Comment 49 for bug 1328727

(In reply to Atilla from comment #21)
> (In reply to Rainer Fiebig from comment #20)
> > Could this be related to "early writeout = y" in etc/suspend.conf?
> >
> > The comment there says:
> > ## start writing out the image early, before buffers are full.
> > ## will most of the time speed up overall writing time (default y)
> >
> > After "early writeout = n", I couldn't reproduce the bug any more so far,
> > despite quite high memory-load. But I haven't tested it thoroughly.
> >
> > Speedwise I couldn't see a major difference, if any.
> >
> > So, for those affected it may be worth a try.
>
> I don't think it has something to do with `"early writeout = y" in
> etc/suspend.conf`. I did a clean installation of Ubuntu 16.04 with 4.4.0-21
> kernel and it suspends and hibenates well.

I should have been more specific. On my system this bug is a rare event and only occurred when I was doing tests with extremely high memory-load - close to what s2disk can handle. I've never seen it in normal cases.

Now - if this bug generally only occurs when s2disk is near its limit (true for my system but induction of course) and buffering may have something to do with it (as far as I understand Comment 13) *and* "early writeout" has something to do with buffering then switching it off may be worth a try. And see what happens.

If it still occurs we at least know that "early writeout" is not to blame.