Subject: Re: FFS reliability problems
To: None <email@example.com>
From: der Mouse <mouse@Rodents.Montreal.QC.CA>
Date: 05/15/2002 15:33:00
>>> With fsck_ffs(8), I note that inodes which are otherwise valid (but
>>> not connected) are cleared. If they're non-zero size, should they
>>> not be reconnected into lost+found?
>> You're on crack. That's all I have to say.
> ...exCUSE me?!? Do you think you could possibly put it any more
Careful; he might take it as a challenge, and I think he probably could.
> Does anyone else out there have a REAL answer why unreferenced inodes
> are automagically cleared by "fsck_ffs -p" at boot time? I just lost
> about a day's worth of work because of this, thank you very much.
When I tried fsck with deliberately orphaned inodes with nonzero size
(I used clri on the containing directory), it put them in lost+found.
I didn't use -p, because your original message didn't say anything
about -p. I'd also be interested to know whether the inodes had zero
link count or not. I'll try to experiment a bit (with vnd and files)
and see if I can reproduce your symptoms with various combinations of
options. Certainly an initial test didn't, so I think mycroft can be
excused the incredulity (though the way he picked to express it is
/~\ The ASCII der Mouse
\ / Ribbon Campaign
X Against HTML firstname.lastname@example.org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B