platform. The most current version at the time of this writing was 1.2-RC4, which I found to be stable for my needs as I had been using the PC based version for several weeks.
2.] Unzip the downloaded file (the version I downloaded isn’t a tar archive so we only need to use gunzip):

gunzip pfSense-1.2-RC4-Embedded.img.gz

This will expand the file, leaving a file called pfSense-1.2-RC4-Embedded.img.

3.] Now insert the CF card into the card reader on your PC. To find out the resource Linux is using to access the drive, type the following at the command line:

dmesg

and look at the last several lines which might look something like:

[ 7377.984000] sd 2:0:0:0: [sdb] 700560 512-byte hardware sectors (359 MB)
[ 7377.984000] sd 2:0:0:0: [sdb] Write Protect is off
[ 7377.984000] sd 2:0:0:0: [sdb] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA
[ 7377.984000] sd 2:0:0:0: [sdb] 700560 512-byte hardware sectors (359 MB)
[ 7377.984000] sd 2:0:0:0: [sdb] Write Protect is off
[ 7377.984000] sd 2:0:0:0: [sdb] Mode Sense: 00 3a 00 00
[ 7377.984000] sd 2:0:0:0: [sdb] Write cache: disabled, read cache: enabled, doesn't support DPO or FUA
[ 7377.984000] sdb: sdb1
[ 7377.984000] sd 2:0:0:0: [sdb] Mode Sense: 00 3a 00 00
[ 7378.020000] sd 2:0:0:0: [sdb] Attached SCSI removable dis
[ 7378.020000] sd 2:0:0:0: Attached scsi generic sg2 type 0

From this output we can see that Linux is accessing the CF card as /dev/sdb.

4.] Next we will use the Linux utility “dd” to write the image to the CF card

dd if=pfSense-1.2-RC4-Embedded.img of=/dev/sdb

where “if=” is the input file (the pfSense image) and “of=” is for the output file. In our case we are writing the output to the file that is the CF card (remember, in Linux almost everything is a file). While the process is running, unfortunately you won’t get any output or indication of what it is doing. However, when it is finished you should get some output such as:

239144+0 records in
239144+0 records out
122441728 bytes (122 MB) copied, 187.177 seconds, 654 kB/s

5.] Unmount the CF card from your computer. If you’ve got a nifty neato X session going, and a handy dandy automounter, then chances are good that you can just right-click on the respective desktop icon for your CF card and select “Unmount Volume”. If not, hit the command line and do a:

mount

which should show where the drive is mounted:

/dev/sdb1 on /media/disk type vfat (rw,nosuid,nodev,shortname=mixed,uid=1000,utf8,umask=077,usefree)

6.] Now that we know where the drive is mounted, let’s unmount it:

sudo umount /media/disk

7.] Remove the CF card from your computer and insert it into the CF adapter on the ALIX board.
8.] Connect one end of the null-modem cable to your computer’s serial port and the other end to the serial port on the ALIX.
9.] Fire up your favorite terminal emulation software such as minicom (or Hyperterminal on Windows) and use the following settings:
  • Baud rate: 38,400
  • Data: 8 bit
  • Parity: None
  • Stop: 1 bit
  • Flow control: None
  • Terminal: ANSI
10.] Now apply power to the ALIX. If you are connected correctly, you should start to see the ALIX BIOS text.
11.] While the BIOS is going through the memory test press the “s” key to enter the BIOS setup.
12.] If have successfully entered the BIOS setup, you should see the text with some different options. Do the following:
  • Press “9″ to set the baud rate at 9600
  • Press “q” to quit the BIOS setup
  • Press “y” to save the settings to flash
13.] If you start seeing gibberish ASCI characters instead of text, then you need to set your terminal emulation software to 9600 baud instead of the 38,400 we set it at earlier.
14.] Now reboot the ALIX by power cycling the unit (unplug the power, plug it back in).
15.] With the terminal set to 9600 baud, we should see the boot-up process and if all is well it should look akin to a Free-BSD boot.
16.] If all goes well and pfSense discovers your hardware, then you are good to go. To get started, you need to:
  • Assign the interfaces
  • Give the LAN interface an IP address that works for your internal network (i.e. 192.168.1.1)
17.] Once you plugged the LAN interface into your network, then fire up your web browser and surf on over to the IP address you gave for the LAN interface (http://192.168.1.1)

Unfortunately for me pfSense did not properly detect the NIC modules. After some digging around on the ‘net it looked like a BIOS upgrade would do the trick as the version on my ALIX board was 0.98b. The latest and greatest was 0.99. So here is how to flash upgrade the BIOS:

1.] Download the FreeDOS bootable image from PC Engine’s website.
2.] Insert another CF card into your computer. If you only have the one CF card, then you’ll have to re-do the previous  instructions to get the pfSense image back on the card after you’re done updating the BIOS.
3.] Unzip the image from the download if necessary (it was a Zip file when I downloaded it)

unzip freedos3.zip Archive: freedos3.zip
inflating: freedos_alixupdate_0.99.img

4.] Now write this image to the CF card:

dd if=freedos_alixupdate_0.99.img of=/dev/sdb

5.] Unmount your CF card from the computer, and insert it into the ALIX.
6.] Power on the ALIX and press “s” to enter the BIOS setup.
7.] Change the drive configuration to LBA by pressing “L”.
8.] Now press “q” to exit, and “y” to save your changes to flash.
9.] The system should boot the FreeDOS image and automatically run the BIOS flash utility (sb.com) to reprogram the flash.
10.] Once it is done, power off the ALIX and reinsert your pfSense imaged CF card. If you only had one card you’ll now have to go back and rewrite the pfSense image to the card.

Oct 31, 2017

For TP Routing: Modify firmware size, unlock U-boot partitions, add Fullflash partitions

TP-Link TL-WR703N factory default has only 4MB of flash and 32MB RAM, if you want to play with the OP is simply powerless, so I changed the 8M Flash and 64M RAM.

First, modify the firmware size

OpenWrt in the compile time will be based on each model profile to generate the firmware, if the generated firmware integrated too much software and more than 4MB it will be error, you will find that the firmware can not be generated.

Specific symptoms See my last Post start compiling and compiling openwrt.

Compilation passed, but did not find the firmware in ./openwrt/bin/ , turned up the compile output prompt, and found a similar

  1. [mktplinkfw] kernel length aligned to 914864
  2. [mktplinkfw] *** error: images are too big
  3. make[3]: [install] Error 255 (ignored)
Here's how to modify the default firmware size of the TP series router


Actually very simple, find the target/linux/ar71xx/image/Makefile, search for the model you want to modify, such as 703N, find the relevant 703N parameters

Then change the tplink-4mlzma to tplink-8mlzma on the line, and then

  1. define Device/tl-wr703n-v1
  2. $(Device/tplink-8mlzma)
  3. BOARDNAME := TL-WR703N
  4. DEVICE_PROFILE := TLWR703
  5. TPLINK_HWID := 0x07030101
  6. CONSOLE := ttyATH0,115200
  7. endef

After you save the exit, you can compile the 8M firmware for tl-wr703n.

Note: Some outdated tutorials on the web say that you also need to modify tools/firmware-utils/src/mktplinkfw.c after I test, the latest source code inside this file is not about 703N and other router parameters, you do not need to modify.


II. Unlocking u-boot Partitions

The horse has the hoof, the person has the slip, does not have to die u-boot how dares the confidence to play the OP? Here we will talk about U-boot, U-boot is the embedded Linux system boot, equivalent to the computer BIOS. The traditional u-boot is only responsible for booting the firmware, once the firmware does not start, then the entire router is brick, there is a TTL is also easy, no one can only on the programmer. But there is not dead u-boot, in fact, with the Web Brush Machine interface U-boot, MA Ma no longer have to worry about my machine into bricks, completely without the demolition machine can save bricks. Want to know their own Google, recommended Enshan hackpascal breed, and brush into U-boot tutorial , I do not discuss how to use u-boot, mainly brush into the u-boot will encounter

  1. could not open mtd device u-boot ,can't open device for writing

This is because OpenWrt is locked by default, unless it is unlocked when the firmware is compiled. U-boot

Locate /target/linux/ar71xx/files/drivers/mtd/tplinkpart.c, search for U-boot, find

  1. parts[0].name = "u-boot";
  2. parts[0].offset = 0;
  3. parts[0].size = offset;
  4. parts[0].mask_flags = MTD_WRITEABLE;
Then remove parts[0].mask_flags = MTD_WRITEABLE; This line will be recompiled.

WARM hint: TP series Brush finished breed must change the MAC address for a valid value, otherwise wireless will hang!

III. Add Fullflash partitions

Now the bricks can be saved, but save back the data are all gone, but also start the configuration, want to think all big ah, the good habit of nurturance is to fall roar! Add Fulllash partitions so that you can back up the entire programmer firmware with the DD command, and there is no such thing as an art loss.
Same as /target/linux/ar71xx/files/drivers/mtd/tplinkpart.c to add the following code

  1. parts[5].name = "fullflash";
  2. parts[5].offset = 0;
  3. parts[5].size = master->size;

As shown in the figure


After recompiling and brushing, you can see the Fullflash partition.

View partitions with cat /proc/mtd

Effect as shown


You can then simply back up the programmer firmware through the dd if=/dev/mtd6 of=/tmp/fullflash.bin !

5aimiku

Jan 20, 2015

Pear OS Linux Panther v3.0 Mac OS X Look Alike

Pear OS Linux has announced the release of Pear OS 3.0, an Ubuntu-based desktop distribution with a Mac OS X-like look and feel (as well as slogan): "Pear OS 3.0 'Panther' available. 


From the desktop you see when you start up your PC to the applications you use every day, everything is designed to be simple and intuitive. Of course, making amazing things simple requires some seriously advanced technologies, and Pear OS is loaded with them. 

Working and playing on a PC is all about applications, so Pear OS makes it simple to find and open those applications fast. The Dock is a handy place on your desktop for storing and launching your favorite applications, and it makes switching between them a breeze."

Download links:
Pear-OS-Linux-Panther-v3.0-x32.iso (895MB, MD5)
Pear-OS-Linux-Panther-v3.0-x64.iso (964MB,MD5)

The release announcement doesn't offer any details, but the product page has some screenshots alongside much marketing talk.

