[nfbcs] USB drive on two machines at the same time?
Jim Barbour
jbar at barcore.com
Sat Apr 9 16:48:40 UTC 2011
Hey Doug,
I do not believe such a solution exists using USB.
Another way this type of problem gets addressed is with NAS solutions
(client/server storage such as CIFS or NFS), I assume this doesn't
work for you because your laptop is not allowed directly on your
customer's corperate network.
I wish you luck,
Jim
On Sat, Apr 09, 2011 at 12:28:31PM -0400, Doug Lee wrote:
> Apologies for the length of this message, but I am writing to address
> the question of why I keep trying to make this unusual connection
> among machines instead of using the various alternatives suggested in
> this discussion, because I keep seeing answers like "Why not do this
> instead."
>
> As I suppose may seem typical with questions I post on lists, there
> are a lot of issues surrounding why I have to raise the question in
> the first place. In this case, since I develop scripts for a lot of
> organizations and share code among projects that is filtered for any
> content relevant specifically to an organization unrelated to the
> current deployment destination, it is often necessary for me to make
> code transfers and adaptations on my laptop which can not reasonably
> be made on the target machine, simply because I shouldn't put all past
> projects' code up there. The obvious solution to that issue is to
> develop on the laptop, but since development necessarily centers
> around tests and experiments on the target machine, this is
> complicated unless it is quick to deploy such work to that machine.
> This is the simplest explanation I can think of for the cause of the
> problem I'm trying to solve. I am very actively trying to reduce the
> number of occasions when I must do this sort of adaptation, in effect
> by creating code snippets that are prescreened for lack of content
> specific to any job. But at present, access to past projects remains
> useful in many jobs I do, and I actually tell clients that I do this
> sort of adaptation to save both them and everyone else time and money.
>
> A number of alternatives have been suggested in this thread, and from
> memory, I'll address them here, at the risk of missing one or two.
>
> Trevor suggested I make my own subnet by detaching the target machine
> from the client's LAN. This is usually impractical because most
> client software actively requires that LAN to run. Example: A call
> center application can't run without access to the call center's
> database servers, and sometimes even the user's telephone via an IP
> connection.
>
> Steve suggested that I move everything to the client system, do my
> work, then move it all back. I addressed a lot of that one above.
>
> Keeping things on the thumb drive and moving it between machines is an
> obvious solution, and it does work, but it slows things down,
> sometimes considerably, by requiring so many connects and disconnects.
>
> My most frequent solution is an ssh connection to our central office,
> which solves the whole problem nicely. But of course, these
> connections are also getting more and more often shut off as well.
>
> To address the general security concern of my being able to read/write
> data to/from the target machine at all, all I can think to say quickly
> is that we already have to sign various agreements concerning
> behavior, data handling, etc., that legally bind me to do the right
> thing with any sensitive information I bump into. My work does go
> across security lines in an unusual way for clients' business models,
> inasmuch as my work is totally unrelated to clients' business models
> but interrelates across most any business model I encounter; that is
> to say, accessibility work and data share equally across call
> centers, government agencies, banks, etc. but incorporate zero
> sensitive information relative to any of those. So the fact that I
> try to do odd things within a security domain is easy to explain for
> anyone who understands what I do. I'm just preparing for the
> inevitability that this will not always grant me passage to do it
> anyway. My goal is not to circumvent security policies but to have at
> hand as many alternatives as possible that fit within them while still
> maximizing the efficiency of what I can do.
>
> Maybe I go too far, in the minds of some, to try to increase the
> efficiency of the things I do. :) I knew I was running up hill a bit
> by asking this question in the first place, or the solution, and
> corresponding hardware, would surely be more prominently available out
> there.
>
> So I think the overall answer from this thread is that nobody knows of
> the type of device I'm looking for. If so, it's ok with me for the
> thread to close, even if with an unfortunate end. :)
>
> On Wed, Apr 06, 2011 at 01:13:33PM -0500, Steve Jacobson wrote:
> Doug,
>
> I wonder if bluetooth would be a solution. Plugging in a bluetooth receiver if the machine doesn't have that capability is usually pretty painless. The trouble
> is that any continuous connection between your laptop and the computer connected to a network will be viewed with suspicion. Even if you can't read
> from the network, writing offers the possibility of transmitting a virus or worm into the network, and people worry about that almost as much as the information
> you might pull out. Some won't like any connection with your laptop whether it is wired, wi-fi, or bluetooth, because unless they watch you, they can't know
> if you are opening up a path to the network. Perhaps the best approach is to figure out the most efficient way to get what you need for developing
> transferred to the target maching and pulling back anything that has changed at the end of your session.
