chore(specs): retire scheduler-config-version-tag as superseded

Its false-ERROR-on-reconnect symptom was already fixed by bug-1854 and
bug-1886; today's HA soak shows zero scheduler ERRORs on reconnect.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This commit is contained in:
Rongsong Shen 2026-09-29 13:03:15 +00:00
parent c2b3bf6473
commit b01eb3a37c
Signed by untrusted user: rshen
GPG key ID: 80533536B0EB651C

View file

@ -1,9 +1,9 @@
{
"feature_name": "scheduler-config-version-tag",
"created_at": "2026-08-28T10:41:46Z",
"updated_at": "2026-08-28T11:20:00Z",
"updated_at": "2026-09-29T13:03:15Z",
"language": "en",
"phase": "tasks-generated",
"phase": "superseded",
"approvals": {
"requirements": {
"generated": true,
@ -18,7 +18,7 @@
"approved": true
}
},
"ready_for_implementation": true,
"ready_for_implementation": false,
"gitea": {
"milestone_id": 45,
"issues": {
@ -26,5 +26,6 @@
"implementation": 199,
"validation": 200
}
}
},
"superseded_note": "Retired 2026-09-29 without implementation. The motivating symptom -- a burst of false ERROR log lines on every HA-failover scheduler reconnect -- was removed by bug-1854 (663ea63c: resourceApplyDef treats an identical re-pushed resource as RES_APPLY_SAME, logged at DEBUG) and bug-1886 (3e873f0a: a re-applied config updates queues in place). The 2026-09-29 3-round HA soak (xce-bug1885-soak) logged 0 scheduler ERROR entries across 163 join/reconnect lines. The remaining benefit (skipping redundant re-application) did not justify the new risk of skipping the shared resvals an unchanged push carries."
}