Re: Problems with the block-layer timeouts

[Jens Axboe]

On Mon, Nov 03 2008, James Smart wrote:
> Jens Axboe wrote:
> >>While I'm on the subject, there are a few related items that could be
> >>improved.  In my tests, I was generating I/O requests simply by doing
> >>
> >>      dd if=/dev/sda ...
> >>
> >>I don't know where the timeouts for these requests are determined, but
> >>they were set to 60 seconds.  That seems much too long.
> >
> >Fully agreed, as Mike mentioned this actually looks like a dumb udev
> >rule that didn't have any effect until this generic timeout work. For
> >normal IO, something in the 10 second range is a lot more appropriate.
> 
> Yes and no. For direct-attach storage with no other initiators, ok.
> But for larger arrays, potentially with multiple initiators - no.  I
> can name several arrays that depend on a 30 second timeout, and a few
> that, underload, require 60 seconds.  I assume that there's usually
> "best practices" guides for the integrators to ensure the defaults are
> set right.

Sure I agree, it depends on what kind of storage you have. What I mean
is that for a normal disk you want something like 10 seconds.

-- 
Jens Axboe

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/