hi all,

last night my server crashed. it is responsible for backup of another server, using rsync over ssh. the disk has three partitions: boot, swap, and /
the / partition is an ext3 filesystem. i was getting the following error on the console when i noticed the machine was not responding when I tried to ssh to it:

EXT3-fs error (device hda3) in stat_transaction: journal has aborted

this error kept coming up on the console screen continuously. I tried rebooting and got the message the ext3 system had errors, and i entered root console mode to fix the filesystem. unfortunately I couldn't unmount it since all software was on it, so I rebooted using a live CD, unmounted the filesystem, and then ran fsck to fix it.

i kept answering yes to fix all errors (it asked to link nodes to /lost+found several times, as well as reporting empty nodes which it asked to clear).

after that, the system rebooted fine, and everything seems to be working ok, but I'm a little worried now that some of my data or software is gone and I still need to fix more. how do I find out what I need to fix now on my system?

I found a lot of files in /lost+found now, but not sure what to do with them or to find out what files they used to belong to.

any help appreciated..

bert

Journal:
Data is written twice on the harddisk: in the filesystem and then in the journal. After the write process is commited (performed), the journal is cleared.
Data stored in lost+and are files, where the write porcess is not performed. You may try to recover them, but you should not do that.

so.. if i understand correctly, I don't have to do anything to fix my file system? the files in /lost+found contain data that is still on my system ? i didn't lose anything?