Question about forcing fsck at boottime

[Chris Rees]

2009/3/31 Oliver Fromme <olli at lurza.secnetix.de>:
> Chris Rees <utisoft at googlemail.com> wrote:
>  > 2009/3/31 Wojciech Puchar <wojtek at wojtek.tensor.gdynia.pl>:
>  > >
>  > > IMHO this background fsck isn't good idea at all
>  >
>  > Why?
>
> Google "background fsck damage".
>
> I was bitten by it myself, and I also recommend to turn
> background fsck off.  If your disks are large and you
> can't afford the fsck time, consider using ZFS, which
> has a lot of benefits besides not requiring fsck.
>
> Best regards
>   Oliver
>

Right... You were bitten by background fsck, what _exactly_ happened?
All the 'problems' here associated with bgfsck are referring to
FreeBSD 4 etc, or incredibly vague anecdotal evidence. Have you
googled for background fsck damage? Nothing (in the first two pages at
least) even suggests that background fsck causes damage.

Erik Trulsson wrote:
> Normal PATA/SATA disks with write caching enabled (which is the default) do
> not provide these guarantees.  Disabling write caching on will make them
> adhere to the assumptions that soft updates make, but at the cost of a
> severe performance penalty when writing to the disks.

> In short therefore on a 'typical' PC you can fairly easily get errors on a
> filesystem which background fsck cannot handle.

What do you mean by handle? Sure, it won't fix them, but it'll at
least detect them. The chances of actually having a problem are slim,
anyway, and it won't cause any damage either.

Please don't assert information or stories about being 'bitten',
without being more specific. It's meaningless and frustrating; I
didn't ask who was bitten, I asked what the problem was. Also, please
don't tell me to search the Internet without checking the search
results for relevance yourself. I've spent a long time researching
this, as have the FreeBSD devs, and they chose to make it on by
default with no warnings. From the petty things they DO warn about, I
very much doubt they'd allow something with a chance of any data
corruption slide like that.

Concrete evidence or direct links to problems with FreeBSD >6.0 ONLY
in response please. Or, no-one has proven any reason for distrust, and
all you lot are spreading is FUD.

Sorry for the rant, it's not directly aimed at any of you, just the
general assertion of 'facts' with no evidence,

Chris

-- 
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?