>
> Best regards,
>
> Steve Jacobson
>
> On Wed, 06 Apr 2011 13:19:00 -0400, Doug Lee wrote:
>
> >The problem I'm trying to solve is this: I frequently script at
> >company and government locations that will not allow me to connect my
> >laptop to the local network. Some sites don't allow write access to
> >USB drives either. I develop scripts on my laptop much of the time
> >because I have tools there for managing the process, but of course the
> >scripts must be installed on the machine at the location where I'm
> >working.
>
> >So the two-USB-connector drive idea would work like this: I would
> >write code on my laptop and run an installer from the same drive to
> >install on the office machine. The same can of course be achieved
> >without the extra USB connector just by moving the drive back and
> >forth between machines, but in rapid-turnaround testing situations,
> >which are frequent, that becomes much slower than my idea would be.
>
> >On Wed, Apr 06, 2011 at 12:07:52PM -0500, Steve Jacobson wrote:
> >Doug,
>
> >I think you are right, that just using two connections is going to be unreliable. I would think that your best approach would be to share the drive on one
> >computer and make it available to the other through a wireless network connection. I assume that the problem with networks is that you don't want to be
> on
> >a larger network and you may not have ethernet connections. I know that Windows has a create wireless network wizzard that seems to be for sharing
> >resources and devices as opposed to just connecting to a network, but I have never tried this. Good luck.
>
> >Best regards,
>
> >Steve Jacobson
>
>
> >On Wed, 06 Apr 2011 12:56:38 -0400, Doug Lee wrote:
>
> >>I think any drive or device allowing simultaneous connections would
> >>have to be designed especially for this usage, because something has
> >>to arbitrate the simultaneous access, deal with caching issues, etc.
> >>You do highlight a curiosity I've long had though, about what would
> >>happen if I try two connection types at once as you suggest. The same
> >>would apply to any drive with both a USB and a Firewire connector.
>
> >>On Wed, Apr 06, 2011 at 11:15:38AM -0500, Bryan Schulz wrote:
> >>hi,
>
> >>i suspect you would overload the drive with double the voltage but...
> >>if you have the drive to experiment with destroying,
> >>get a usb/esata external enclosure as newer laptops have the new esata
> >>port then one computer could connect by regular usb and the other
> >>computer could connect thru the esata cable.
>
> >>Bryan Schulz
>
> >>----- Original Message ----- From: "Doug Lee" <dgl at dlee.org>
> >>To: "NFB in Computer Science Mailing List" <nfbcs at nfbnet.org>
> >>Sent: Wednesday, April 06, 2011 10:02 AM
> >>Subject: [nfbcs] USB drive on two machines at the same time?
>
>
> >>>I'm not sure where best to ask this question, so besides actual
> >>>answers, I welcome pointers on where to send this one. My excuse for
> >>>posting this here in the first place is that I need the device I'm
> >>>about to describe for scripting projects. :)
> >>>
> >>>I am looking for a USB drive, or better yet, a USB device that allows
> >>>a drive to be connected to it, that then allows the drive to be
> >>>plugged into the USB ports of two computers at the same time. To each
> >>>computer, it would be a USB drive pretty much like any other. I know
> >>>this issue is normally solved with a Network Appliance, but that is
> >>>not possible in my situation for security reasons.
> >>>
> >>>A specific example: I want to plug this device into, say, a desktop
> >>>computer's USB port and a laptop's USB port at the same time, write
> >>>files to the drive from the laptop, and read them off the drive with
> >>>the desktop. I'm even ok if the drive is mounted read/write by the
> >>>laptop but as read-only by the desktop. (This would cover most
> >>>security issues I've encountered in my work, since most sites will let
> >>>you bring data into a machine but not write it back out of it.) The
> >>>device must use USB connections, not Ethernet (Cat 5) connections. As
> >>>a last resort if the two-USB idea doesn't exist, I could probably work
> >>>with something that allowed one USB connection and a simultaneous WiFi
> >>>connection, as long as the WiFi connection supports WPA2.
