Clean preloaded disposable remnants on next boot #742
Conversation
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #742 +/- ##
==========================================
- Coverage 70.40% 70.38% -0.02%
==========================================
Files 61 61
Lines 13682 13691 +9
==========================================
+ Hits 9633 9637 +4
- Misses 4049 4054 +5
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
OpenQA test summaryComplete test suite and dependencies: https://openqa.qubes-os.org/tests/overview?distri=qubesos&version=4.3&build=2025102219-4.3&flavor=pull-requests Test run included the following:
New failures, excluding unstableCompared to: https://openqa.qubes-os.org/tests/overview?distri=qubesos&version=4.3&build=2025081011-4.3&flavor=update
Failed tests12 failures
Fixed failuresCompared to: https://openqa.qubes-os.org/tests/149225#dependencies 84 fixed
Unstable testsPerformance TestsPerformance degradation:17 performance degradations
Remaining performance tests:163 tests
|
qubes/vm/dispvm.py
Outdated
| :param bool force: Force cleanup even if enabled, else it might \ | ||
| be handled by ``domain-shutdown``. |
There was a problem hiding this comment.
Leftover from earlier version? This function doesn't have force parameter
There was a problem hiding this comment.
Yes, the remnant PR had remnants... It was dealt with the running check.
If the disposable was preloaded, in progress feature would return False, now it check is qube was a preloaded disposable that completed but is not running to signal improper shutdown. When calling cleanup, if the qube is not running, the "_bare_cleanup()" was skipped on "cleanup()", now this is handled. There is something that triggers "domain-remove-from-disk" on qubesd start for unnamed disposables, didn't detect the origin, but it is now handled by removing the preloaded disposable from the list. Fixes: QubesOS/qubes-issues#10326 For: QubesOS/qubes-issues#1512
ben-grande
force-pushed
the
preload-dirty-shutdown
branch
from
5b6d695 to
3e842d7
Compare
| running = False | ||
| # Full cleanup will be done automatically if event 'domain-shutdown' is | ||
| # triggered and "auto_cleanup=True". | ||
| if not running or not self.auto_cleanup: |
There was a problem hiding this comment.
Some strange issue appears when using not running here.
The not running statement was intended to deal with halted qube (remnant from previous boot). I guess it is better to accept a force parameter to not default to running state.
There was a problem hiding this comment.
But still doesn't explain the issue... was cleanup called twice?
3 participants
If the disposable was preloaded, in progress feature would return False, now it check is qube was a preloaded disposable that completed but is not running to signal improper shutdown.
When calling cleanup, if the qube is not running, the "_bare_cleanup()" was skipped on "cleanup()", now this is handled.
There is something that triggers "domain-remove-from-disk" on qubesd start for unnamed disposables, didn't detect the origin, but it is now handled by removing the preloaded disposable from the list.
Fixes: QubesOS/qubes-issues#10326
For: QubesOS/qubes-issues#1512
How can I simulate stopping qubesd on integration tests? I tried to do unittest and firing
domain-load, but I could get the magic right.