Apr 9, 2013

Microsoft Is Using Linux Kernel In Windows 8 ?

ZDNet's Mary Jo today wrote an interesting article about why Microsoft is keeping a tight lid on the Windows 8 test builds. Earlier such builds were leaked by OEMs or employees to give users an idea of what was coming. This time Microsoft has taken extra measures to ensure this version of Windows doesn't fall in wrong hands.


The company specifically wrote:
Unauthorized use or disclosure in any manner may result in disciplinary action up to and including termination of employment (in the case of employees), termination of an assignment or contract (in the case of contingent staff), and potential civil and criminal liability.
However, there are some leaked copies available on the Chinese torrent site taxfreewindows.com. We were able to grab a copy of Windows 8. The file was 20GB large which expanded to 50GB after installations. It is actually a 3boot OS which can run on old PCs, Linux PCs and Macs. The OS is expected to run on Nokia tablets as well.

This build also disclosed why Microsoft was so tight-lipped about Windows 8. Microsoft is using a fork of Linux kernel in this version. The OS was fast and detected all our hardware -- a clear indication of Linux' presence. We found further evidences when we looked at the changelog. There were patches by Greg-KH in the kernel log, which verified that Microsoft is in fact forking the upstream kernel tree. There was a message by Linus Torvalds which Microsoft engineers forgot to remove:
Anything else? I'm sure I've forgotten something really exciting. But on the whole I think this should be one of those "solid, boring progress" releases.

Knock wood. I like boring.

~ Linus
When asked about this alleged copying or piracy, Microsoft Strategist Miguel De Icaza wrote on his blog:
There are hundreds of possible answers as to why someone would copy someone else's work. For instance, perhaps I am a better implementor than an inventor.

That being said, it can be argued that implementing a system that is compatible with something else has a lot of value. You get to reuse knowledge, you get to reuse experience and you get to focus your efforts in new directions, instead of redoing things from scratch.

Source

Oct 27, 2013

How-To Create Linux BackTrack USB Flash Drive on Windows

How to Make a BackTrack Linux Flash Drive using Windows. BackTrack is a Live Linux distribution based on SLAX that is focused purely on penetration testing. Distributed by remote-exploit.org, BackTrack is the successor to Auditor. It comes prepackaged with security tools including network analyzers, password crackers, wireless tools and fuzzers. Although originally designed to Boot from a CD or DVD, BackTrack contains USB installation scripts that make portable installation to a USB device a snap. In the following tutorial, we cover the process of installing BackTrack to a USB flash drive from within a working Windows environment.


Distribution Home Page: http://www.backtrack-linux.org

Minimum Flash Drive Capacity: 2GB+

Persistent Feature: Yes (Backtrack 4)

USB Ultimate Boot CD (UBCD) prerequisites:
  • Universal USB Installer (does the USB conversion)
  • BackTrack ISO
  • 2GB+ USB flash drive (fat32 formatted)
  • A windows host PC to perform the build
How to install BackTrack to a USB device from Windows:
  1. Download and launch our Universal USB Installer, select Backtrack and follow the onscreen instructions
  2. Reboot and set your BIOS or Boot Menu to Boot from the USB device and proceed to boot
Note: Once Backtrack has loaded, you must type startx at the prompt, to start the graphical X environment.

If all went well, you should now be running from your very own Portable Backtrack on USB!

Jun 5, 2013

BackTrack 5 UNetbootin USB Installer

Required Tools and Preparation
  • USB Thumbdrive (Minimum USB Drive capacity 4 GB)
  • Backtrack 5 ISO file, md5sum verified.
  • UNetbootin - A tool to transfer an iso image to a USB drive

UNetbootin allows you to create bootable Live USB drives for Ubuntu, Fedora, and other Linux distributions without burning a CD. It runs on both Windows and Linux. You can either let UNetbootin download one of the many distributions supported out-of-the-box for you, or supply your own Linux .iso file if you've already downloaded one or your preferred distribution isn't on the list. This method of getting a live install to a USB drive is the simplest available using Unetbootin. Note that we will format the USB drive and erase its contents.

Note: you can now save changes made to the system across reboots (persistence).
  • Download the latest BackTrack iso from the BackTrack Download page.
  • Download UNetbootin, latest version (Linux, Windows, OSX)
  • Plug in your USB Drive
  • Format the USB drive to FAT32
  • Start Unetbootin and select the "diskimage" option, choose a BackTrack 5 ISO file.
  • Select the amount of space to use for persistence in MB
  • Select your USB drive and click “OK” for creating a bootable BackTrack USB drive.
  • Once rebooted, remember that the default username and password are root / toor.

Nov 3, 2013

Android OS For Personal Computer

Google's answer to Symbian and iOS is Android. It is a Linux avatar, specifically designed for mobile phones and other portable devices like tablet computers. The Linux pedigree means that portability is inherent in this operating system. Very few people in Android fan communities know that Android OS for a PC has been a reality since August, 2009. The developer team of Android launched a live CD for computer users, where they could test the prowess of this new operating system. Now customized distributions of the latest Android operating systems are made available online, for free. The Android-x86 website, hosted by a community of developers provides latest customized builds of Android for computers.


The concept of a live CD lets you test an operating system, in spite of having a fully functional installed operating system on your PC. So people who are thinking of switching to a new operating system, can check it out, before they decide to completely migrate over to it.

If you feel like testing the Android OS, before buying an Android phone, getting a live CD or live USB version and running it, is the best way to test it. Making this live CD or live USB version available was one way of increasing the credibility of the operating system among the masses. That's exactly what the folks at Android-x86 have done for you. If you like the live CD version, you may install it on your netbook, laptop or computer as a standalone or alternative operating system. Let us see how you could go about it.

How to Get Android OS For Your PC?

So how and where can you get Android for PC setup? Like all developers of the Linux open source community, the Android team has made the live CD version, as well as the live USB version available on its official website. Visit the Android-x86.org web page or any other mirror sites to download the setup files. Various images of builds based on the succeeding Android versions (Cupcake, Donut, Eclair, Froyo, Gingerbread, Honeycomb and Ice Cream Sandwich) are available. Some builds are exclusively designed for specific brands of netbook and laptop computers like Asus, Lenovo and Dell. Each ISO file can be used to create a live CD, which also provides the option of hard drive installation.

Know that these versions work only on x86 systems. Read all the instructions regarding download and installation on the website. A glance through the FAQ section will also be very helpful. Let me walk you through the installation.

How to Install and Run Android OS on Your PC?

Once you get the setup files on your computer, the rest is quite easy. There are two ways in which you can run Android on a computer, without installing it on the hard drive. Either you could burn the live CD ISO image on a CD and boot from your optical drive or USB drive or alternatively, you could run it using VMware or VirtualBox. If you like Android, you may go ahead and install it on your hard drive. Here are step-by-step instructions for installing Android x86 versions.

Step 1: Download Latest Build From Android-X86 website
The latest version available, at the time of writing is Android-x86 4.0-RC1. You can get the whole ISO image directly, by downloading it from the main website or any one of the mirror sites. These builds support fat32, ntfs, ext3 and ext2 file systems. You also have the option of running Android on VMware or VirtualBox.

Step 2: Burn Image on a CD or Created Bootable USB
Firstly, you must prepare the Android ISO image for installation. Burn it on a blank CD or install it on a USB using UNetbootin. Now you are ready for installation.

Step 3: Run Installation
To run the installation, simply plug in the USB or put the installation CD into the optical drive and boot up the PC. The installer will provide you with a couple of options. Choose 'Run Andorid Without Installation' (to boot from CD) or 'Install Android to Hard disk' (for a permanent installation). Android can coexist on the hard drive with other operating systems in multiboot mode.

Step 4: Choose Partition and File System Format
On choosing hard drive installation, you will be asked to choose a hard drive partition for installation. You may choose one of the available partitions or create a new one. You may also choose to install it on a removable USB drive. To go for the third option, choose the 'Search Devices' option. Then you must choose whether formatting is to be implemented on the partitions and if yes, choose the type of file system.

Step 5: Install Grub
Lastly, you will be asked whether Grub (the boot chain loader) should be installed. Get it installed if you are going for a multiboot setup. With that done, installation will begin immediately. Within a few minutes (depending on the processing speed of your PC), the installation will be finished. Within a few minutes, you will be greeted with the stunning Android interface, which you can begin using!

As you will realize, the power of Android lies in its multitasking prowess and light use of computing resources. Through Android Linux is reborn as a new mobile incarnation. The future is ANDROID for sure.

Aug 5, 2018

How To Backup Fiberhome AN5506-04-FA MTD

First and foremost before doing the backup of the ONU/ONT Fiberhome AN5506-04-FA make sure you are the owner of the device, secondly you responsible for the damage you have done, doing this is a risky and you might be able to brick the equipment if mishandled incorrectly. As a precaution, preferably you have a spare ONU/ONT on your hand in the case of disaster you might not disrupt your internet connection.

How do we backup the ONU/ONT Fiberhome AN5506-04-FA mtd aka firmware?


There are two available terminal that we can gain access on AN5506-04-FA one is via Telnet terminal and the other is via Serial communication port. These two accessible console is also applicable to the other Fiberhome ONU/ONT AN5506-XX-XXX series if you want to explore your own device to prevent your Internet Fiber ONU device being remotely updated upon updated its firmware for you not to have the vast features configuration and all the other setting being hidden by your greedy ISP. To begin with, lets check if the port of Telnet is open we can use NMAP to verify it or other similar port scanner that may give us same result. If the Telnet port is close then we do not have other choice but to open the top cover of the ONU/ONT AN5506-04-FA because Serial Console header is residing on the surface of the board.



What we need for serial communication for us to be able to communicate with AN5506-04-FA Serial Console is a serial com port, or any USB-to-Serial converter device. Then a HyperTerminal, you can also use Putty,you might want to be comfy Hercules will do or any other application alike will serve for data communication between two digital devices. Once you are plugged in then the console terminal is ready. Communication baud rate is 115200 8n1, if everything is properly hooked up, once we fire up the power ON you should see the U-Boot message just started like this.

