max / makenotwork

pom: note that the CI host's timezone is load-bearing astra moved from America/Los_Angeles to America/Denver, matching fw13. goingson's a_relative_event_keeps_its_wall_clock_across_a_move asserts a civil time that only holds at UTC-6, and it reads /etc/localtime rather than TZ, so exporting TZ from run-ci.sh does not move it. With the host on Denver the goingson suite is green. The test's dependency on the host clock is a real fragility and is not fixed by this; any contributor outside Mountain time hits it too.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

Author: Max Johnson <me@maxj.phd> · 2026-08-06 21:55 UTC

Signed with PGP, not checked

Commit: 294355cde9fc23305c721c5d4464430272efe0c5

Parent: c305c07

1 file changed, +8 insertions, -0 deletions

M pom/deploy/README.md +8

			@@ -101,6 +101,14 @@
101	101		compiling against the committed `.sqlx` cache, which goes stale silently whenever
102	102		a migration lands without a `cargo sqlx prepare`.
103	103
	104	+	The host timezone is load-bearing. astra was on `America/Los_Angeles` and is
	105	+	now on `America/Denver`, matching fw13. goingson's
	106	+	`a_relative_event_keeps_its_wall_clock_across_a_move` asserts a civil time that
	107	+	only holds at UTC-6, and it reads `/etc/localtime` rather than `TZ`, so exporting
	108	+	`TZ` in this script does not move it — only the host setting does. That is a
	109	+	latent fragility in the test worth fixing at the source; until then a CI host on
	110	+	any other zone reports goingson red for no reason.
	111	+
104	112		To reproduce a red suite by hand:
105	113
106	114		```