apport hook tags too many bugs regression-update
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
xdiagnose (Ubuntu) |
Fix Released
|
Undecided
|
Unassigned | ||
Oneiric |
Fix Released
|
Undecided
|
Unassigned | ||
Precise |
Fix Released
|
High
|
Unassigned |
Bug Description
The stable release update team has subscribed to bug reports that are tagged either regression-proposed or regression-update and has noticed that quite a few bugs are tagged regression-update which should not be.
The question in xdiagnose that leads to tagging bugs regression-update is a bit ambiguous:
"The problem began right after system software update"
This could mean either installing a package from -updates or happened after upgrading from one release to another. Due to the quantity of false positives the best thing seems to be just to remove the question.
This should be SRU'ed for both Oneiric and Precise.
TEST CASE:
1) run ubuntu-bug xorg
2) At the first dialog box you will see a radio button labelled:
'Continue. The problem began right after system software update.'
In the proposed version of the package this question should not appear.
REGRESSION POTENTIAL:
This question just affects the tags being added to the bug and no regression in functionality of the apport hook is possible.
Changed in xdiagnose (Ubuntu Precise): | |
status: | New → Fix Committed |
importance: | Undecided → High |
Changed in xdiagnose (Ubuntu Oneiric): | |
status: | New → Fix Committed |
description: | updated |
Fixed by just dropping the question.
This removes the feature for indicating possible regressions, so if that's useful to anyone it'll need to be identified some other way.