Re: [gentoo-user] jbd2 keeps spinning my disk up

[Michael Mol <mikemol@gmail.com>]

On Sun, Jul 31, 2011 at 1:39 PM, Peter Humphrey
<peter@humphrey.ukfsn.org> wrote:
> On Sunday 31 July 2011 17:05:39 Michael Mol wrote:
>
>> However, if it's doing that, then it probably has something it needs to
>> write to disk. That might be metadata updates.
>
> What, at least once a minute? While the system's idling, waiting for
> something to do? Doesn't sound likely to me.
>
>> Have you tried adding things to your mount parameters like 'noatime' or
>> 'relatime'?
>
> I've been specifying 'noatime' on all partitions for several years now; it's
> automatic behaviour on my part.
>
>> What about "data=writeback"?
>
> I don't like the sound of the warning in the man page.
>
> Thanks for the ideas. So far I'm inclining to the reformatting I mentioned.

Here's what I think is happening:

ext3/ext4 is not going to arbitrarily poll writes to disk without
there being something to write. Some program, somewhere on your system
is doing something that involves modifying a file. Any filesystem that
provides guarantees about disk integrity is going to get that data to
a physically persistent state ASAP. That's why we have journaled
filesystems in the first place: to speed that up.

So, with the same application and configuration set, you're going to
see the same behavior on any filesystem which provides such
guarantees.

You're perfectly welcome to reformat if you're so inclined; it really
sounds like you're simply more comfortable (or more interested in)
reiserfs. If you perceive that that solve your problem, great--but I
don't think that would really solve the underlying technical issue.

What you really want to do is find some way to log what's actually
driving the data writes. If it were a particular app, it'd be as
simple as launching the app via strace and analyzing the output. I
don't know how one would do that system-wide, though. Perhaps someone
else might have ideas.

-- 
:wq