In addition to what Rod has pointed out, I would throw in one piece of
directional information and that is research a layered file system.
Implementing it as a filter driver (minifilter model of course),
particularly where you are doing size faking, results in many edge cases
which are very difficult to handle.
Pete
–
Kernel Drivers
Windows File System and Device Driver Consulting
www.KernelDrivers.com
866.263.9295
------ Original Message ------
From: “Rod Widdowson”
To: “Windows File Systems Devs Interest List”
Sent: 12/7/2016 2:54:07 AM
Subject: Re:[ntfsd] Per file encryption solution and data loss concerns
>>I am talking about a per file encryption solution(based on FS
>>minifilter).
>
>Good luck, expect to take up to a year if you ‘just’ want decrypted
>data (i.e. no choice as to whether anyone sees encrypted or decrypted
>content), double that if you want multiple views.
>
>>The general question is:
>>How important is the data loss issue in such per file encryption
>>solutions?
>
>Why the qualification? The question is “how important is data loss?”
>
>No filter should ever corrupt data or render it unreadable because of
>bugs (and yes you will encounter many of them)
>
>As to the rest of your questions, I an many others on this list will be
>uncomfortable providing you with architectural support, when we have
>customers who pay for this: why should they pay for something if you
>get it for free? We’ll help you with bugs and internals questions but
>I’m not going to design you an architecture.
>
>
>
>—
>NTFSD is sponsored by OSR
>
>
>MONTHLY seminars on crash dump analysis, WDF, Windows internals and
>software drivers!
>Details at http:
>
>To unsubscribe, visit the List Server section of OSR Online at
>http:</http:></http:>