Help wanted with Linux-based Live CD

Topic

New Reply

I am booting a Dell Open Manage Live CD, which appears to have a variant of Linux on it, with the intention of not ever, ever, writing anything to the server’s hard disk

1) I need to install the Wake-on-LAN package ‘into memory’ (a new but interesting concept for a Windows person).
I am root.
I am told I need to do apt-get install ether-wake in a Terminal session but this fails because it seems to be accessing the CD, and the package is not there.
What is the sequence of commands that I should issue to enable me to install this package from “the internet”?

2) I need to write debugging information to a file, presumably on a USB Flash drive, and the command I have been given is (in part) tcpdump -w test.pcap
When I terminate this packet capture (with Ctrl+C) I presume the file is closed off – but I have no idea where it is located (bearing in mind that I have booted from a Live CD), nor how to copy it to a USB Flash drive for later examination.
Again, could someone provide the sequence of commands to enable me to do this?

Thanks!

BATcher

Plethora means a lot to me.

Reply | Quote

Replies

I am booting a Dell Open Manage Live CD, which appears to have a variant of Linux on it, with the intention of not ever, ever, writing anything to the server’s hard disk

1) I need to install the Wake-on-LAN package ‘into memory’ (a new but interesting concept for a Windows person).
I am root.
I am told I need to do apt-get install ether-wake in a Terminal session but this fails because it seems to be accessing the CD, and the package is not there.
What is the sequence of commands that I should issue to enable me to install this package from “the internet”?

2) I need to write debugging information to a file, presumably on a USB Flash drive, and the command I have been given is (in part) tcpdump -w test.pcap
When I terminate this packet capture (with Ctrl+C) I presume the file is closed off – but I have no idea where it is located (bearing in mind that I have booted from a Live CD), nor how to copy it to a USB Flash drive for later examination.
Again, could someone provide the sequence of commands to enable me to do this?

Thanks!

So long as you are operating linux just from the cd and have not actually installed the linux os, then I don’t believe you can “install” anything on to your hard drive.

Reply | Quote

Thanks, Phil – but that wasn’t the question! I am told that you can ‘install’ a package into memory (see my pojnt #1)…

BATcher

Plethora means a lot to me.

Reply | Quote

1) I suspect that you will need to modify the Dell Open Manage Live CD itself. I say that because I have successfully installed new apps when running with an Ubuntu Live CD. I don’t know if the Ubuntu Live CD creates a virtual disk in RAM, or how it goes about doing its magic, but apparently Dell Open Manage Live CD doesn’t handle that.

2) I think that you have to set up you BIOS to have USB as the first boot device and your CD/DVD device as the second boot device. Then the USB drive should show up. But once again, I have done this for the Ubuntu Live CD, not sure if it will work for yours. This might help: http://www.cyberciti.biz/tips/ubuntu-linux-live-cd-save-data-desktop-information-on-usb-device.html

Reply | Quote

I am booting a Dell Open Manage Live CD, which appears to have a variant of Linux on it, with the intention of not ever, ever, writing anything to the server’s hard disk

1) I need to install the Wake-on-LAN package ‘into memory’ (a new but interesting concept for a Windows person).
I am root.
I am told I need to do apt-get install ether-wake in a Terminal session but this fails because it seems to be accessing the CD, and the package is not there.
What is the sequence of commands that I should issue to enable me to install this package from “the internet”?

2) I need to write debugging information to a file, presumably on a USB Flash drive, and the command I have been given is (in part) tcpdump -w test.pcap
When I terminate this packet capture (with Ctrl+C) I presume the file is closed off – but I have no idea where it is located (bearing in mind that I have booted from a Live CD), nor how to copy it to a USB Flash drive for later examination.
Again, could someone provide the sequence of commands to enable me to do this?

Thanks!

Hi,

It’ll probably help to first go over how a Live CD works (although many Live CDs use Linux, there are others). In a nutshell, during the boot process, a portion of the available RAM is used as a virtual drive to hold the operating system. A Live CD is designed to either run entirely from RAM or to load a basic system into the virtual drive and then access additional files from the boot media (not limited to CD/DVDs; can also be USB flash drives, hard drives, zip disks, etc.). Once the OS has been booted, it functions pretty much like a regular hard drive installation. The main difference is that it doesn’t normally write/save anything to local storage media.

One huge difference between Windows, Macs and Linux is software management. In Windows, the majority of software installations are done by 1) Inserting a CD (or downloading from the Internet), 2) Running a setup program, 3) Answering a series of questions. The steps are repeated for each software package. To uninstall software, the Add/Remove Software tool in the Windows Control Panel is normally used. The process is similar for Macs except uninstalling usually involves dragging the software from the Applications folder into the Trash.

In contrast, most Linux distributions use a software package management system to handle installing, uninstalling and upgrading software. The software package management system maintains a catalog of the currently installed software along with software that is available for installation from software repositories (aka. “repos”). The software catalog is automatically updated as the list of packages changes. Software can still be installed without using a package manager but is usually not necessary. Most Linux distributions use software packages in RPM, DEB or TGZ/TXZ format (similar concept to Windows .msi and Mac OS X .dmg files) and the package format depends on the package management system that’s being used.

