Bug 1130227 - storage proposal weirdness
storage proposal weirdness
Status: CONFIRMED
Classification: openSUSE
Product: openSUSE Distribution
Classification: openSUSE
Component: YaST2
Leap 15.1
Other Other
: P5 - None : Normal (vote)
: ---
Assigned To: YaST Team
Jiri Srain
https://trello.com/c/5ql4ijMp/
:
Depends on:
Blocks:
  Show dependency treegraph
 
Reported: 2019-03-22 14:20 UTC by Steffen Winterfeldt
Modified: 2019-03-25 07:23 UTC (History)
1 user (show)

See Also:
Found By: ---
Services Priority:
Business Priority:
Blocker: ---
Marketing QA Status: ---
IT Deployment: ---


Attachments
test case (2.87 KB, application/x-ruby)
2019-03-22 14:22 UTC, Steffen Winterfeldt
Details
y2log (91.39 KB, text/plain)
2019-03-22 14:23 UTC, Steffen Winterfeldt
Details
rspec output (1.80 KB, text/plain)
2019-03-22 14:25 UTC, Steffen Winterfeldt
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Steffen Winterfeldt 2019-03-22 14:20:18 UTC
While reviewing some test scripts for bug 1094927 I came across this one

https://github.com/yast/yast-storage-ng/blob/master/test/y2storage/initial_guided_proposal_test.rb#L200-L213

It passes but the proposed layout is wrong.

With the fix for bug 1094927 it passes and the layout is correct so this looks
like a non-issue - BUT the way it failed originally is interesting and might
point to a weakness in the proposal process.

I'll attach the singled-out test case and the y2log it originally produces.
Comment 1 Steffen Winterfeldt 2019-03-22 14:22:38 UTC
Created attachment 800967 [details]
test case
Comment 2 Steffen Winterfeldt 2019-03-22 14:23:08 UTC
Created attachment 800968 [details]
y2log
Comment 3 Steffen Winterfeldt 2019-03-22 14:25:22 UTC
Created attachment 800969 [details]
rspec output
Comment 4 Steffen Winterfeldt 2019-03-22 14:33:37 UTC
As you see in the log the problem is that (as root is on gpt) it needs a
bios_boot partition. But due to circumstances it tries to create it on some
other disk which happens to be msdos partitioned.

This fails, which is ok - but this also leaves a good-for-nothing half-finished
/dev/sdc1 partition in the proposal. Which is imho not ok.

It will not happen again in this particular case with the fix for the above
mentioned bug but I think maybe we are not correctly backing out of errors
generally.
Comment 5 Steffen Winterfeldt 2019-03-22 14:38:07 UTC
Just noted a minor incorrectness: with the new code it also does not pass but
for another reason - the layout is correct, however.