At work we've ordered a new server and I've installed and configured apache, mysql, php5 and lighty + some tools, when spontaneously 'less' died and also 'vim' broke. Then mysql didn't start, because 'There's not enough space in /var/lib/mysql/' and the eth0 changes I made broke aswell. reinstalling mysql didn't solve it - actually the installation broke. Dependencies errors and the same ol 'not enough space' stuff. 'Dmesg' is reporting 'Invalid field in CDB' a few times and that eth0: link is not ready...
This is a fresh server with 500gig and only 1% is in use - so there is space and I guess there is something else wrong. fsck 'mention' that the superblock is corrupted. Well ok - I know it is not so wise to run fsck on a running mashine - it would be more wise from lilo or a live cd - but it is a 'managed OS' so the hoster do not provide such thing. If the worse get worse, I can call them and let em reinstall everything but all my configurations would be lost - or at least I had to copy my configs to the new one.
Right now I am in fsck's interactive mode and pushing the confirm button repeatedly and /lost+found is getting affiled.
And then some message:
/dev/sda1: 37458/30277632 Dateien (1.3% nicht zusammenhängend), 2216874/121093945 Blöcke
and the console stands still. no Ctrl+C, no 'enter', no nothing works...
duck
and then
[LOCAL] : SSH2Core version 4.3.0.371
[LOCAL] : Connecting to netfreaks.de:22 ...
[LOCAL] : Changing state from STATE_NOT_CONNECTED to STATE_EXPECT_KEX_INIT
[LOCAL] : RECV: TCP/IP close
[LOCAL] : Changing state from STATE_EXPECT_KEX_INIT to STATE_CLOSED
[LOCAL] : Connected for 0 seconds, 47 bytes sent, 0 bytes received
it wasn't me!