U-Boot 2010.03-svn671412 (May 27 2017 - 09:37:39)

DRAM:  128 MB
Boot From NAND Flash
CHIP ID = 51152100
NAND:  Special Nand id table Version 1.23
Nand ID: 0xC2 0xF1 0x80 0x95 0x02 0x00 0x00 0x00
Nand(Hardware): Block:128KB Page:2KB Chip:128MB*1 OOB:64B ECC:4Bytes
128 MiB
env0 ok ~~~~~~~~~~~
In:    serial
Out:   serial
Err:   serial
hi_lsw_init
hi_lsw_init_t
MEM_MODE = MEM!
tmp_cmd a =kk=112 mem=240M console=ttyAMA1,115200 root=/dev/mtdblock5 rootfstype=jffs2 mtdparts=hinand:128k(startcode),1M(u-bootA),1M(u-bootB),1M(envA),1M(envB),18M(kernel_rootfsA),18M(app_binA),20M(app_exA),18M(kernel_rootfsB),18M(app_binB),20M(app_exB),12160k(cfg)
kernel_rootfs_mtd_offset = 0x420000
Hit enter to stop autoboot:  2

This is just the first boot, let the system to goes on to the second boot until you see the message like this and then it will tell you to Press Ctrl+C to stop auto setup in 3 seconds, from 2sec to 0sec you must be quick.

CFE adapter module install successfully ...!

CFE hw_adpter_l3 module install successfully ...!

CFE module install successfully ...!
dapter multicast module install successfully, version: Jul 12 2017 10:28:18

 CFE_FH_MARK module install successfully ...!
initialize.sh...

Press Ctrl + C to stop auto setup 0

You are now in the root directory

~ #
Once you type the following Linux command like this cat /proc/mtd then you will see now the list of all MTDs.

~ # cat /proc/mtd
dev:    size   erasesize  name
mtd0: 00020000 00020000 "startcode"
mtd1: 00100000 00020000 "u-bootA"
mtd2: 00100000 00020000 "u-bootB"
mtd3: 00100000 00020000 "envA"
mtd4: 00100000 00020000 "envB"
mtd5: 01200000 00020000 "kernel_rootfsA"
mtd6: 01200000 00020000 "app_binA"
mtd7: 01400000 00020000 "app_exA"
mtd8: 01200000 00020000 "kernel_rootfsB"
mtd9: 01200000 00020000 "app_binB"
mtd10: 01400000 00020000 "app_exB"
mtd11: 00be0000 00020000 "cfg"

Your USB pen drive or Flash drive must be formatted in FAT32 by default in any Microsoft Windows OS. After formatting it eject then plugged it onto the USB port of the AN5506-04-FA device. It will then pop you a message like this.


usb 1-2: new high speed USB device using hiusb-ehci and address 3
scsi1 : usb-storage 1-2:1.0
scsi 1:0:0:0: Direct-Access     TOSHIBA  TransMemory      1.00 PQ: 0 ANSI: 4
sd 1:0:0:0: [sda] 7574304 512-byte logical blocks: (3.87 GB/3.61 GiB)
sd 1:0:0:0: Attached scsi generic sg0 type 0
sd 1:0:0:0: [sda] Write Protect is off
sd 1:0:0:0: [sda] Assuming drive cache: write through
sd 1:0:0:0: [sda] Assuming drive cache: write through
 sda: sda1
sd 1:0:0:0: [sda] Assuming drive cache: write through
sd 1:0:0:0: [sda] Attached SCSI removable disk
fat
open /dev/console successed.
usb led 0  off
usb led 1  on

Verify with the df  Linux command. It will show you the list of all devices in the system.

~ # df
Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/root                18432      9864      8568  54% /
tmpfs                   119696         4    119692   0% /dev
none                    524288         0    524288   0% /tmp
none                    524288         4    524284   0% /var
/dev/mtdblock11          12160       856     11304   7% /fhcfg
/dev/mtdblock6           18432      5796     12636  31% /fh/bin
/dev/mtdblock7           20480     16564      3916  81% /fh/extend
/dev/sda1              3777936    215868   3562068   6% /dev/shm/usb/media/sda1

The USB flash drive is detected as sda1 device (/dev/sda1) and the mount point is located at (/dev/shm/usb/media/sda1), now we have to unmount the USB pen drive device first by not unplugging from the USB port of the AN5506-04-FA. Then mount again the USB flash drive with this following Linux command.


umount /dev/sda1

mount /dev/sda1 /dev/shm/usb/media/sda1

We are ready now for backing up all the MTDs of Fiberhome ONU/ONT AN5506-04-FA, we'll make first folder on the USB drive with this command.


mkdir /dev/shm/usb/media/sda1/backup


Now use these following Linux commands for back up the list of all the AN5506-04-FA MTDs.


~ # dd if=/dev/mtd0 of=/dev/shm/usb/media/sda1/backup/startcode.bin

256+0 records in
256+0 records out
131072 bytes (128.0KB) copied, 0.018642 seconds, 6.7MB/s


~ # dd if=/dev/mtd1 of=/dev/shm/usb/media/sda1/backup/u-bootA.bin

2048+0 records in
2048+0 records out
1048576 bytes (1.0MB) copied, 0.147251 seconds, 6.8MB/s

~ # dd if=/dev/mtd2 of=/dev/shm/usb/media/sda1/backup/u-bootB.bin

2048+0 records in
2048+0 records out
1048576 bytes (1.0MB) copied, 0.146912 seconds, 6.8MB/s

~ # dd if=/dev/mtd3 of=/dev/shm/usb/media/sda1/backup/envA.bin

2048+0 records in
2048+0 records out
1048576 bytes (1.0MB) copied, 0.147150 seconds, 6.8MB/s

~ # dd if=/dev/mtd4 of=/dev/shm/usb/media/sda1/backup/envB.bin

2048+0 records in
2048+0 records out
1048576 bytes (1.0MB) copied, 0.146535 seconds, 6.8MB/s

~ # dd if=/dev/mtd5 of=/dev/shm/usb/media/sda1/backup/kernel_rootfsA.bin

36864+0 records in
36864+0 records out
18874368 bytes (18.0MB) copied, 2.636288 seconds, 6.8MB/s

~ # dd if=/dev/mtd6 of=/dev/shm/usb/media/sda1/backup/app_binA.bin

36864+0 records in
36864+0 records out
18874368 bytes (18.0MB) copied, 3.495190 seconds, 5.1MB/s

~ # dd if=/dev/mtd7 of=/dev/shm/usb/media/sda1/backup/app_exA.bin

40960+0 records in
40960+0 records out
20971520 bytes (20.0MB) copied, 2.980738 seconds, 6.7MB/s

~ # dd if=/dev/mtd8 of=/dev/shm/usb/media/sda1/backup/kernel_rootfsB.bin

36864+0 records in
36864+0 records out
18874368 bytes (18.0MB) copied, 5.694926 seconds, 3.2MB/s

~ # dd if=/dev/mtd9 of=/dev/shm/usb/media/sda1/backup/app_binB.bin

36864+0 records in
36864+0 records out
18874368 bytes (18.0MB) copied, 2.767045 seconds, 6.5MB/s

~ # dd if=/dev/mtd10 of=/dev/shm/usb/media/sda1/backup/app_exB.bin

40960+0 records in
40960+0 records out
20971520 bytes (20.0MB) copied, 2.995862 seconds, 6.7MB/s

~ # dd if=/dev/mtd11 of=/dev/shm/usb/media/sda1/backup/cfg.bin

24320+0 records in
24320+0 records out
12451840 bytes (11.9MB) copied, 1.870105 seconds, 6.3MB/s

Now we are done, we able to backup all the MTDs of Fiberhome ONU An5506-04-FA. Soonest I post an article on How-To upload the MTD int the device incase you bricked you equipment you can repair it by you own.

Dec 12, 2014

Yet Another AoE vs. iSCSI Opinion

That’s right, folks! Yet another asshole blogger here, sharing his AoE (ATA over Ethernet) vs. iSCSI (Internet SCSI) opinion with the world!

As if there wasn’t already enough discussion surrounding AoE vs. iSCSI in mailing lists, forums and blogs, I am going to add more baseless opinion to the existing overwhelming heap of information on the subject. I’m sure this will be lost in the noise but after having implemented AoE with CORAID devices and iSCSI with an IBM (well, LSI) device and iSCSI with software targets in the past I feel I finally have something share.

This isn’t a technical analysis. I’m not dissecting the protocols nor am I suggesting implementation of either protocol for your project. What I am doing is sharing some of my experiences and observations simply because I can. Read on, brave souls.

Background

My experiences with AoE and iSCSI are limited to fairly small implementations by most standards. Multi-terabyte and mostly file serving with a little bit of database thrown in there for good measure. The reasoning behind all the AoE and iSCSI implementations I’ve setup is basically to detach storage from physical servers to achieve:
  1. Independently managed storage that can grow without pain
  2. High availability services front-end (multiple servers connecting to the same storage device(s))
There are plenty of other uses for these technologies (and other technologies that may satisfy these requirement), but that’s where I draw my experiences from. I’ve not deployed iSCSI or AoE for virtual infrastructure which does seem to be a pretty hot topic these days, so if that’s what you’re doing, your mileage will vary.

Performance

Yeah, yeah, yeah, everyone wants the performance numbers. Well, I don’t have them. You can find people comparing AoE and iSCSI performance elsewhere (even if many of the tests are flawed). Any performance numbers I may accidentally provide while typing this up in a mad frenzy are entirely subjective and circumstantial… I may not even end up providing any! Do you own testing, it’s the only way you’ll ever be sure.

The Argument For or Against

I don’t really want to be trying to convince anyone to use a certain technology here. However, I will say it: I lean towards AoE for the types of implementations I mentioned above. Why? One reason: SIMPLICITY. Remember the old KISS adage? Well, kiss me AoE because you’ve got the goods!

iSCSI has the balls to do a lot, for a lot of different situations. iSCSI is routable in layer 3 by nature. AoE is not. iSCSI has a behemoth sized load of options and settings that can be tweaked for any particular implementation needs. iSCSI has big vendor backing in both the target and the initiator markets. Need to export an iSCSI device across a WAN link? Sure, you can do it, never mind that the performance might be less than optimal but the point is it’s not terribly involved or “special” to route iSCSI over a WAN because iSCSI is designed from the get-go to run over the Internet. While AoE over a WAN has been demonstrated with GRE, it’s not inherent to the design of AoE and never will be.

