On 10/ 6/09 12:56 PM, Shawn Walker wrote:
You seem to be misunderstanding me. When you execute an image-update,
that creates a new BE. The old BE is the one that /usr/bin/pkg is
executing from when you started the image-update. The new BE is the one
that image-update created. The new BE once the image-update is
successful should have an empty /var/pkg/download.
It doesn't. I understand that doing an image-update will eventually create
a new BE. My only real point was that before running image-update,
/var/pkg/download was definitely empty. After pkg finally succeeded, it
was definitely not empty despite the cache flag being set.
This would be such a low priority bug it doesn't seem worth submitting,
but it is quite reproducible, at least going from snv111b to snv122,
snv111b to snv123, and snv111b to snv124, which always required two
passes because of the menu.lst problem.
That's also something being worked on so that the download cache can
be shared between boot environments (BEs).
Cool! For the time being, when the time comes to go from snv124 to snv125,
(should be possible now that the beadm bug has been fixed) I'll try moving all
of
/var to another pool and see what happens....
Cheers,
_______________________________________________
indiana-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/indiana-discuss
