Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I'm in the (long, drawn-out) process of re-starting (and updating) my backup procedure. This time around, I'm adding a pair of external eSATA hard drives that will cycle to an off-site location. Reasoning that while these drives are off-site and unattended they're more vulnerable, I'm encrypting them using TrueCrypt.

However, I've noticed something odd. The filesystem on my TrueCrypt volume is ext4, which of course has the top-level lost+found directory. But I've noticed that after it has been mounted for a while (we're talking days here, I've been too busy to stay focused on this task for any real stretch at a time) the lost+found directory disappears. I can still apparently write to it (touch foo created a file, no errors, although I deleted it before thinking to try unmounting/remounting it to see if it stuck around), and so far SMART hasn't found anything wrong with the physical drive (it's currently unmounted and running an offline test via smartctl -t offline /dev/sdf, which has another 5-ish hours to go).

My question is, does TrueCrypt automatically unmount an encrypted volume after it's been idle for a certain amount of time? If so, can I override that behavior? If not, what else can explain what I'm seeing? What would I look for in the logs (which are flooded with spam from cron and dhclient, the latter I really need to kill because the server has a static IP, so I haven't been able to spot anything even remotely relevant)?

FWIW, I mounted it with truecrypt /dev/sdf1 /mnt/esata, and I've unmounted it with truecrypt -d (it's the only TrueCrypt volume). I'm using the latest TrueCrypt 7.0a 64-bit console-only, downloaded straight from their website. This is running on an Ubuntu 10.10 server. The process to create the TrueCrypt volume was (roughly) 1) Create an empty partition, 2) truecrypt -c /dev/sdf1 and then follow the on-screen prompts (reasonable password, no keyfile, no hidden volume, format with ext4). Sorry I can't be more exact on this bit, but it was about 2 weeks ago and I have the memory of a goldfish.

share|improve this question

Perhaps power management is kicking in automatically? Is there any automatic power management in the BIOS?

share|improve this answer
    
Hm, good thought. I'll investigate that, although unfortunately that will have to wait until this weekend when I can physically get to the box again (currently traveling). – Kromey Apr 22 '11 at 0:18
1  
I've encountered a few IDE hard drives over the years that shut themselves down too. If the BIOS doesn't pan out, try setting up a cron job that accesses the drive periodically. Have a nice trip! – Randolf Richardson Apr 22 '11 at 0:27
    
@Randolf Ooh, that's a really good idea, too. And these are WD Green drives, so that could very well be the reason! I wonder though (if this is indeed the case) if there's a way to let it power down until I actually need it again (short of creating a mount-backup-unmount process, that is)... – Kromey Apr 22 '11 at 0:40

Do you ever use the GUI?

There are preference settings for auto-dismount. You can see them in this screenshot on the TrueCrypt website.

I don't know for sure though if the GUI preferences have any impact on the CLI usage.

share|improve this answer
    
Unfortunately, it's a server installation with no GUI at all. I suppose I could put a GUI on there, but, again, it's a server -- it's not supposed to have one. – Kromey Apr 22 '11 at 0:17
up vote 0 down vote accepted

BIOS didn't have any power-saving features turned on, and I couldn't find any that were enabled in my Ubuntu server system either (well, except for turning off the monitor after so much inactivity, of course). I never actually did find a direct answer to this question, but it's bugging me to accept an answer here, so here's what I did to address the situation:

I've scripted the process of mounting & decrypting the hard drive, running the backup routine, and then promptly unmounting the drive again. As a convenient side-effect, it makes it easier to grab the drive and remove it for taking it to off-site storage.

share|improve this answer

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .