[email protected] wrote:
On Wed, Jul 08, 2009 at 05:21:22PM -0700, Brock Pytlik wrote:
[email protected] wrote:
Folks,
The new transport made sure that we always ran the captive portal test,
if it hadn't already been executed, prior to performing a network
operation. However, in some cases we really don't care whether the
network operation succeeds or fails -- it's purely informational.
In those cases, running the test prohibited users from performing
installs/uninstalls when all of the content was already on disk.
With the new fix, I've verified that it's once again possible to
install/uninstall with all bits present on local disk. The webrev is
below. (It's small).
http://cr.opensolaris.org/~johansen/webrev-9929/
Thanks,
-j
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
I don't really see a reason to make captive_portal a wrapper. The check
starting on line 396 might get moved up to 372 so that the directories
aren't make if the captive portal check fails.
Other than that, LGTM.
Ok, thanks. I've incorporated comments from you and from dduvall.
The new webrev is here:
http://cr.opensolaris.org/~johansen/webrev-9929-2/
Overall, looks fine.
Although, a test case would be nice, I swear we used to have one.
It should be simple:
1) start a depot
2) install a package
3) stop the depot
4) uninstall a package
Cheers,
--
Shawn Walker
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
