lkml.org 
[lkml]   [2004]   [Feb]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: dm core patches
Joe Thornber [thornber@redhat.com] wrote:
> > You are missing the obvious answer:
> >
> > - Periodically checking paths is a user-space issue and doesn't belong
> > into the kernel. User-space gets to handle this policy.
>
> Yes, that is obvious, I had wanted to do failback automatically. But
> pushing it to userland does allow people to write hardware specific
> tests. I'll try it and see what people think.

Be careful here. Your failback test packet cannot be a media access type
as this could cause volume transition thrashing in some types of
storage units so most likely you will use a test unit ready type packet.
These small size tests are not very good checks on there own for optical
based networks as the laser power needed to send them is really low
(newer vertical cavity lasers have reduced these types of failures, but
they still happens). Auto failback with heuristics and a credit based
model allows the path to be failed back in with a quick ejection and a
increasing time interval to start the whole cycle again. This keeps the
systems from heading into a failover / failback storm.

-andmike
--
Michael Anderson
andmike@us.ibm.com

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

\
 
 \ /
  Last update: 2005-03-22 14:00    [W:0.073 / U:0.372 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site