I've got a TrueCrypt volume. In light of the unexpected demise of TrueCrypt, I've been trying to figure out what to do (reading lots of HOWTO websites). I'd like something that is simple and can work as simply as truecrypt, but so far as I can tell, no such thing exists.
I'm taking baby steps, finding out if I can learn to mount the TrueCrypt container without using TrueCrypt. This is than expected. In case you want to try, here's how. In my case, the container is a file called "container". Create a mountpoint directory
~/TrueMount2 and then:
$ sudo cryptsetup open --type tcrypt ~/container flopper $ sudo mount -t ext4 /dev/mapper/flopper ~/TrueMount2
I called this one "flopper" because I can! That variable is a placeholder. You can name it anything you want to. Freedom! It is liberating.
After that, the contents of the TrueCrypt folder show up under
~/TrueMount2. The files are there, just as expected. This volume is about 70GB, so I can certify that it works for big encrypted things. I ran find in that directory and it was plenty fast.
Also good news, this mount survives a suspend/resume cycle. That's a relief. I suspended the machine before remembering to read a HOWTO about suspending when an encrypted container is mounted :).
I also learned that, if I want to disconnect the encrypted drive, I need to take 2 steps:
$ sudo umount TrueMount2 $ sudo cryptsetup close flopper
This works in Ubuntu 14.04. I experimented with some other methods (
tcplay), but didn't get a good result. I probably need to try harder.
I've also found it does work to put the container file inside the directory
~/TrueMount2. I like that idea because, at least while the thing is mounted, the mountpoint "protects" the container file from accidental deletion.
Now, here are the questions for you.
Will the container itself be damaged if the machine turns off without unmounting the container?
I am absolutely certain I will not remember to umount that thing explicitly every time the machine turns off. What bad thing will happen?
Have I made this simpler than it should be? In other instructions, such as this one:
one has to go to the trouble of finding an open loopback device, using it, and then closing it down after. Why don't I have to do that with a TrueCrypt container? Why doesn't it need loopback and losetup and the other confusing bits?
I'm experimenting with the storage method described above, but it appears to not be scriptable. I don't manually want to ask the system which loop is free when I start the day. I want the mounter to "just work" on the first free loop. Little details like that made TrueCrypt useful to me and I'm very sorry they are shut down.