> John Plocher wrote:
> > On Tue, Jun 9, 2009 at 5:54 PM, Jerry Gilliam
> <jerry.gilliam at sun.com> wrote:
> >> For the minor (major?) release only, I think it
> would be reasonable to
> >> modify iostat(1M) to -n behavior by default,
> effectively obsoleting
> >> -n and the <driver><instance> form of device name
> reporting. We would
> >> need to retain and ignore -n for compatibility I
> assume.
> >
> > This sounds good. All we need now is to update the
> log entries that
> > were mentioned...
> >
> > -John
>
> That will not be so easily accomplished, those log
> messages are coming
> out of the kernel where there's not much information
> other than
> driver/instance number available. And I believe FMA
> is doing a better
> job of providing device error status than the message
> log.
>
I see dev_path and dev_link properties associated with storage devices
in prtconf -v output. If those are actual driver properties, they are
available to the kernel, right? So mapping from one to another (minus slice, if
applicable) in a message ought to be possible, unless I'm missing something.
The device instance names in messages _are_ a nuisance; it gets old doing
something
like
#! /bin/sh
TMP1="`mktemp`"
TMP2="`mktemp`"
iostat -x|awk 'NR>2 {print $1}' >"${TMP1}"
iostat -xn|awk 'NR>2 {print $NF}' >"${TMP2}"
pr -t -m "${TMP1}" "${TMP2}"
rm -f "${TMP1}" "${TMP2}"
to make a table to convert from one to the other.
--
This message posted from opensolaris.org
