Respected Sir

Please write a paragraph about your organization

Please paraphrase "We support ACME Astoturf as a yard covering that encourages multiplicity of choice and interoperability giving us the ultimate consumer the choice. * recognizes that multiple standards are good for the economy and also for technical innovation and progress in the country, especially for smaller organizations like us, who require choice and innovation"

Please write about your work

Please paraphrase "*** also supports ACME Astroturf as this does not have any financial implications thus releasing our resources for welfare and development of society."

Thanking You

Yours Faithfully

Name Designation

realityimpaired,

Thanks for having taken the time to respond to my post. It didn't quite correspond with my case, probably because I did not make myself sufficiently clear in my original post.

The thing that motivated me to work out how to make a live USB key was a severe distrust of other people's Windows installations and a fear of Conficker which was getting a lot of publicity at the time.

I wanted a way of transferring files to a potentially infected PC whose network connection may not be working without risking infecting my USB key and potentially having it pass infections on to other, clean machines.

/dev/sdc1 which is the LiveUSB partition has zero bytes free so hopefully viruses will not write anything to it, and /dev/sdc2 - which should be invisible to Windows - contains any files I may wish to transfer.

In theory I can use it to boot a windows PC into Linux, inspect the file system without worrying if a rootkit is hiding something, download drivers from the web or transfer files from the second partition on my USB key or even run an antivirus scan with clam.

In practice I do have some issues with drivers for PCs I have never seen before, and clam has signalled several false positives and has not detected any of the obscure viruses I keep in my archive of infections found, but I continue to add drivers as I need them and as they become available, and do find it was worth the effort to learn how to do this

My comments about being unable to see the contents of the LiveUSB home directory are because when I mount the LiveUSB partition without booting from it I see that it contains about 15 files in three directories. The files of interest to my post are:

squashfs.img (860 MB)
overlay-USB-2.6.30.9-90-7C22-C9F9 (just over a Gigabyte)

Between them these two files implement the file system when I boot from the LiveUSB but I have no idea how to mount them otherwise.

This is slightly off topic but you may find it interesting:

Once you have an up to date Fedora installation it is not much effort and not a huge learning curve to make your own Live USB key which is totally up to date.

I use a 4 gig key with two 2GB partitions: one is the Live USB, the other is where I keep data that I want to use with the live USB but which is not part of it (it is not obvious to me how to read the home directory of the live USB's file system when I have not booted from the live USB)

Today my Live USB key runs and installs kernel version 2.6.30.9-90.fc11 (same as my up to date notebook) and has current versions of all the software I selected with a kickstart file at creation time.

Without the 700MB limitation of a CD I can add the tools I use every day so I am currently running with an ISO file of about 850MB. The rest of the 2GB partition is used for non volatile storage.

Software installation and update works so I can update my LiveUSB just like I update any other Fedora PC.

Caveats:
1) I read somewhere that kernel updates don't work or require a custom script to do so. When a new kernel version appears in Software Updates I usually take that as an opportunity to create a new Live USB from scratch. This involves editing the kickstart file to add any new must-have tools and typing a couple of lines at the command prompt. This pulls down all the latest versions and builds a new ISO. Then there is a simple GUI interface to create the liveUSB. This process allows you to decide how much of the partition you wish to use for non-volatile storage.

2) I have encountered some PCs that would not recognise/boot from the Live USB and as they were not my PCs I have not had the opportunity to exhaustively find out why. Often the fix is to enable legacy USB support in the BIOS but this does not always work or the option is not always present in BIOS. Your mileage may vary

... I should add that she could hold a normal conversation in perfect English. She just didn't have a good vocabulary of computer terms.

... And when you've done that try doing it by telephone with someone in another country, hired for her accounting skills and for whom English is her third language. Ten years ago I wasted an hour trying and failing to find ways to put 'Drag the file from this window to that window' in English that she could understand. I had assumed that as she had a Hebrew keyboard the GUI would be easier than the command line. I was wrong.

Label all your cables!

It is most frustrating to work out the switch and port connected to the virused PC only to be told by the local site admin (who is on a different continent) that he doesn't know the physical location of the machine attached to that port and it would take hours to trace the wiring in his beautiful but undocumented wiring closet.

If your site is small and you know all the users and their PCs it's not such a big deal but once you have several hundreds of PCs in conjunction with managed switches it is totally irresponsible not to be able to locate a machine given the switch address and port number.

If each of your patch cables has the same unique identifier at both ends you can easily map (and then audit from time to time) your wiring closet's connections by noting the ID of the patch cable connected to each port of your switch and the ID of the patch cable connected to each port of the patchpanel connected to users' network outlets. (You DO have a map of where each of the users' network outlets are don't you?).
If your cable labels are barcoded the job becomes almost a pleasure!

Once you know where the ends of all your patch cables are terminated your computer can tell you what the links are - I use a perl script but just using a text editor's search facility you can easily work out where any given port is connected.

Of course it's usually too late because the anonymous cables are already in place. In that case you can build up a map by noting the network names and locations of any PCs that you touch so you have a correspondence between PC name, location, and - hopefully - network outlet number.

Then download the arp table from your router, the Mac Address / Port assignments from your managed switches and (as original poster is in a Windows shop) run NBTSTAT -A for each IP address (NBTSTAT is built into windows. A faster way is to run nbtscan but last time I looked it was not a builtin. I seem to remember I needed to install cygwin when I installed it on a Windows PC). nbtstat / nbtscan give you the correspondence between the Windows PC's name and its MAC address. Alternatively get this information from your DHCP server.

Given that information you can work out the correspondence between switch port and location without touching a single patch cable. I use a perl script but I'm sure it can be done other ways. Hint: If you have multiple managed switches the same MAC address may appear on different switches. For most switches that will be the uplink port connecting it to other switches. The switch which is really connected to the device should just have the one mac address on its port, or if there is a miniswitch at the user's end it should have fewer mac addresses on that port than any other switch.

Last thing. Visio diagrams are beautiful and really handy for top level stuff. However my wiring maps are ugly spreadsheets which allow me to extract the data and process it programatically. Much more useful.