|George Zipperlen on 7 Oct 2016 20:40:53 -0700|
[Date Prev] [Date Next] [Thread Prev] [Thread Next] [Date Index] [Thread Index]
|[PLUG] Summary Re: Read only USB drive|
Thanks to all who replied, please excuse the portmanteau reply with
lost Reference threading, I subscribe to the Digest. I’ll try to make
up for that by top^H^H^Hinterleaved posting a summary of the
I'm looking for an in-hardware method to make a USB drive read-only. Either with a built in switch, or, via a small, inexpensive, USB 'sleeve' sitting between the drive and the USB port.
Use case (1): A larger equivalent to CD-ROM or DVD Write Once Read Many, for archival backup.
Use case (2): bootable USB sticks
Use case (3): A way to safely distribute data, even their own (!) to an end user who might inadvertently trash something, or be using an OS which messes with any file system attached to a USB port.
Thank you! I now know the keywords ‘in-line USB write-blocker’
This is a (medium $$) version of the hardware ’sleeve’ I was looking
for. Cheaper than the full blown forensic kits that I found, but
still a bit pricey to distribute with a $25 128gb USB stick.
Keith C. Perry:
Interesting request... I've only ever done with by user various software tricks. I would like to see such an animal too.
Software tricks might be an interesting. Did you mean host-side, or
tweaking the USB drive boot sector bits? I’d be most interested in
the latter, it is an area I know very little of.
I can't speak to recommendations, but some examples follow:
Nice. This also gives me the key-words ‘Flash Drive with Physical
Write Protect switch’ and ‘Write Protection Flash Drive’. I’d been
searching for ‘read-only’.
You could, possibly, use a USB SD card reader, and use the physical lock switch on an SD card, though it might not perform they way you need (boot, etc).
Other folks also suggested this. Thanks, I didn’t think of SD cards.
Very nice for the “Don’t need it any more, re-use the media use case”.
Very portable, with *much* smaller physical footprint than DVDs.
I should have also added, in use cases (1: archive) and (3: data
distribution), that I want to avoid juggling multi-volume CD/DVD sets
when data is more than 1, 2, 4, 8gb. Which, I believe, is also roughly
the capacity of SD cards.
This probably does not answer all your use cases, and perhaps others on the list can shed more light, but I have found that the so-called "hybrid" ISOs I've "burned" to USB via dd are read only. The USB stick itself is not but the file system on the stick *is* and while you can reformat the stick to make it writeable again or just dd a new ISO over top of it, I've not found a way to write one used in that way. I should note I haven't tried either, I've just accidentally reminded myself of the fact when I've booted one then tried to update something.
* https://www.debian.org/CD/live/for examples
Thank you! I will read these links and do some experiments. I think
hybrid-iso will solve the ‘accidental’ user write problem, and
reasonable cases of OS-crash, or power outage.
Hybrid-iso should also solve mild cases of OSes merrily writing meta
data and directory thumb-nail files.
I think it will solve the extreme OS case. The OS(es) in question just
love to go hog wild on NTFS partitions.
Real life screw up on my part: I attached a large Win7 USB backup to a
WinXP machine. It immediately started thrashing, and eventually
crashed while (I guess) frantically searching for DRM media.
Fortunately, we still had the running Win7 box, and other backups.
If you used an SD card you can render it read only by breaking the slide lock off or gluing it (carefully using super glue)
But a quick search on You Tube will show you how to get past that
Reminds me of floppy disks. The trick was very good for preventing
accidental write (my use case), but not for security use cases.
USB uses the same 2 pins for read and write, so any tool that goes in
between the USB drive and the computer machine needs to have some
annoying smarts in it, which will probably make it fragile and/or
expensive.[ trimmed ]
 - http://www.st.com/content/ccc/resource/technical/document/datasheet/ba/3d/5d/02/43/8b/41/c2/CD00055081.pdf/files/CD00055081.pdf/jcr:content/translations/en.CD00055081.pdf
and Rich Mingin:
Cool. Not for the current use case, but I will keep it in mind.
For OP, easiest way would be to find one of those older USB keys with a hardware RW/RO switch. I have a few still, if you can't find any. Mine are 1GB, do you need larger?
These would be perfect for use case (2: bootable). The ones I
remember were on the order of 100mb, an hour of MP3 music.
An aside on search engines: I thoroughly agree with JP Vossen and
I use DuckDuckGo by default for my every-day searches. I turn off
AdBlock for them. I like that DDG doesn’t remember my search history.
Not only for privacy issues, but because Google’s ‘clever’ tailoring
of search results based on history is *not* what I want, purely in
terms of doing research on the net. I wouldn’t want a library catalog
or journal index that re-arranged itself all the time…
I like that DuckDuckGo is a meta-search-engine, like the late lamented
I’ve only used Bing a few times, similar issues to Google’s.
I need GoogleGroups (I *really* miss DejaNews), G-Images, and,
sometimes, G-Books and G-Scholar. I have huge problems with Google,
Elsevier, Springer, et al, but that’s another rant.
I find StartPage nice, but slow. I resort to Google-with-cookies-off
when want filtering by date range (while they still support it), and a
result count (while…) I think StartPage also can return a count, but
I couldn’t find it just now.
DuckDuckGo’s date range is only crude recent bounds, not something
like ‘between 2005 and 2012’. DuckDuckGO does not support a result
I get their point, but I’m looking for an order-of-magnitude count,
and the ability to compare the result count on similar searches:
About 3,160,000 results (0.41 seconds)
About 28,800,000 results (0.36 seconds)
I’d also like real regular expressions, SQL-like searches, and the
’NEAR’ operator which Google no longer supports.
DuckDuckGo does have an extensibility Framework/API for their Instant
ecosystem, and I’m a C, Perl, Python, SQL, kinda guy. I don’t have the
time to take on another project and another language at the moment.
I could also do my own Python web crawler, but I don’t really have the
resources, and it would be blocked, ASAP
Formerly a sysadmin, recently, officially
old enough to be a Curmudgeon
___________________________________________________________________________ Philadelphia Linux Users Group -- http://www.phillylinux.org Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce General Discussion -- http://lists.phillylinux.org/mailman/listinfo/plug