Hi Matt
Thanks for the review.
I have modified the fix. Please find the updated webrev :
https://cr.opensolaris.org/action/browse/caiman/nirmal27/7143082-2/webrev/
I have tested with the following scenario's :
--> traditional whole_disk=true, with no partition/slice children
specified at all
--> whole_disk=true with partitions/slice specified to cause conflict
--> whole_disk=false with partition/slice specified to cause conflict,
which you do not ignore
Thanks
Nirmal
On 02/21/12 15:21, Matt Keenan wrote:
Nirmal,
Just slightly concerned that we should not be ignoring the
SliceInUseError in all scenarios.
If wipe disk scenario is not in place, and the manifest attempts to
provide a layout, then SliceInUseError's I think are actually valid. In
this scenario AI will attempt to merge the manifest specification with
what is already on the disk, and attempting to overwrite a slice already
in use is a valid error.
Have you tested this with manifest that has whole-disk=false ?
cheers
Matt
On 02/20/12 18:00, Nirmal Agarwal wrote:
Hi all
Could I please get a code review for the following CR:
7143082 In Solaris11 AI, specifying <disk whole_disk="true"> in
ai_manifest will not wipe out disk
Webrev :
https://cr.opensolaris.org/action/browse/caiman/nirmal27/7143082/
-- Source is pep8 clean.
slim_test result
----------------
/net/indiana-build.us.oracle.com//export/home/na210770/ai/7143082/slim_test
Manual Tests :
-- created a custom image and tested the fix with the manifest using
<disk whole_disk="true"> tag. AI install succeeds.
Let me know if I need to run some other tests.
Thanks
Nirmal
_______________________________________________
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
