feat: recover experimental Googlevideo media autonomously

Preserve operator-owned profiles while rechecking fresh media and bounded native relearning in the deployed daemon. Keep mixed safety failures fatal and expose active coverage separately from qualification. Document source-only acceptance and retain existing recovery budgets and privacy boundaries.

Co-authored-by: GPT-6 <gpt-6@openai.com>
This commit is contained in:
ZeusinaandGPT-6 Bot committed 2026-10-12 20:19:16 +03:00
1 parent 1a35cb3019
commit a60f62de22
14 files changed
+878 -41

No files matched your search

+22 -4
View File
@@ -419,7 +419,25 @@ do not blindly retry it. Use the matched whole-state/image rollback above.
Candidate-ID CLI apply cannot select/inherit the experimental policy; explicit
native-circular authorization is required. Internal journal recovery is unchanged.
Resource probes still run. This policy remains `DEGRADED`/GGC-unresolved even
when a bounded media check succeeds: neither playback READY nor family
qualification is inferred. Automatic frontend/image repairs preserve the
operator suffix; automatic GGC recovery does not replace or qualify it.
Resource probes still run. In service mode (legacy or adaptive), healthy actual
image/media probes short-circuit recovery, but operational health does not mean
family qualification or playback READY. Experimental media recovery preserves
the operator suffix and all committed profile IDs: a validated transfer failure
permits one guarded restart of that same deployed daemon, then at most 10 native
relearning attempts within 120s and a final fresh signed media 3/3 for each of
TLS1.2 and TLS1.3. Learning stays in the persistent daemon's RAM; no temporary
NFQ worker or replacement/qualified candidate is created. Expiring media is
rediscovered, never rewritten or persisted. Missing inventory remains degraded,
not strategy absence; mixed, guard, infrastructure and cancellation errors do
not authorize relearning. Unrelated missing image inventory may coexist with
usable media, but successful media recovery does not repair or qualify it.
`managed status` reports active-profile `covered=` separately from qualified
`groups` and `unresolved_groups`; `managed active-ids` adds `covered_groups` and
`googlevideo_scope` to its JSON. An experimental Googlevideo profile can be
covered yet qualification-unresolved; one image profile does not establish full
asset-inventory success. Status is read-only and reports live health NOT CHECKED.
The new autonomous recovery/classifier/status behavior is source-only pending
field checks, not a deployment or production-acceptance claim. Generic production
media behavior and signed-URL/SS secret privacy remain unchanged, as do the
persisted minimum two-hour strategy-absence pause and 90-minute whole-job budget.