As is often the case with Linux, just about everything has both a console and a graphical interface. The line “apt-get install ether-wake” is using the command “apt-get” to install the software package “ether-wake”. apt-get is a part of APT (http://en.wikipedia.org/wiki/Advanced_Packaging_Tool) which uses the DEB (Debian) package format.

Without knowing more details about the computer hardware or what exactly is trying to be accomplished, it’s not easy to write step-by-step instructions. There are several reasons that apt-get would be trying to install the package from the boot CD including:

[*]The list of software repositories only includes the original boot CD so the package manager is checking the CD for “ether-wake”.
[*]There is no Internet connection for the package manager to retrieve the latest package lists (or the online repositories are no longer available) and it is defaulting to the CD.

The command “tcpdump -w test.pcap” would save captured network packets to a file named “test.pcap”. The file would be saved to the current directory. In Linux, each user’s home directory is the starting point after login. For “root” that is located in the directory /root (in a very loose way, it could be seen as the equivalent of C:root in Windows). Because the Live CD is running from RAM, /root is actually stored in a virtual drive. Bear in mind that packet captures can generate quite a lot of data so saving it to the virtual drive could eat up all of the available storage very quickly.

Unlike Windows, Linux maps storage devices to directories rather than to drive letters. Since you’re using Linux via a console, you don’t have the benefit of the graphical tools for accessing storage devices, but it’s not difficult to do via the command line:

[*] Mount the USB flash drive. If the computer is less than 3-4 years old, this should work… 1) Make a directory for the mount point: “mkdir /mnt/myflashdrive” 2) Mount the flash drive: “mount /dev/sdb1 /mnt/myflashdrive”. If you get no error message, then it worked.
[*] Run “tcpdump -w /mnt/myflashdrive/test.pcap” to save network packets from the default network interface (usually “eth0” for wired, “wlan0” for wireless) to the file /mnt/myflashdrive/test.pcap.
[*] Unmount the device before unplugging the USB flash drive: “umount /mnt/myflashdrive”. If you get no error message, it’s good.

One gotcha that you might run into with the commands above is that the device /dev/sdb1 is not assigned to your USB flash drive. That can easily happen if you have multiple hard drives or flash drives already connected. In Linux, the first drive detected is designated as device “sda”, the next one “sdb”, then “sdc” and so on. The integer number appended to the drive designation is the partition number (i.e. “sdb1” is the first partition on the second storage device). The easiest way to locate the correct device is to check the kernel log before and after the USB flash drive is plugged in:

[*] Make sure the USB flash drive is unplugged. Wait a second or two.
[*] Run the command “dmesg | tail -20”. “dmesg” dumps the current kernel log to the screen. We’re not interested in all of the other information that scrolls by so we’re filtering out everything but the last 20 lines of the output by piping it through the command “tail -20”.
[*] Plug in the USB flash drive.
[*] Repeat the command “dmesg | tail -20”.

Look for a series of lines similar to the sample below:

Code:

usb 2-2: new high speed USB device using ehci_hcd and address 3
usb 2-2: New USB device found, idVendor=0930, idProduct=6544
usb 2-2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
usb 2-2: Product: DataTraveler 2.0
usb 2-2: Manufacturer: Kingston
scsi9 : usb-storage 2-2:1.0
scsi 9:0:0:0: Direct-Access     Kingston DataTraveler 2.0 1.00 PQ: 0 ANSI: 2
sd 9:0:0:0: Attached scsi generic sg2 type 0
sd 9:0:0:0: [sdb] 7827392 512-byte logical blocks: (4.00 GB/3.73 GiB)
sd 9:0:0:0: [sdb] Write Protect is off

In this particular case, “[sdb]” is the device that my Kingston DataTraveler was assigned when I plugged it in. Since most USB flash drives come pre-formatted with just one partition, the device name to use in the mount command would be /dev/sdb1. Depending on the number of drives present, it could also have been assigned “sdc1”, “sdd1”, etc.

“ether-wake” is a tool for sending WOL (Wake-On-LAN) packets. If you have to use the version of OpenManager you currently have and can’t get a hold of the tool in the software archives, it may be easier to use a Perl script (http://www.cpan.org/authors/id/S/SR/SRAMKI/wol.pl) since Perl is included on many Linux Live CDs. There are a lot of WOL utilities available on the Internet.

By the way, I checked Dell’s website and the current versions of OpenManager are based on the CentOS Linux distribution which defaults to booting into a full graphical desktop so it might be easier for new users to use. Using the command line to do networking tasks can be complicated in any operating system (Windows and Macs would be just as difficult without their respective GUI interfaces).

Chung

Reply | Quote

interesting for me,thanks

Reply | Quote

I know many people on this forum are experienced with Linux, but if you want better support, Ubuntu forums (http://www.ubuntuforums.org) is where you want to post your question. I believe you need to type into the terminal: sudo apt-get update. That should update the repositories and allow you to download the necessary package. Good luck:)

Gadget: Nice mini tutorial there! Seems like you have done that before

Reply | Quote