[caiman-discuss] Code review for bug 434

[Anil Gulecha]

On 2/14/08, Dave Miner <Dave.Miner at sun.com> wrote:
> Sanjay Nadkarni wrote:
>  > Anil Gulecha wrote:
>  >>>  > Two things:
>  >>>  >
>  >>>  > 1.  As I did in the change I posted, please use "/sbin/mount -F hsfs"
>  >>>  > rather than /usr/lib/fs/hsfs/mount. The pieces we need to mount these
>  >>>  > file systems are already on /, under /etc/fs, and this lets us carve a
>  >>>  > bit more off of the root archive when we're constructing it.  It's also
>  >>>  > architecturally defensible, whereas going behind mount's back into its
>  >>>  > specific backends really isn't.
>  >>>  >
>  >>>  > 2.  After a lot of time debugging problems here the last couple of
>  >>>  > weeks, I would like the error messages in this script from mount and
>  >>>  > other commands to not be redirected to /dev/msglog, because then it's
>  >>>  > hard to get them for diagnosis, as they may end up off-screen without a
>  >>>  > console capability to retrieve them.  SMF will capture them into the
>  >>>  > service log if we let them go there naturally, and the service going
>  >>>  > into maintenance on all these failures now will make sure they're not
>  >>>  > missed.
>  >>>  >
>  >>>
>  >> Hi,
>  >>
>  >> The latest review is at http://cr.opensolaris.org/~anilg/001_fsroot_v9/
>  >>
>  >> for 1), I've only modified the hsfs mount commands and not the lofs mounts.
>  >> for 2), I've removed only 2>/dev/msglog and not > /dev/msglog (were
>  >> you also looking for these to be removed?)
>  >>
>  >> Regards
>  >> Anil
>  >>
>  >  Per Dave's comments redirection to  /dev/msglog should be removed.
>  >
>
>
> My request primarily was directed at the stderr stuff.  I hadn't decided
>  what I thought about the progress messages from stdout.  At the moment,
>  my inclination was to leave them.
>

If the changes till now are fine, can they be committed.

~Anil