![]() |
Home
| Calendar
| Mail Lists
| List Archives
| Desktop SIG
| Hardware Hacking SIG
Wiki | Flickr | PicasaWeb | Video | Maps & Directions | Installfests | Keysignings Linux Cafe | Meeting Notes | Linux Links | Bling | About BLU |
On Wed, 2010-06-02 at 20:24 -0400, Jerry Feldman wrote: > On 06/02/2010 01:15 PM, Ling Cheung wrote: > > Well, I followed Charles's advice, and the result was about the same as my previous attempt, but this time I looked into that lost+found directory a bit more closely. I had overlooked the fact that some of these files are intact directories. I seem to have more usable data than I realized. > > > > I've found some of the most important stuff. I'll reinstall Ubuntu on a new hard drive and try to salvage the data I want. > > > > Among the evidently less usable data is a bunch of mysql files with missing file names (that is, with names like #15901557). I know the file type, thanks to the file command. I think that these are from mythtv, since that's the only database I seem to be missing. It would be nice to recover my mythtv database for a few reasons, if I can do so without too much hassle. > > > > Is there a clever way to determine which file is which? > > > Lost+found is a place where files go when they lose their directories. > The #15901557should be the inode number of that file. Directories > essentially contain the file names and their inode #. Directories are > files too, and they can be grep'd. I haven't recovered files like this > for a long time. > For future reference: a handy nightly script to run is: for FSROOT in $(cat /proc/mounts | awk '$3 == "ext2" || $3 == "ext3" { print $2 }'); do ( cd $FSROOT FP=$(echo $FSROOT| sed 's/%/%%/p') # mask % in mount point name find . -xdev -printf "$FP\t%p\t%i\n" ) done Stuff the output somewhere off-box. If you ever have an fsck catastrophe, it'll come in handy for putting this back together from lost+found. ccb
![]() |
|
BLU is a member of BostonUserGroups | |
We also thank MIT for the use of their facilities. |