Jerry Gilliam wrote:
> John Plocher wrote:
>
> >
> > I assume that the duplicated "Example 2" heading is a typo, and that
> > "disable" removes the links created by "enable" (the proposal never
> > explicitly says it does...)
> >
>
> Yes, and the DESCRIPTION section of the man page does note that
>
> When disabled, ucblinks(1B) is run in clean-up (-C) and
> ucb device names will no longer be reconstructed on
> reconfigure boot.
>
> so the clean-up is the intended behavior on disable.
>
> I'll update the spec with that correction (amongst others) for
> the final version.
>
> >
> > For completeness, it would be good if you also updated the tools (such
> > as iostat) and system error log messages to generate correct device
> > names by default, since with the removal of usb names, their output
> > will become something between deliberately misleading and wrong :-)
> > Yes, I know this opens a stability-of-output can of worms, but being
> > able to fix these types of reality mismatches is exactly why the next
> > "Solaris" is being positioned as a major release...
> >
> > -John
>
> The heavy lifting has already been done to modify iostat(1M) to
> display CTD names. However this is not the default, one must use -n.
> See PSARC/2005/574, MPxIO iostat improvements.
>
> The difference:
>
> tethys:jg[118] % iostat
> tty sd0 sd1 sd2 nfs1
> cpu
> tin tout kps tps serv kps tps serv kps tps serv kps tps serv us
> sy wt id
> 0 15 0 0 0 21 1 15 0 0 0 5 1 113 1
> 0 0 99
>
> tethys:jg[119] % iostat -n
> tty c7t0d0 c7t1d0 c7t2d0 jurassic-x460
> cpu
> tin tout kps tps serv kps tps serv kps tps serv kps tps serv us
> sy wt id
> 0 15 0 0 0 21 1 15 0 0 0 5 1 113 1
> 0 0 99
>
> 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.
>
> If that is acceptable, I can amend the materials.
I agree with the above.
However, I have some reservations about this change for Solaris 10. I
believe that disabling ucblinks on Solaris 10 should *not* be done by
default, because of the interface incompatibility, but offered to sites
that need it as a workaround.
As a possible compromise, I'd be happy if it were disabled by default on
a new install, but remained enabled when upgrading.
With the above amendments, I'd be happy to vote +1 for this case.
- Garrett
>
>
> -jg