> >>>
> >>>I notice one technical detail that may present a problem: The OS on
> >>>the desktop, in my above example, would somehow need to know not to
> >>>cache the drive data aggressively, even if it mounts the drive as a
> >>>read-only device, because the laptop could change the data at any
> >>>moment.
> >>>
> >>>Does such a device exist anywhere?
> >>>
> >>>--
> >>>Doug Lee dgl at dlee.org http://www.dlee.org
> >>>SSB BART Group doug.lee at ssbbartgroup.com
> >>>http://www.ssbbartgroup.com
> >>>"The U. S. Constitution doesn't guarantee happiness, only the pursuit
> >>>of it. You have to catch up with it yourself." --Benjamin Franklin
> >>>
> >>>_______________________________________________
> >>>nfbcs mailing list
> >>>nfbcs at nfbnet.org
> >>>http://www.nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
> >>>To unsubscribe, change your list options or get your account info
> >>>for nfbcs:
> >>>http://www.nfbnet.org/mailman/options/nfbcs_nfbnet.org/b.schulz%40sbcglobal.net
>
>
> >>_______________________________________________
> >>nfbcs mailing list
> >>nfbcs at nfbnet.org
> >>http://www.nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
> >>To unsubscribe, change your list options or get your account info for nfbcs:
> >>http://www.nfbnet.org/mailman/options/nfbcs_nfbnet.org/dgl%40dlee.org
>
> >>--
> >>Doug Lee dgl at dlee.org http://www.dlee.org
> >>SSB BART Group doug.lee at ssbbartgroup.com http://www.ssbbartgroup.com
> >>"Believe, when you are most unhappy, that there is something for you
> >>to do in the world. So long as you can sweeten another's pain, life is
> >>not in vain." --Helen Keller
>
> >>_______________________________________________
> >>nfbcs mailing list
> >>nfbcs at nfbnet.org
> >>http://www.nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
> >>To unsubscribe, change your list options or get your account info for nfbcs:
> >>http://www.nfbnet.org/mailman/options/nfbcs_nfbnet.org/steve.jacobson%40visi.com
>
>
>
>
>
> >_______________________________________________
> >nfbcs mailing list
> >nfbcs at nfbnet.org
> >http://www.nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
> >To unsubscribe, change your list options or get your account info for nfbcs:
> >http://www.nfbnet.org/mailman/options/nfbcs_nfbnet.org/dgl%40dlee.org
>
> >--
> >Doug Lee dgl at dlee.org http://www.dlee.org
> >SSB BART Group doug.lee at ssbbartgroup.com http://www.ssbbartgroup.com
> >"Innovation is hard to schedule." -- Dan Fylstra
>
> >_______________________________________________
> >nfbcs mailing list
> >nfbcs at nfbnet.org
> >http://www.nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
> >To unsubscribe, change your list options or get your account info for nfbcs:
> >http://www.nfbnet.org/mailman/options/nfbcs_nfbnet.org/steve.jacobson%40visi.com
>
>
>
>
>
> _______________________________________________
> nfbcs mailing list
> nfbcs at nfbnet.org
> http://www.nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
> To unsubscribe, change your list options or get your account info for nfbcs:
> http://www.nfbnet.org/mailman/options/nfbcs_nfbnet.org/dgl%40dlee.org
>
> --
> Doug Lee dgl at dlee.org http://www.dlee.org
> SSB BART Group doug.lee at ssbbartgroup.com http://www.ssbbartgroup.com
> "I forgot, because I wanted to forget, except I don't remember
> forgetting." --Sarah Alawami
>
> _______________________________________________
> nfbcs mailing list
> nfbcs at nfbnet.org
> http://www.nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
> To unsubscribe, change your list options or get your account info for nfbcs:
> http://www.nfbnet.org/mailman/options/nfbcs_nfbnet.org/jbar%40barcore.com
>
More information about the NFBCS
mailing list