So what does AoE have that iSCSI doesn’t? Simplicity and less overhead. AoE doesn’t have myriad of configuration options to get wrong, it’s really so straight forward that it’s hard to get it wrong. iSCSi is easy to get wrong. Tune your HBA firmware settings or software initiator incorrectly (and the factory defaults can easily be “wrong” for any particular implementation) and watch all hell be unleashed before your eyes. If you’ve ever looked at the firmware options provided to by QLogic in their HBAs and you’re not an iSCSI expert, you’ll know what I’m talking about.

Simplicity Example: Multipath I/O

A great example of AoE’s simplicity vs. iSCSI is when it comes to multipath I/O. Multipath I/O is defined as utilizing multiple paths to the same device/LUN/whatever to gain performance and/or redundancy. This is generally implemented with multiple HBAs or NICs on the initiator side and multiple target interfaces on the target side.

With iSCSI, every path to the same device provides the operating system with a separate device. In Linux, that’ll be /dev/sdd, /dev/sde, /dev/sdf, etc. A software layer (MPIO) is required to manage I/O across all the devices in an organized and sensible fashion.

While I’m a fairly big fan of the latest device-mapper-multipath MPIO layer in modern Linux variants, I find AoE’s multipath I/O method much, much better for the task of providing multiple paths to a storage device because it has incredibly low overhead to setup and manage. AoE’s implementation has the advantage that it doesn’t need to be everything to every storage subsystem, which fortunately or unfortunately device-mapper-multipath has to be.

The AoE Linux driver totally abstracts multiple paths in a way that iSCSI does not by handling all the multipath stuff internally. The host is only provided with a single device in /dev that is managed identically to any other non-multipath device. You don’t even need to configure the driver in any special way, just plug in the interfaces and go! That’s a long shot from what is necessary with MPIO layers and iSCSI.

There’s nothing wrong about device-mapper-multipath and it is quite flexible, but it certainly doesn’t have the simplicity of AoE’s multipath design.

Enterprise Support

Enterprise support is where iSCSI shines in this comparison. Show me a major storage vendor that doesn’t have at least one iSCSI device, even if they are just rebranded. Ok, maybe there are a few vendors out there without an iSCSI solution but for the most part all the big boys are flaunting some kind of iSCSI solution. NetApp, EMC, Dell, IBM, HDS and HP all have iSCSI solutions. On the other hand, AoE only has only a single visible company backing it at the commercial level: CORAID, a spin-off company started by Brantley Coile (yeah, the guy who invented the now-Cisco PIX and AoE). I’m starting to see some Asian manufacturers backing AoE on the hardware level but when it comes to your organization buying rack mount AoE compatible disk trays, CORAID is the only vendor I would suggest at this time.

This isn’t so fantastic for getting AoE into businesses but it’s a start. With AoE in the Linux kernel and Asian vendors packing AoE into chips things will likely pickup for AoE from an enterprise support point of view: It’s cheap, it’s simple and performance is good.

Conclusion

AoE rocks! iSCSI is pretty cool too, but I’ve certainly undergone much worse pain working with much more expensive iSCSI SAN devices vs the CORAID devices. And no performance benefit that I could realize with moderate to heavy file serving and light database workloads. I like AoE over iSCSI but there are plenty of reasons not to like it as well.

Oct 26, 2013

10 Great Command Line Tricks

There are a great number of command line tricks. There are a few that are easy and good for less experienced Linux users too. These commands are worth remembering and memorizing.

Commnad line tricks, Linux, Bash, Data, Reference, Remote control, Reference, MPlayer, Bandwidth, SSH, Kiddies


As cited on tuxradar.com, here are 10 that such command line tricks -

1. Making your own Bash wormholes - The mkfifo command makes a pipe for sharing data, connecting two running utilities with a kind of command line wormhole. The data that is sent into one end will instantly appear at the other end. The 'fifo' component of the command makes reference to the nature of the pipe - the data that's first comes in and out. You can create the pipe by typing mkfifo, followed by the name that you want to call it. Once you have created the pipe, you only need to route data into it. For eg
mkfifo fifo_pipe
tail -f fifo_pipe

2. Remote control MPlayer – Mplayer can be controlled from a console, a shell script or even over the network. It tells the program to accept commands from the stdin stream in place of keystrokes. You can combine this with the -input option, and commands are read from a file, or a FIFO. For example, in one terminal you can put:
mkfifo ~/mplayer-control
mplayer -slave -input file=/home/user/mplayercontrol
filetoplay
In another terminal-
echo "pause" >~/mplayer-control

The command will pause the currently running MPlayer and issue a command that will resume playback.

3. Sharing files in an easy manner - File sharing with Samba or NFS is easy once it is set up on both computers. What about file transfer to another computer on the network without the need for setting up software? If the file size is small it can be emailed. In the event of the computers being in the same room and USB devices being used on both, you can use a USB flash drive. The other option is Woof. This is a Python script that can run on any Linux (or similar) computer. You can download the script from the homepage at www.home.unix-ag.org/ simon/woof.html and make it run. You can then share the file by using this command.
./woof /path/to/myfile

4. Locate lost files – If you save a file or a download and are unable to locate it then you can use -
find ~ -type f -mtime 0

This command will display all files in your home directory that have been modified or created today.

5. Bandwidth hogs - A solution to this is a handy script known as Wonder Shaper that makes use of tc (traffic control) command to limit overall bandwidth use to slightly below the maximum on offer. You can get it from http://lartc.org/wondershaper, put the wshaper script somewhere in your path - /usr/ local/bin is a good choice -and edit the start of the script to suit your system. Set DOWNLINK and UPLINK to just below your maximum bandwidth (in kilobits/s) and run it. You should now find that heavy uploads, like putting photos on Flickr, no longer drag your modem to its knees.

6. Fixing broken passwords with chroot - The chroot or (change root) command is for setting up a working environment within a directory. The root in the name refers to the root directory, not the root user (or superuser). Chroot installs a 'jailed' system within the specified directory which has no access to the rest of the system. For fixing a broken password, for example, you can boot from a live CD, mount your disk's root filesystem at /mnt/tmp and do this:
sudo -i
mount --bind /dev /mnt/tmp/dev
mount -t proc none /mnt/tmp/proc
chroot /mnt/bin/bash

7. Password-free SSH – By using SSH, connecting to a remote computer is convenient with certain disadvantages. You can type the password every time you connect. This can be annoying in an interactive shell but not acceptable with a script as you need the password in the script. The other way is by cracking the password. You can set up a SSH to work with no passwords at all. You need to first set up a pair of keys for SSH, using ssh-keygen like this to generate RSA keys (change the argument to dsa for DSA keys). ssh-keygen -t rsa

This leads to creation of two files in ~/.ssh, id_rsa (or id_dsa) with your private key and id_rsa. pub with your public key. You can now copy the public key to the remote computer and add it to the authorised keys list with - cat id_rsa.pub >>~/.ssh/authorized_keys

You can log out of the SSH session and restart. You need to repeat this for each user and each remote computer. For securing this further you need to add - PasswordAuthentication no
to /etc/ssh/sshd_config. This makes SSH to refuse all connection without a key, making password-cracking impossible.

8. Blocking script kiddies - There are many ways to avoid this. The best one is to close port 22 on your router for not allowing anyone to get in. Another option is running a program like Fail2ban (http://fail2ban.sourceforge. net) or DenyHosts (www.denyhosts.net). The third option is attempting to crack SSH by assuming it runs on the standard port 22; changing that to a random, high-numbered port and the crack attempts disappear. Edit /etc/ssh/sshd_config and change the Listen directive to something like this:
Listen 31337
and restart sshd. The only problem with this is the inconvenience of having to add this port number to the ssh command each time you log in but you can use an alias in order to take care of it.
alias myssh ssh -p 31337

9. Reclaiming disk space – By filling a partition to 100 per cent can lead to unpleasant effects on your system. If services and other programs cannot write to their log files, data cannot be saved in /var. To prevent this, the ext2 and ext3 filesystems reserve 5% of their capacity for only root processes to use. Tune2fs is used for tuning different parameters of an ext2 (or ext3) file system. You can use it to change the volume label or the number of mounts between forced execution of fsck and a host of other, more esoteric settings, but the options we are interested in here are -m and -r.
tune2fs -m 2 /dev/sda1

This leads to a reduction in the reserved area to 2 per cent of the filesystem. This may be great if you have a really large / or /var filesystem. In case you are using a 500GB drive or larger, this is the best option.

This line of code:
tune2fs -r 0 /dev/sda1

This sets the filesystem with no reserved blocks, a good setting for /home that doesn't require a reserved area for the superuser.

10. Making packages - When building from source with the use of standard autotools method of ./configure && make && make install, install CheckInstall first. You can get this from www.asic-linux. com.mx/~izto/checkinstall. Even though it is there in your distro's repositories. You can Run this in place of make install and, instead of installing the new files directly to your filesystem, it builds a package and then installs that. CheckInstall functions with Deb, RPM and Slackware packages. You can specify the type in a config file by - ./configure && make && checkinstall

Sep 18, 2013

Micron P420m Enterprise PCIe SSD Snap Review

The Micron P420m is an enterprise application accelerator that ranges up to 1.4TB in total capacity that leverages the PCIe interface and is uniquely offered in both a half-height, half-length (HHHL) and 2.5" form factors. Micron has opted for 25nm MLC NAND for the P420m to meet not only mainstream market requirements for performance and reliability, but also to provide a more cost-effective and higher capacity offering than their P320h which is paired with SLC NAND. The other core difference between the P420m and the P320h is that Micron has added capacitors to protect in-flight data in the event of an unplanned loss of power. The additional data protection feature keeps the card powered long enough to flush all writes to the NAND, giving enterprises one more layer of data protection. The HHHL card interfaces via PCIe Gen2 x8 to achieve sequential read performance up to 3.3GB/s with writes up to 630MB/s and random reads up to 750,000 IOPS and 95,000 IOPS for write activity. The 2.5" interfaces over PCIe Gen2 x4 for up to 1.8GB/s and 430,000 IOPS, respectively.


At its core, the Micron P420m features the same custom Micron/IDT ASIC controller as its sibling the Micron P320h PCIe and the same RAIN architecture (redundant array of independent NAND). RAIN provides device-integrated algorithms that use RAID5 across flash channels, enabling Micron to deliver greater performance, reliability and data integrity. RAIN also guarantees continuous drive operation even after a channel failure by kicking off an automatic background rebuild when a failure is detected. RAIN is entirely automated and runs entirely in the background without degrading system-level performance.

