So with my push, I broke all of the target_controller tests. They broke
because we don't have a fully populated DOC for the target controller to
walk through (since they're unittests).
Can I get a quick one-line review of the fix?
diff -r b67956b28780 usr/src/lib/install_target/controller.py
--- a/usr/src/lib/install_target/controller.py Fri Aug 05 15:44:21 2011
-0600
+++ b/usr/src/lib/install_target/controller.py Fri Aug 05 16:22:25 2011
-0600
@@ -539,7 +539,7 @@
return
# force a VTOC label and update the disk's geometry and dev_size
- if disk.label == "VTOC":
+ if disk.label == "VTOC" and not self._dry_run:
# look for this disk in the DISCOVERED target tree to see
if the
# label changed from EFI to VTOC
for discovered_disk in self._discovered_disks:
The tests are run with dry_run set to True so we never try to walk
self._discovered_disks only to find that it's actually set to None.
Thanks!
-Drew
On 8/5/11 2:58 PM, Drew Fisher wrote:
Good afternoon!
Could I please get an eye or two on:
https://cr.opensolaris.org/action/browse/caiman/drewfish/7075669/webrev/
7075669 <http://monaco.us.oracle.com/detail.jsf?cr=7075669>
controller.py shouldn't try to relabel a disk already labeled with a
VTOC label
Karen found this little gem for me with the fix for her P1. I tested
this within a vbox guest using the same ISO Karen created. I tested
setting the disk's label with both VTOC and EFI before starting
text-installer. Both installers successfully completed target
instantiation and proceeded with install.
Thanks!
-Drew
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
