Switch autostart event in favor of only start#758
Switch autostart event in favor of only start#758ben-grande wants to merge 1 commit intoQubesOS:mainfrom
Conversation
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #758 +/- ##
==========================================
- Coverage 70.74% 70.54% -0.20%
==========================================
Files 61 61
Lines 13809 13807 -2
==========================================
- Hits 9769 9740 -29
- Misses 4040 4067 +27
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:
|
ben-grande
force-pushed
the
preload-autostart-refill
branch
from
4207b00 to
48e3c99
Compare
OpenQA test summaryComplete test suite and dependencies: https://openqa.qubes-os.org/tests/overview?distri=qubesos&version=4.3&build=2025121419-4.3&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=2025111104-4.3&flavor=update
Failed tests12 failures
Fixed failuresCompared to: https://openqa.qubes-os.org/tests/158999#dependencies 18 fixed
Unstable testsDetailsPerformance TestsPerformance degradation:11 performance degradations
Remaining performance tests:84 tests
|
ben-grande
force-pushed
the
preload-autostart-refill
branch
2 times, most recently
from
8c987f4 to
a9fa2d7
Compare
Now it is tested... and working. |
ben-grande
force-pushed
the
preload-autostart-refill
branch
from
a9fa2d7 to
7a4e1ad
Compare
|
Why fail with me OpenQA... I don't know, the preloads are taking too much time to get ready for that test. I'd have to see the machine live or get the output of Increased the timeout from 40s to 60s. |
I saw openqa disp* logs, but didn't identify the problem. |
I'm not sure if feasible, but maybe you can extend test with collecting such info on failure? Or is relevant VM killed already at that point? |
ben-grande
force-pushed
the
preload-autostart-refill
branch
from
7a4e1ad to
4d5728e
Compare
Tests passed when timeout increased. VM is not killed at that point. The performance tests run fine with the expected values, just the CI one that times out. The issue this PR attempted to solve was that a 2 preload switch was too much, but in some way, I ended up producing a similar test when cleaning up manually 2 preloaded disposables and then running the autostart script to preload 2. Looking again at the logs, |
ben-grande
force-pushed
the
preload-autostart-refill
branch
2 times, most recently
from
c21f900 to
20de5d8
Compare
The autostart made a refresh, which cleaned all preloaded disposables, served to: - Invalidate previous incomplete remnants - Get fresh ones after a change was made to a template in chain Both of these issues were handled, "domain-load" of the disposable template invalidates incomplete ones while "domain-shutdown" of a template in chain triggers a refresh. For: QubesOS/qubes-issues#1512
ben-grande
force-pushed
the
preload-autostart-refill
branch
from
20de5d8 to
b474909
Compare
|
Superseded by #760 |
3 participants
The autostart made a refresh, which cleaned all preloaded disposables, served to:
Both of these issues were handled, "domain-load" of the disposable template invalidates incomplete ones while "domain-shutdown" of a template in chain triggers a refresh.
For: QubesOS/qubes-issues#1512