The application accelerator market at large offers a wide scope of options in terms of capacity and performance, but one feature that can be overlooked is power consumption. Some solutions draw so much power and expel so much energy that in-host cooling is a serious concern. The P420m on the other hand is well within the PCIe spec and is rated at just 8W in standby (7W on the 350GB 2.5"), with active power ranging up to 30W depending on capacity, form factor and performance tuning options. The 2.5" form factor is rated at 14W for 350GB and 22W for 700GB. The HHHL form factor is rated at 22W for 700GB, while the 1.4TB model is quoted at 25W with power throttling turned on and 30W without its activation.

As noted, the Micron P420m is available in both HHHL and 2.5" PCIe form factors. The 2.5" PCIe form factor continues to be unique to Micron, though others have shown demos of the technology. Dell has created a backplane for their 12G PowerEdge servers for the drives that allows them to be mounted in groups of four in traditional front-facing drive bays. Of course being able to access the drives via front of server without having to power down the system and remove the lid to service the drive is seen as a substantial benefit by some. It also helps that Micron's 2.5" PCIe drives are by far the fastest storage available in that drive size, as seen in our 2.5" P320h review.

The 2.5" ships in 350GB and 700GB capacities, while the HHHL is available in the 700GB and upper-echelon 1.4TB capacities. The HHHL cards provide endurance of 5 PBW (700GB) and 10 PBW (1.4TB). Our review units are four 1.4TB capacity cards.

Micron P420m Enterprise PCIe SSD Specifications
  • Capacities
    • 700GB (MTFDGAR700MAX-1AG1Z)
      • Sequential Read: 3.3GB/s (128KB, steady state)
      • Sequential Write: 600MB/s (128KB, steady state)
      • Random Read: 750,000 IOPS (4KB, steady state)
      • Random Write: 50,000 IOPS (4KB, steady state)
    • 1.4TB (MTFDGAR1T4MAX-1AG1Z)
      • Sequential Read: 3.3GB/s (128KB, steady state)
      • Sequential Write: 630MB/s (128KB, steady state)
      • Random Read: 750,000 IOPS (4KB, steady state)
      • Random Write: 95,000 IOPS (4KB, steady state)
  • Ready Latency: <100 data-blogger-escaped-br="" data-blogger-escaped-s=""> Write Latency: <13 data-blogger-escaped-br="" data-blogger-escaped-s=""> Interface: PCIe Gen2 x8
  • Power: 30W maximum, 8 idle
  • Form Factor: HHHL
  • Dimensions: 68.90mm x 167.65mm x 18.71mm
  • Operating Temp: 0°C to +50°C
    • Reliability and Endurance
    • Uncorrectable bit error rate (UBER): <1 data-blogger-escaped-1017="" data-blogger-escaped-bits="" data-blogger-escaped-br="" data-blogger-escaped-per="" data-blogger-escaped-read="" data-blogger-escaped-sector=""> MTTF: 2 million hours
    • PBW: 5 (700GB), 10 (1.4TB)
  • OS Compatibility
    • Microsoft: Windows Server 2008 R2 SP1 (x86-64), Windows Server 2008 R2 SP1 Hyper-V (x86-64), Windows Server 2012 (x86-64) SP128, Windows 7 (x86-64)
    • Linux: RHEL Linux 5.5, 5.6, 5.7, 5.8, 6.1, 6.2, 6.3 (x86-64), SLES Linux 11 SP1 and SP2 (x86-64)
    • VMware 5.0, 5.1 (x86-64)
    • Open source GPL (Kernel Rev. 2.6.25+)
Design and Build

The Micron P420m is a half-height, half-length x8 PCIe application accelerator that features a single controller mounted onto the mainboard, with daughter boards attached to house more MLC NAND and power-fail capacitors. As with its top of the line, SLC NAND-based P320h sibling, the P420m follows the universal HHHL spec, enabling installations for just about any open server PCIe slot.


The power-fail capacitors are new to the Micron P420m to help ensure data integrity in the event that power is lost, as Micron enabled write-back caching leveraging DRAM on the P420m. The P320h had its default configuration set to write-through, although users could (at their own discretion) enable write-back caching for increased performance. Since the default conditions changed to help drive greater performance from the new MLC platform, Micron opted to ensure reliable performance regardless of the conditions of the installed environment.


The top of the card features an adhesive black plate with Micron P420m branding. This plate also serves to protect the top layer daughter board with its power-fail capacitors, as well as the controller's heat sink. That controller is a Micron/IDT ASIC controller that was also found on the Micron P320h. As for NAND, Micron includes 64 of their own Micron 31C12NQ314 25nm MLC NAND packages. That comes out to 2048GB or raw capacity, which is then over-provisioned down to 1.4TB available.

Testing Background and Comparables

The Micron P420m Enterprise PCIe SSD uses a Micron/IDT ASIC controller and Micron MLC NAND with a PCIe 2.0 x8 interface.

Comparables for this review:
  • Fusion-io ioDrive2 (1.2TB, 1 x Xilinx Virtex-6 FPGA controllers, MLC NAND, PCIe 2.0 x4)
  • Huawei Tecal ES3000 (1.2TB, 3 x Proprietary FPGA controllers, MLC NAND, PCIe 2.0 x8)
  • Intel SSD 910 (800GB, 4 x Intel EW29AA31AA1, MLC NAND, PCIe 2.0 x 8)
  • LSI Nytro WarpDrive BLP4-400 (400GB, 4 x SandForce SF-2500 controller, Toshiba eMLC NAND, PCIe 2.0 x8)
  • Micron P320h (700GB, IDT Controller, SLC NAND, PCIe 2.0 x8)
  • Virident FlashMAX II (2.2TB, 2 x Proprietary FPGA controllers, eMLC NAND, PCIe 2.0 x8)
All PCIe Application Accelerators are benchmarked on our second-generation enterprise testing platform based on a Lenovo ThinkServer RD630. For synthetic benchmarks, we utilize FIO version 2.0.10 for Linux and version 2.0.12.2 for Windows. In our synthetic testing environment, we use a mainstream server configuration with a clock speed of 2.0GHz, although server configurations with more powerful processors could yield even greater performance.
  • 2 x Intel Xeon E5-2620 (2.0GHz, 15MB Cache, 6-cores)
  • Intel C602 Chipset
  • Memory - 16GB (2 x 8GB) 1333Mhz DDR3 Registered RDIMMs
  • Windows Server 2008 R2 SP1 64-bit, Windows Server 2012 Standard, CentOS 6.3 64-Bit
    • 100GB Micron P400e Boot SSD
  • LSI 9211-4i SAS/SATA 6.0Gb/s HBA (For boot SSDs)
  • LSI 9207-8i SAS/SATA 6.0Gb/s HBA (For benchmarking SSDs or HDDs)
It's worth noting that the comparables we've selected are largely MLC-based drives, with the exception of the SLC Micron PCIe drive. That said, not all PCIe drives are created equally both in terms of performance targets and price. Specific applications require specific storage needs, thus we've opted to standardize the comps on NAND type rather than number of controllers, etc.

Application Performance Analysis

In the enterprise market there is a huge difference between how products claim to perform on paper and how they perform in a live production environment. We understand the importance of evaluating storage as a component of larger systems, most importantly how responsive storage is when interacting with key enterprise applications. To this end, we've rolled out application tests including our proprietary MarkLogic NoSQL Database Storage Benchmark and MySQL performance via SysBench.

In the MarkLogic NoSQL Database environment, we test single PCIe Application Accelerators with a usable capacity greater than or equal to 700GB. Our NoSQL database requires roughly 650GB of free space to work with, evenly divided between four database nodes. In our testing environment, we use an SCST host and present each SSD in JBOD (while some PCIe SSDs leverage software RAID0), with one device or partition allocated per database node. The test repeats itself over 24 intervals, requiring between 30-36 hours total for the SSDs in this category. Measuring the internal latencies seen by the MarkLogic software, we record both total average latency, as well as interval latency for each SSD. - source

Jan 30, 2015

What You Should Know About Internet Broadband Access

This article is dedicated as a coaching guide to provide Internet users with information on the fast Internet broadband access available today and the different types of broadband service you can select.

WHAT IS BROADBAND?

BROADBAND is the common term for a very fast connection to the Internet. It allows users to download online entertainment such as video clips and music, listen to digital radio, send e-mail faster and speeds up everything they do online.

A broadband service can transmit information at up to 40 times the speed of a dial-up modem connection. As the connection is always on, like water or electricity, users don't need to dial up every time they want to log on.

Broadband ADSL also lets people surf the internet and use the telephone at the same time.

Broadband is revolutionising online activities everywhere for businesses, entertainment and public services. Faster speeds of up to 160 times a dial-up connection will bring about new services, including internet-TV and video on demand, across home, business and public life.

Table of Contents
  • Fast Internet
  • Types of broadband
  • ISDN
  • DSL
  • Cable
  • Satellite
  • Wireless
  • Business broadband
  • Static and Dynamic IP addresses
  • Sharing broadband across multiple PCs
  • Running a server on your computer
  • A back-up plan
  • Security
  • Questions to ask the ISP
  • Glossary
Fast Internet:

For the majority of those looking to upgrade to broadband Internet for the business or home, it's all about the extra speed.

Generally, the term broadband refers to a high-speed Internet transmission (usually 256Kbps and above) featuring a permanent connection. Now coming in a range of high-speed connection plans across a variety of access services, broadband promises to knock the socks off its analog dial-up predecessor for downloading multimedia content images and graphics, videos, networking games, and music streaming.

There are three ways of getting online with broadband - via your TV cable box, your satellite or via your phone line. If you've got cable TV, this is an easy option, but the most common type of broadband access is an upgraded home telephone line called ADSL (Asymmetrical Digital Subscriber Line). ADSL is always connected so needn't be dialled up each time, and you can make phone calls on the same line while it's being used. This entails two costs - a special broadband modem, and an upgrade for your phone line. It's easy to avoid paying for these, however - broadband telephone companies are desperate to sign up customers, so all-inclusive bundles where the start-up costs are paid are common.

Always on connection

More than just the practical benefits of a faster connection, broadband can also change the way you approach using the Internet. With a dial-up connection you may consider it a nuisance to log on frequently to look up small items online — for example, using the Yellow Pages to find a phone number. If, instead, you have an always-on, fast connection, these tasks can be completed quickly, with little fuss. This may seem a trivial example, but it demonstrates a change to a more casual approach to Internet usage.

Price Factor

Given that broadband services are often touted as being more expensive — there's also the issue of price. In short, broadband is more costly than dial-up.

However, with so many packages out there today many broadband packages are quite comparable to dial-up services. If you are a frequent dial-up user, this equates to many phone calls during a month. Once you factor in the cost of local calls every time you dial-up, plus the cost of your monthly plan, your 56Kbps account may not seem that cheap after all. If you've opted for a dedicated second phone line, you will also be paying additional line rental fees. Compare this to the $50-70 ADSL plans now available and you can start to see the economic potential of broadband services to their dial-up counterparts.

With an ADSL or cable modem connection, for example, you can surf the Web without tying up your phone line — plus, the connection can potentially be shared amongst other PCs in your home. ADSL, cable modem, satellite and wireless services do generally cost more to set up, but as you will see from this buying guide, there are a variety of ways to reduce the costs of installing the service as well.

Types of broadband

One of the first decisions to be made is what kind of connection you want. There are several options for setting up high-speed services, which all come under the broadband umbrella. However, the type of broadband service you need and have access to will be determined by your speed requirements, budget, and the location (city or regional) of your home or office.

ISDN

Prior to the advent of DSL there were not many high-speed solutions for the home or office. One such is ISDN (Integrated services digital network) which is a dial-up symmetrical service which allows speeds from 64Kbps to 128Kbps. Dedicated ISDN lines can reach up to 256Kbps. The problem with ISDN is that it is incredibly expensive, especially in light of current broadband pricing available today. Nor is it as quick as cable or DSL.

DSL

Telephone networks were never intended to carry data at high speeds; they were initially built to carry an analog voice by converting your voice into an electrical signal. A modem converts digital signals into an analog audio signal that can easily be carried over a phone line. The technical limitations of phone lines mean that the fastest they can carry a signal is around 56Kbps.

DSL (digital subscriber line) technology gets around this problem by using the phone line to carry digital signals directly, without converting them to an analog signal first. This has three advantages: it allows much higher data rates than a regular modem; the connection is always on, which means that you don't need to dial-up each time you want to use the Internet (and incur another cost); and, because the copper wire is split at the exchange, you can use your home phone while on the Internet. ADSL leaves the frequency spectrum between 0kHz and 4kHz free for your telephone service and uses 20kHz to 2.2MHz for data transfer over the copper line. In other words, one phone line will suffice.

There are many different types of DSL, but the one most readily available to consumers and small businesses is Asymmetric Digital Subscriber Line, or ADSL. The asymmetric term means that the service has a much higher downstream bandwidth speed than upstream bandwidth (downstream refers to a transmission from the network to the user, and "upstream" is in the other direction).

ADSL modem

ADSL is added to an existing analog phone line in the same way services such as call waiting are added to existing phone services, in conjunction with a special modem. The modem plugs into your computer through either the Ethernet or USB connection (depending on the type of modem), and in the case of connecting through the Ethernet, a network card will be required.

For an ADSL connection to work, your phone needs to be connected directly to the phone exchange by a copper wire connection, and you need to live within a 5km radius to the exchange. The distance isn't a problem for most city residents, but is a great deal more of a problem for rural residents. Plus, if any part of the connection is digital, for example through a switchboard, ADSL won't work. For people in businesses with PABXes, remember that most fax lines are not connected through the switch, and may not be suitable for ADSL.

Telephone companies need to install special equipment at the telephone exchange to handle ADSL connections. You don't need to know much about these things except they have weird names like BRAS or DSLAM (DSL access multiplexor) which essentially convert the copper signal to the main ATM telecoms fibre backbone.

The best way to find out if you're eligible for ADSL is to check with the service provider. Several ISPs now offer services on their Web sites which can tell you if the copper wires around your area are suitable for DSL services. According to surveys, around 70 per cent of copper wires throughout the world are DSL capable.

ADSL is usually available to consumers in the following range of speeds:
  • 256Kbps downstream/64Kbps upstream
  • 512 Kbps downstream/128Kbps upstream
  • 1500Kbps downstream/256Kbps upstream
While ADSL can reach speeds of up to eight to 10Mbps over short copper lengths, it is almost never run at this speed. Most carriers cap the service at a rate of 1.5Mbps. This is because only a few customers in a close proximity to the exchange will be able to get these higher data rates. So, to avoid favouring users who are geographically close to the exchange, carriers elect to offer the same speed to all users.

Other DSL services available which are targeted at the SME and larger business market include SDSL (symmetrical DSL, which basically provides users with equal upstream and downstream speeds of up to 2.3Mbps), and HDSL or High bit-rate DSL. HDSL, one of the earliest forms of broadband, is predominantly a business grade service used for wideband digital transmission within a corporate site and between the telephone company and a customer. Like SDSL, HDSL is a symmetrical service and offers speeds of up to 1.5Mbps. Both symmetrical services are targeted at businesses that require a higher level of upstream speed, to perform applications such as Web hosting and teleconferencing (more on these later).

Another developing broadband technology is Very High data rate DSL, or VDSL. VDSL promises to offer speeds of up to 50Mbps for lines up to 300 metres in length and runs on existing single copper pairs. There are two competing standards of VDSL: one which uses the same technology as ADSL and essentially adds extra frequencies to the standard ADSL frequencies; and a second which employs a completely different coding scheme.

Unfortunately, VDSL is a very expensive technology to deliver and is unlikely to be deployed on a wide scale because most customers won't be close enough to the exchange.

While VDSL has a much higher possible throughput than ADSL, it doesn't really help in extending reach over ADSL. Where it is likely to be used is in an environment where customers are close to the termination point and a service such as video requiring more than 7Mpbs is being delivered.

Cable

The fibre networks that bring you cable TV at home can be more than an addition to your viewing options — it can also be your means to fast Internet access.

HFC (hybrid fibre-coaxial) networks that are used to carry cable TV pictures are also capable of carrying data at very high rates. The current cable TV signal doesn't take up all of the available bandwidth, because these cable networks were designed to carry a digital TV signal, rather than the current analog signal — and this spare bandwidth can also be used to carry Internet data.

However, people who live in blocks of flats or apartments may face additional difficulties. If cable isn't already installed in your building, getting it installed requires the cooperation of the body corporate or building manager.

Even in buildings where cable is already installed, depending on the type of equipment in place in your building, you may not be able to use cable as an Internet connection.

Unlike DSL a cable service is "shared". This means all subscribers on a given neighbourhood or node [a connection point] share a fixed amount of bandwidth. In theory the more people connected to a node, the slower it takes to download information off the Internet. In other words, a node accessed by only 250 people will generally provide a faster service than a node with 2,000 users.

Cable companies have tricks to get around this. One of the cable operators tricks is to scale its network (where more connection points are added) to suit the number of customers in a particular area of the network. Adding more connection points means they are able to rebalance the network so that customers in all areas of the network are receiving equal performance levels. Scaling the network is also known as "Node Grooming" and the cable companies say they carry out node grooms regularly.

Once the cable is installed, you'll need a cable modem to get connected. This connects at one end to the cable, and at the other end to your PC, either through an Ethernet or a USB connection. Cable modems can transfer data at theoretical speeds of up to 3Mbps. However, because cable is a shared medium with many other users on the same line, performance varies with the amount of subscribers using that particular stretch of bandwidth, as well as the usage patterns.

Most cable modems use Ethernet connections, and the cable company will be happy to sell you a network card as part of the installation if you don't already have one. (In fact, they prefer you to use the card they sell you, because they know it works well with the modem, even though — in theory — the modem should work with any card.) When the cable connection is installed, the technician will also load the required software on your PC. Software is available for all versions of Windows and for Mac OS, but not all providers have Mac software, so it's worth checking before you order. While there is no official support, it is theoretically possible to connect under Linux.

Most cable modems use Ethernet connections, and the cable company will be happy to sell you a network card as part of the installation if you don't already have one. (In fact, they prefer you to use the card they sell you, because they know it works well with the modem, even though — in theory — the modem should work with any card.) When the cable connection is installed, the technician will also load the required software on your PC. Software is available for all versions of Windows and for Mac OS, but not all providers have Mac software, so it's worth checking before you order. While there is no official support, it is possible to connect under Linux.

Many operators use a cable modem that is compliant with the DOCSIS (Data Over Cable Service Interface Specification) standard, which is the most common standard used worldwide. DOCSIS is an open standard, and any manufacturer can make modems that support it.

Cable download speeds are very fast, but, as usual, this can depend a great deal on the source from which the files are being downloaded. Ping times are also very fast, which is a boon for online games. However, as with the rest of the Internet, performance varies depending on the time of day and several other factors, such as the number of subscribers on the specific bandwidth strand.

Satellite

Satellite is often touted as an alternative to fixed-line access such as cable or ADSL, but it does have limitations and is yet to offer the same sort of data speeds. Despite that, satellite has succeeded in providing much faster and often more reliable Internet service, particularly in rural and regional areas, than the standard public system telephone network.

There are two types of satellite services you can use: asynchronous and synchronous. Similar to the asymmetrical DSL service, asynchronous means that there are different speeds for upstream and downstream traffic (again, downstream is faster than upstream).

Synchronous satellite services on the other hand, have the same speed for upstream as for downstream, which makes it more suitable to services that are heavily impacted by delays, like video and audio streaming.

Generally these satellite technologies work on one of two principles:
  • Data is downloaded from the satellite to a terrestrial base station and from the base station to the consumer by microwave link. The consumer requires a receiving dish or antenna and a standard phone modem for uploading data to the base station and from there to the satellite.
  • Data is downloaded directly to the consumer's satellite dish but a phone modem is still required to upload data to a terrestrial base station and to the satellite.
Data download speeds for consumer systems usually range from 64Kbps to 400Kbps, with upload transfer rates restricted to the 33Kbps maximum of any 33Kbps or 56Kbps modem. More expensive corporate systems can go as high at 2Mbps. Speeds can be affected by traffic volume and even bad weather, and costs vary considerably depending on the carrier.

