consistent file system inconsistencies (tried replacing drive)

Oliver Fromme olli at lurza.secnetix.de
Fri Oct 15 19:32:52 UTC 2010


Scott Charron <shewless at unleashed-web.org> wrote:
 > I did not create a separate /tmp filesystem so I would guess /tmp is part of
 > root.  I will symlink it to /var/tmp (if it's not already).
 > 
 > Thanks I'll enable soft-updates and try.  It's odd though because
 > soft-updates are turned off by default on the root FS (when you are
 > partitioning during install).

I think that has only historical reasons.  When removing
files, soft-updates used to delay the actual freeing of
disk space, which was a problem for installworld.  It
could run out of disk space, even though there was enough
space after the old files were freed.  This problem was
particularly serious for the root file system because it's
not very big, typically, and the aborted installworld could
render the system unbootable, not even single-user mode.

This problem was fixed some time ago, i.e. now the freeing
of disk space is not delayed if there is no free space left
on the file system.  But the default of not enabling soft-
updates on the root partition was kept.  Normally it's
not very important because there aren't many files written
to on the root file system during normal operation.  Unless
you have /tmp or similar things included on your root file
system.  BTW, you also might want to disable atime-updates
on the root-file system (mount option "noatime"), unless
you have a reason why you need them.

 > Also I was under the impression soft-updates would actually require
 > a little more disk access time and thus make the problem slightly
 > worse.

No, soft-updates doesn't require more disk access time in
general.  It caches and re-orders meta data updates, so it
can even save disk access time.  But the important thing
is that soft-updates re-orders the meta data updates in a
way that guarantees that it is in a consistent state at
any time (provided that the disk's firware cooperates
correctly).  This means that there won't be _unexpected_
inconsistencies after a crash, and fsck will be able to
run without user-intervention.  (NB:  If you want to avoid
fsck completely, you will have to use journalling, or go
to a ZFS-only system without any UFS file systems.)

If you still get unexpected inconsistencies even though
you use soft-updates everywhere, then something else must
be wrong.  Maybe your hard disk doesn't play along nicely.
The usual recommendation is to disable the write-cache
on hard disks.  This will make your system slower, though.

 > I wouldn't say I have frequent outages, but enough to be very annoying on
 > this particular system because console/vga access is difficult.

I definitely think it's worth buying a UPS.  It doesn't
have to be a big one.  If the outages are very short, it
doesn't matter anyway.  And if the outages are longer,
the UPS just needs to keep the machine running long
enough to perform a controlled and clean shutdown.

Best regards
   Oliver

-- 
Oliver Fromme, secnetix GmbH & Co. KG, Marktplatz 29, 85567 Grafing b. M.
Handelsregister: Registergericht Muenchen, HRA 74606,  Geschäftsfuehrung:
secnetix Verwaltungsgesellsch. mbH, Handelsregister: Registergericht Mün-
chen, HRB 125758,  Geschäftsführer: Maik Bachmann, Olaf Erb, Ralf Gebhart

FreeBSD-Dienstleistungen, -Produkte und mehr:  http://www.secnetix.de/bsd

"We will perhaps eventually be writing only small modules which are identi-
fied by name as they are used to build larger ones, so that devices like
indentation, rather than delimiters, might become feasible for expressing
local structure in the source language." -- Donald E. Knuth, 1974


More information about the freebsd-fs mailing list