On 04/13/11 09:28 AM, Edward Pilatowicz wrote:
On Mon, Apr 11, 2011 at 01:44:36PM -0700, Brock Pytlik wrote:
On 04/11/11 01:33 PM, Edward Pilatowicz wrote:
On Wed, Mar 30, 2011 at 09:36:05PM -0700, Brock Pytlik wrote:
Here's the system repository work that Tim and I have been working
on for quite a while now.
http://cr.opensolaris.org/~bpytlik/ips-sysrepo-v1/
some initial comments below.
ed
---------
src/brand/attach
src/brand/common.ksh
src/brand/pkgcreatezone
- i don't see the zones proxy included in this wad. if that's the case
then i don't think we can remove the publisher syncing logic just yet
because without the zones proxy pkg(1) inside a zone would break
because it wouldn't be able to access the system repo. (have you
tried running the pkg(1) command within a zone with these changes?)
Dan asked me about this to, and while I do remember testing these, I
don't remember to what extent. What I think I did was test that I
could install and update a zone. What I'd like to think I did was
then, from the global zone, use -R to reference the zone created,
and make sure the right things happened. I can't swear with 100%
certainty that I did that though.
the case i'm worried about is not pkg -R from the global zone (i'm
assuming that will work since the sysrepo will be available. it's the
zlogin(1) into a zone and then run the pkg(1) command that i think will
have problems (since the sysrepo won't be accessible).
I understand and that's correct. However, if pkg -R from the global zone
didn't work, then it's highly unlikely that entering the zone and doing
pkg commands would work either. Just wanted to make it clear that
everything except the actual zone -> GZ transport had been tested.
Brock
[snip]
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