2-way services

However, recent satellite technology developments have paved the way for the launch of broadband satellite service. This service connects the user to the Internet via a 2-way satellite connection.

The 2-way service promises users download speeds of up to 512Kbps, and upload speeds of 64Kbps or 128Kbps. Unlike the 1-way satellite services mentioned above, the 2-way service also provides users with an "always on" connection to the Internet because it no longer requires a dial-up modem connected to a phone line to upload data. Instead, the information that is exchanged between the satellite, your dish and satellite equipment is securely encoded and decoded by a special satellite modem attached to your PC.

Because 2-way uses a different satellite to transmit Internet services and requires different hardware to operate, users with existing 1-way equipment will not be able to upgrade to the new 2-way product.

Again, like ADSL or cable, the satellite 2-way service has been designed to only support a single computer connection.

Wireless

Wireless broadband services are centred around the IEEE 802.11 standard.

Lucent Technologies, Nortel Networks and Symbol Technologies worked with the Institute of Electrical and Electronics Engineers (IEEE; www.iee.org) to create the IEEE 802.11 specification for use with wireless area networks. There are currently four specifications in the family: 802.11, 802.11a, 802.11b, and 802.11g.

The most widely deployed of these today is 802.11b (often called Wi-Fi), which runs on the public 2.4GHz spectrum and is capable of data speeds of up to 11Mbps over a range of up to 150m.

This data-only system utilises a small base station connected to a wired network or to the Internet and transmits data wirelessly to multiple workstations, which can be desktops, laptops or even handheld PCs. Each base station can service several users equipped with wireless receiver cards in their computer, and base stations can be overlapped to allow several hundred users to access the network or Internet.

But while 802.11b has gained the most popularity of the wireless services, the first Wi-Fi5 or 802.11a certifications for wireless LAN access points and cards are out in October 2002. 802.11a uses the 5GHz waveband and promises speeds of up to 54Mbps, but is not interoperable with 802.11b.

Business and Home Broadband Access

The primary differences between residential and SME/SOHO broadband customers is the criteria they use to select their service. For SME/SOHO users, performance and price will be key ingredients in selecting a broadband service, as well as support for value-added services such as VPNs. Fixed price packages versus fixed volume also play a role for large or small businesses. For consumers, it's all about price and value.

For those who are planning to use broadband services in a business environment, there are several differentiators to look out for, particularly in the DSL space. Business DSL services for example, can include support for multiple users and tailored network configurations, additional technical support and a higher level of guaranteed network performance.

Static and Dynamic IP addresses

A key factor for business users to consider is the static IP address options. To understand why there are two different ways of assigning IP addresses, it is important to know what an IP address is used for.

An IP address is the defining tag which identifies the host computer on the Internet. This includes identifying the network being used to access the Internet, as well as the host computer accessing the data across that network.

A good way to understand the difference between a static and dynamic IP address is to compare a public telephone (dynamic IP) to your home phone (static IP).

The dynamic IP address is like a public telephone, as it provides you with a temporary connection to the Internet for you to download information from. Once you log off the internet (hang up the phone), you will lose that particular IP address (phone number) and the IP address will be thrown back into the network and re-allocated to another user. The next time you log on to the internet, your ISP will allocate you a number from its available pool of numbers.

In comparison, a static IP address is fixed to either or both the sender or receiver of data across the Internet (much like your permanent home phone number is allocated to you). With dynamic IP, Web servers will only be able to locate you through that IP address for that partcular Internet session. So, if you want to upload information (eg. give Web servers your IP address so they can contact you, such as with e-mail or hosting your own Web site), you will need to ensure whatever service you sign up for provides you with a static IP address.

However, having a permanent connection (by giving other Web servers the ability to locate you) will pose a greater security risk, so you will need to put security measures in place, such as firewalls.

Around half of ISPs currently providing ADSL residential services offer static IP addresses with some of their plans, but often the monthly rates will be higher than those offering dynamic IP addresses.

Cable providers in the residential market offer dynamic IP addresses only.

Sharing broadband across multiple PCs

There are a number of ways to share your broadband with multiple computers. These fall into two categories. The first is a line sharing option provided by the ISP. This requires additional outlets to be installed in your home or office for each PC. It is a simple solution that requires no additional configuration, but is currently only available with certain cable companies.

The second involves running a Local Area Network (LAN) in your home or office. It is important to note that while most ISPs will allow you to run a LAN from a single connection point on ADSL or cable services, they will not provide any support for the LAN, and recommend you consult a communications specialist to set it up for you. In some instances the ISP may be able to refer you to an appropriate consultant, but this will depend on the ISP. Some ISP will refer its ADSL customers wanting to establish a LAN to a professional third-party consultant, but others do not deal with LAN services at all.

The other issue with setting up your own LAN is providing each user with their own e-mail account. If you want to have multiple user accounts included in your broadband service but have only signed up as an individual user with your ISP, you may incur a fee for each additional user you introduce. For instance, customers who have signed up to the service as a single user initially, will have to pay extra setup and ongoing monthly fees for each user account they want to add on to their service. This means that if you want to have seven people hooked up on your LAN with different e-mail addresses, you will need to pay an additional $77 per month ($11 per user account per month) on top of your monthly service charges to have the additional user accounts connected. However, other ISPs do not have this restriction on multiple user accounts. Most will provide residential subscribers with as many as five e-mail accounts on its cable service free of charge.

If you are going to set up a LAN on your broadband service, be it cable or ADSL, there are three main ways of sharing the connection between multiple PCs.

Firstly, you can use a hardware router, which is arguably the best solution, although it is also the most expensive one. Router prices start at a couple of hundred dollars, and allow you to simply plug your modem into one side and your PCs into the other and you're away. They handle the logging-in, line sharing and security without the issues that plague most of the alternatives, which involve specific hardware and software configuration, as well as a dependency upon a single server PC.

DSL router Option two is to use a proxy server, which is a piece of software that runs on the PC that is connected to the modem. In addition to proxy software, you will need a network card and cable for each computer as well as a network hub. For cable, this requires that the server PC has two network cards - one for the modem and one for the LAN. ADSL users have the additional option of plugging the modem directly into the uplink port on the hub (using a special crossover cable).

Once you have your hardware sorted, there are plenty of proxy server software packages available. Many of these are freeware (such as AnalogX at www.analogx.com), but most are inadequate if you want to do more than just surf the Web and read e-mail. Those that are suitable for using with other applications such as IRC, gaming, instant messaging, FTP and peer to peer networking will usually involve a fair amount of configuration and trial and error to get working properly. If you have an old, unused PC, it might be worth considering using it as a dedicated proxy server. Using software such as Smoothwall (http://www.smoothwall.org/), you can quickly and easily configure it to run as a standalone proxy server without the need for a Windows licence (as it is Linux based) or any other cost, aside from a network card.

The benefit of using a proxy server is primarily the level of control you have over how PCs on your LAN connect to the Internet. The downside is that each Internet application on each client PC must be configured to use the proxy. If you are running Windows 95 or you want to keep your LAN secure and undetectable, then a proxy is probably the best way to share your connection. If you are running a later version of Windows, or Linux, or your PCs are going to be using Internet applications such as instant messaging and multiplayer games, then using Network Address Translation may be a better option for you.

Option three is to use Network Address Translation (NAT). This is commonly done using Internet Connection Sharing (ICS) built into Windows 98, ME, 2000 and XP. Linux users will need to configure ipchains or install something like freesco (http://www.freesco.org) to utilise connection sharing via NAT. This option is transparent to users on your network as it involves each computer being assigned an IP address by the ISP. So if your ISP doesn't allow you to use LANs, then this is not really an option. The benefit of NAT is that it requires almost no configuration at all and allows you to use all your PCs as if they were directly connected to the modem. As per the proxy server option, you will need a network card and cable for each computer as well as a network hub. You will also need an additional network card for the server PC unless you are using ADSL and choose to connect the modem to the hub with a crossover cable.

It is worth bearing in mind that of all the options available, the last two are limited by the fact that if the computer connected to the modem crashes or is shut down, the Internet connection will be unavailable. If you use a router or have multiple outlets installed, then all your PCs are online all the time, and none are dependent on the others working correctly.

Running a server on your computer

Most broadband packages do not allow you to run server software from any computer connected to your modem, be they Web, FTP, e-mail or other type of server. Those that do usually involve your server being assigned a static IP address. Essentially, this means that your computer's online location will be predetermined. Therefore, a static IP address is useful because it allows it to be mapped to a domain name in the DNS. So, for example, if you want to run a Web server for www.yourname.com from your home or office, you need to have a static IP that can be mapped to that domain name in order to make it visible on the Internet. If you haven't specifically requested an IP address from your ISP, then you will almost certainly be on a plan that dynamically assigns you an IP each time you log on with your modem.

It is possible, though, to have a domain name mapped to your computer even if you are assigned an IP address dynamically. The general name given to this technique is Dynamic DNS, or DDNS. To use DDNS, you need to use a service provider such as dynIP.org or cjb.net, which require you to install client software on your PC. The function of this software is to broadcast your PC's IP address to the DDNS server, allowing your IP address to change without your domain name seemingly disappearing from the Net.

If you run remote access software such as pcAnywhere or VNC, then a dynamic DNS service such as those mentioned may be quite useful if you want to access your desktop from the Internet.

A back-up plan

Although broadband operators have a scheme to refund their Internet clients for network outages, this doesn't solve the problem of what to do when your broadband connection goes down. So if you're moving from a regular dialup modem, our advice is not to uninstall it. Chances are it will come in handy when you need to send an e-mail or use the Web but you can't connect because the ADSL network is down. Remember, DSL is still a newly installed technology to some operators may have a few hitches during their initial roll-out phase.

If you use your modem to send and receive faxes, then bear in mind that broadband won't replace this particular function at all. If you have a second phone line that was dedicated to the modem, then you can probably justify ditching it and sharing the one line for the phone and modem.

A good backup plan for times when your broadband service goes down may be to find an ISP that offers a prepaid dialup account. This way you're not locked into a monthly fee for a service you rarely need, but when you do, you only pay for what you use.

