Skip to content

omamobile-audio: the watchdog's c9 rise test no longer races the clock - #315

Merged
derekross merged 1 commit into
mainfrom
audio-watchdog-test-race
Nov 2, 2026
Merged

derekross merged 1 commit into
mainfrom
audio-watchdog-test-race

Conversation

@derekross

Copy link
Copy Markdown
Owner

The omamobile-audio watchdog test "in stage c9 a rise of 8 C above the measured value mutes" was timing-sensitive, and on 2026-11-01 alone it failed three times: on #256, on #313, and on main.

It also blocked the development channel. The nightly publish on the desktop builds and tests every package first. This test aborted it every night, so pkgs.omamobile.org/development is still at omamobile-shell 0.2.45, while main is at 0.2.75.

Cause: the "measured" 33 °C was set 0.2 s after the watchdog started, and the baseline was expected at 0.6 s. Under load the step landed after the baseline was taken, so the watchdog saw 25 -> 41 °C and the grep for "33.0 -> 41.0" failed.

Fix: the step now lands at once, the baseline is taken at 1.0 s and the rise comes at 2.0 s. That leaves a second's margin either side, with no change to the watchdog itself. Passed three runs in a row locally.

🤖 Generated with Claude Code

https://claude.ai/code/session_01BeEcdVeSfyLCbijF8XQF1d

The test set the "measured" 33 C 0.2 s after the watchdog started and
expected its baseline at 0.6 s. On a loaded machine the step sometimes
landed after the baseline was taken, so the watchdog saw 25 -> 41 C
instead of 33 -> 41 C and the test failed for timing alone. It failed
three times on 2026-11-01 (pull requests #256, #313, and main), and it
aborted every nightly development publish, which builds and tests every
package first: pkgs.omamobile.org/development stayed at omamobile-shell
0.2.45. The step now lands at once, the baseline is taken at 1.0 s and
the rise comes at 2.0 s, a second's margin either side. No change to the
watchdog itself.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BeEcdVeSfyLCbijF8XQF1d
@derekross
derekross merged commit ff2eed1 into main Nov 2, 2026
2 checks passed
@derekross
derekross deleted the audio-watchdog-test-race branch November 2, 2026 03:30
derekross added a commit that referenced this pull request Nov 2, 2026
)

* omamobile-audio: give the watchdog's c9 settle tests a 10 s window

#315 moved the c9 rise test's steps apart, but CI still failed it
(#319's run): the watchdog logged its start and nothing else, because on
the slower runner the steps arrived after its 80 polls (4 s) were over.
The c9 settle tests now poll 200 times (10 s at 0.05 s). A test that
expects a mute still ends at the mute.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BeEcdVeSfyLCbijF8XQF1d

* omamobile-audio: the c9 rise test starts at the measured coil, before the watchdog runs

Still failing on CI with the 10 s window: the test set its starting
temperature only after it saw the watchdog's "watching" line, and the
watchdog's output to a file is buffered, so on CI the line came late and
the baseline was taken from the earlier 25 C. The watchdog then muted on
"25 -> 33" and the test's check for "33 -> 41" failed. The rise test now
sets the measured 33 C before the watchdog starts, so the baseline can't
depend on when its output shows; the step from the reference to the
measured value in the settle window stays covered by the c9-step test.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BeEcdVeSfyLCbijF8XQF1d

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
robertzas pushed a commit to robertzas/omarchy-mobile that referenced this pull request Nov 14, 2026
derekross#315)

The test set the "measured" 33 C 0.2 s after the watchdog started and
expected its baseline at 0.6 s. On a loaded machine the step sometimes
landed after the baseline was taken, so the watchdog saw 25 -> 41 C
instead of 33 -> 41 C and the test failed for timing alone. It failed
three times on 2026-11-01 (pull requests derekross#256, derekross#313, and main), and it
aborted every nightly development publish, which builds and tests every
package first: pkgs.omamobile.org/development stayed at omamobile-shell
0.2.45. The step now lands at once, the baseline is taken at 1.0 s and
the rise comes at 2.0 s, a second's margin either side. No change to the
watchdog itself.

Claude-Session: https://claude.ai/code/session_01BeEcdVeSfyLCbijF8XQF1d

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
robertzas pushed a commit to robertzas/omarchy-mobile that referenced this pull request Nov 14, 2026
…erekross#321)

* omamobile-audio: give the watchdog's c9 settle tests a 10 s window

derekross#315 moved the c9 rise test's steps apart, but CI still failed it
(derekross#319's run): the watchdog logged its start and nothing else, because on
the slower runner the steps arrived after its 80 polls (4 s) were over.
The c9 settle tests now poll 200 times (10 s at 0.05 s). A test that
expects a mute still ends at the mute.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BeEcdVeSfyLCbijF8XQF1d

* omamobile-audio: the c9 rise test starts at the measured coil, before the watchdog runs

Still failing on CI with the 10 s window: the test set its starting
temperature only after it saw the watchdog's "watching" line, and the
watchdog's output to a file is buffered, so on CI the line came late and
the baseline was taken from the earlier 25 C. The watchdog then muted on
"25 -> 33" and the test's check for "33 -> 41" failed. The rise test now
sets the measured 33 C before the watchdog starts, so the baseline can't
depend on when its output shows; the step from the reference to the
measured value in the settle window stays covered by the c9-step test.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01BeEcdVeSfyLCbijF8XQF1d

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant