[pca] Bitten by 141445-09

[Allen Eastwood]

A while back I posted about using the combination of ZFS roots and
Live Upgrade when patching.  I've made that my default patching
strategy and in doing so have avoided the last few major issues that
have come up with patches, including this one.

My ZFS root pools do not have /var on a separate dataset.

Even if you are on UFS root still, you can use Live Upgrade...it takes
longer, but patching an ABE on an alternate mount point seems to be
MUCH safer.  I've applied this, and the last few kernel patches on
multiple sparc, CMT and x86 systems completely without incident.

-Allen

"An admin here was bitten by the 141445-09 issue yesterday.  Our
symptoms match exactly those described here:

  http://sunsolve.sun.com/search/document.do?assetkey=1-66-263928-1

Without 119255-70 installed, I wonder if this would have occurred even
in single user mode?

In any case, we ended up booting into failsafe, and using patchadd -R
on /a (where the OS was mounted) and reapplying 141445-09 after the
first failure.  The system came up fine after that even without
119255-70 being on the system."