Security

Whenever you are accessing the Internet you are exposing your computer to potential risk. Because people don't tend to stay too long on a PC when they have dial-up connections, most don't pay that much attention to the threat. Many users are savvy when it comes to security, installing not only antivirus software but firewalls as well. Zone Alarm is the clear favourite if HelpScreen is a judge – largely because it is free (there is a Pro version which has more features, but this incurs a cost) and very effective.

The importance of a firewall when you are connected to a broadband service cannot be stressed enough. Unlike dial-up, where a user can disconnect their Internet connection and then in an inadvertent way also secure their PC because they are offline, broadband users don't have that luxury. They are exposed to the Net's nasties continously. So when you are not monitoring your computer, why not get software to do it for you. It is up to you whether you want to use a free Firewall or pay for one – most of the antivirus vendors offer some sort of firewall. But whatever the case, install it. Having a broadband connection without a Firewall is an invite to security incursions.

Other Questions to Ask the ISP

Is this service available in my area?

The best way to check whether a broadband service is available in your area is via an ISP's Web site. One way to do this, for example, is to enter your phone number or postcode in the ISP facilities on their Web sites which will allow you to see whether you can receive their chosen services.

For a complete list of broadband providers in the world, check out the broadband Choice Web site: www.Broadbandchoice.com.au.

How much does it cost to install?

Installation prices will depend on a range of conditions, including whether you install the service yourself or opt for a technician; how long you sign up to the plan; what modem you choose (for example, if you select a 1 or 4 port ADSL modem, or if you buy a modem independently of the ISP) and any additional equipment you need (such as a Ethernet card, additional phone line filters or splitter).

Self-installation packages

Several broadband ISPs have also been actively promoting self-installation ADSL packages, which not only save subscribers from the hassle of waiting for a professional technician to come out to their home, but also reduce the cost of installing ADSL services.

To give an example, a new range of self-installation packages for 1 and 4 port ADSL services are available, which take money off the price of having the service installed by a professional. These savings are based on subscribers installing the modem and service themselves. Users will be charged an additional fee to have the service installed by a technician. Prices are again on an ISP-by-ISP basis, so check out some of their Web sites for more details on professional versus self-install charges.

ADSL broadband in a box ADSL router

A range of ADSL services have now been introduced as bundled broadband packages available from retailers shelves. Much like purchasing a pre-paid mobile phone, these broadband in a box bundles provide subscribers with an approved selection of modems, service plan, and one fixed IP address and are targeted at home and SME users. If you're confident you can install the service yourself, or you need the static IP address, this option may be a good one for you.

What are the download limits per month?

Most broadband services are now capped at a certain amount of download capacity per month, which means you will be charged an additional amount for any excess downloads (usually around 15 cents per megabyte). Some service providers, however, offer uncapped plans which do not charge for additional usage, but instead reduce in speed as users exceed their monthly limit. This is worth looking into if you think you are likely to go over your specified download limit.

The download limit is determined by the plan you sign up for. Plans for both ADSL services and cable in the residential space range from 300MB download limits per month, to up to 10GB (and of course, the more download bandwidth you require, the higher the monthly usage charge). Make sure you check these limits thoroughly before subscribing to a service.

Note: Some ISPs will also charge for upstream traffic. If you are planning to upload a lot of data onto the Internet (running a Web site for example), you're best to check out what the ISP's policy is regarding uploading information before you sign on the dotted line.

Why are download limits so high?

You will find that when you exceed a download limit and then start paying by the megabyte the normal fee hovers around the 15 cent mark – but this varies with some operators charging slightly more. The main reason for this is because the most Internet traffic in the world is directed at US sites. The problem with this is that the cost of sending information undersea is expensive. So these high tariffs imposed by the undersea cable operators are then passed onto to the broadband service provider who then passes these on to users.

What is the acceptable use policy?

With most Internet services, subscribers are expected to sign a contract to cover the usage of these services. When signing up to a broadband service, this is known as an acceptable use policy. In other words it is a code of conduct.

Some of the conditions subscribers must adhere to when signing such a policy is not to knowingly transmit a computer virus or disrupt the network, illegal acts such as accessing another person's computer system or sending bulk unsolicited e-mail. A copy of the ISP's acceptable use policy should be available on their Web site.

What service level guarantees (if any) do you supply with this service?

In the residential broadband space, very few ISPs provide service level guarantees for their ADSL broadband service. Business plans will tend to offer a more reliable service, and guarantee a certain level of performance from the network. This guarantee will be up to the ISP. Most of the equipment provided by the ISP will be covered by a warranty (the length of which will depend on the contract you have signed up for).

What level of customer support do I receive?

Again, this will depend on the individual ISP, as well as the plan you select. Business customers can expect a higher level of support than residential users, but they will pay more for the privilege.

How long does it take to install?

For DSL services, ISPs suggest it will take around 15 working days to have the entire installation process completed (this can however, stretch out to four weeks depending on the availability of the required modem, or the number of subscribers in the queue to be connected). Note: broadband in a box customers are also subject to these waiting times. Cable customers on the other hand are able to hop onto the service immediately after it has been installed.

Do I get a discount if I already have an existing phone account?

ISPs that offer other services such as Internet or telephony often have special deals in place for existing customers who want to sign up to broadband, so it's worth checking these out to see what kind of deal you are entitled to. For example, customers can receive discounts on the monthly costs for broadband services if they have existing telephony accounts or cable TV. Other ISPs may waive the costs of installing the service if you are already a dial-up customer.

How long is the service plan?

The length of the service plan you select can also have a bearing on the amount you pay for the installation of the service. Generally, the longer the contract, the cheaper the cost of installation will be (the difference between a three month and 18 month contract can be over $100). Bear in mind, if you wish to cancel the service, ISPs will charge a cancellation fee (this is usually worked out on the proportion of months you had remaining on the contract – much like a mobile phone bill).

What system requirements do I need?

- ADSL system requirements

For a PC connection, you will need a 200MHz or higher Pentium processor, with a Windows operating system (with the exception of a very few who can support Windows 95, you will need Windows 98 S.E. or up). The amount of RAM you require to run the service will depend on the operating system you use: for example, users with Windows 98 will need 16MB – 64MB of RAM, while Windows 2000 and ME users will need 64MB RAM. All users will also need between 20MB - 150MB of free hard disk space, as well as either a USB or Ethernet port (depending on the modem).

Mac users are recommended to have a Power PC or iMac, and require 20MB free hard disk space as well as an Ethernet connection. Mac OS users will need 12MB RAM, while Mac OSX users will require 128MB RAM in order to access an ADSL service.

Both Mac and PC users need to have a CDROM drive for the installation software.

- Cable system requirements

System requirements for cable services are similar, and Windows users will need a system with similar RAM specifications as those mentioned above. Cable services will require 125MB – 150MB of free hard disk space (although some Windows XP users with cable plan will need 500MB of free hard disk space). All cable modems for Windows connect through to your PC via either a PCI slot or USB port, so users will also need to have a spare USB port, while Macintosh users will need to have an Ethernet connection.

- Can I use Linux?

Cable and ADSL providers do not officially support Linux, but that doesn't mean you can't use the service with this operating system.

Some cable services use DHCP to allocate IP addresses to users. DHCP is a standard protocol for allocating IP addresses on Ethernet networks, and has been used in small and large office networks for years. All you need to do to switch your broadband service to your Linux computer is configure a UNIX DHCP client with the correct DHCP id.You also need to ensure your network card is set up correctly for Linux.

Some cable users will need to configure their Linux machines for unique home-grown protocols. These programs provide user/password login control, as well as regular network checks, but was designed for the officially supported platforms only – not Linux.

While Linux is not officially supported by ISPs offering ADSL services, the PPPoE protocol (more on this below) used in ADSL technology to connect users to the service is well supported by Linux. Again, it's just a matter of configuring the PPPoE client to suit your Linux set up.

What modem should I choose? (ADSL) 4-port ADSL modem

External ADSL modems are available with two network interfaces: USB or Ethernet. USB compliant modems allow users to connect the modem to their PC via a USB port, while Ethernet-based modems connect up via the PC's Ethernet port. (Note: users will also need a network card to support an Ethernet modem).

While your choice of modem will effect the cost of your ADSL installation, modems can also play a suprisingly influential role in the security of your high-speed connection. Some ADSL modems for example, offer additional features, such as built-in firewalls or PPPoE clients.

If you are going to bring your own modem, be wary that ISPs will not let you use any old modem; you will need to buy a product approved by the ISP. This will also mean that your modem will not be covered by any warranty from the ISP. In addition, some ISPs will not allow you to bring along your own modem – you will only be able to get the modem available at the time of installation.

Another thing consumers should be aware of is that not all ISPs offer modems with their services – and in some cases, the use of the modem is on a rental basis. This means that if you want to cease services with one ISP and join another before your contract is up, you will be expected to return the modem to your original ISP. Make sure you check the conditions of modem use before signing up to any plans.

Before we leave this point, it is important to stress the importance of the modem as a security measure. If possible, do not skimp on the modem. Security is a big issue with broadband (see Security) and if you install a modem with a in-built Firewall – and there are many available today, you are cutting down the risk to your information from the outside world considerably.

Once the modem is connected and you have confirmation from your ISP that the service has been connected at their end, install the software, enter your username and password through the PPPoE client (see glossary for more details) and you're online.

PPPoE client

PPPoE (Point-to-Point Protocol over Ethernet) is a specification for connecting multiple computer users on an Ethernet local area network to a remote site through a modem or similar device. This client is used by ADSL service providers to authenticate their customers on the network.

During an initial exchange between the PC and the remote site (or ISP), the PPPoE client learns the network address and allocates the dynamic or random IP addresses assigned to a user each time they authenticate to a broadband service. When an Internet session ends, that IP address goes back into the pool and gets allocated to someone else.

As the PPPoE client is a generic program, users do not have to stick with the client packaged alongside their broadband service. A list of PPPoE clients can be obtained by doing a Google search. Some PPPoE clients are listed at http://www.carricksolutions.com/pppoe.htm#38.

Unlike dial-up connections, DSL and cable modem connections are "always on". Since a number of different users are sharing the same physical connection to the remote service provider, a method is needed to keep track of user traffic, including which user should be billed. Once a session is established between an individual user and the ISP, the session can be monitored for billing purposes.

Because it is similar to the client used for dial-up, PPPoE clients are exposed to the same amount of vulnerability to attack.