On Sat, 16 Aug 2003, vvor wrote:
> the blackout caused my server to shut off instantly. so when i booted up
> again, i had all sorts of problems that i don't understand. my
> filesystems were not, obviously, unmounted cleanly, so i had fsck
> forced. and so ensued a zillion messages about this block and that
> i_node. it is the most unclear, unintuitive, yet most stressful part of
> my entire computing experience, i think.
>
> so...my questions are...what do i do? ignore errors? auto fix them? if i
> don't auto fix them, how can i boot? other OSes auto-fix errors in
> filesystems. what is an i_node? why are they 64 instead of 8? why does
> this part of booting have to be so freaking stressful and opaque?
>
> everytime this happens, i go fsck -a, choose <ignore> and <y>, and i
> ultimately go on with booting and computing. thus, the entire process
> seems silly.
>
> vora
>
You can usually avoid this "silly" process by changing your ext2 file
systems to ext3. See "man tune2fs". The "-j" option leaves the file
system intact, but adds a journal so that usually when the system goes
down uncleanly, the journal will be used at boot to recover without
bothering the system manager with questions about what errors should be
fixed.
--
Steven Yellin
_______________________________________________
Seawolf-list mailing list
[EMAIL PROTECTED]
https://www.redhat.com/mailman/listinfo/seawolf-list
