feat(monitor): wire isolated target recovery

Co-authored-by: GPT-6 <gpt-6@openai.com>
This commit is contained in:
ZeusinaandGPT-6 Bot committed 2026-10-10 09:45:34 +03:00
1 parent 2d4e4878ae
commit 5f3a6953ee
16 files changed
+1919 -73

No files matched your search

+31 -13
View File
@@ -21,7 +21,11 @@ binding. It uses full target-local firewall/daemon restart and Player API post-c
atomic rollback on failure. See [MANAGED_TARGET.md](MANAGED_TARGET.md) for guards
and provisioning prerequisites. An isolated pinned deployment definition is in
[deploy/managed-target](deploy/managed-target/README.md); build/run are deferred.
No autonomous loop, remote activation or Xray changes are included. Image
An injected bootstrap/saved-first recovery adapter now connects the persisted
monitor foundation to blockcheck, immutable candidates and transactional apply.
`managed monitor` and the dedicated target entrypoint now wire these adapters;
this increment is not tested by the producer or live/runtime verified.
No remote activation or Xray changes are included. Image
compatibility and route/rollback gates are not yet remotely proven; adapter
tests/build are deferred to the orchestrator.
@@ -35,6 +39,14 @@ and roll back together, including after a crash.
## One-shot Player API health
Internal recovery building blocks enumerate qualified immutable candidates
newest-first with a 1–10 limit and recheck saved arguments on an exact endpoint
through an injected combined TLS checker. Only complete, valid two-TLS 3-pass
nonpassing evidence permits the nonpass result; malformed reports and runner,
filesystem or context failures remain infrastructure errors. These helpers do
not update stored evidence, apply configuration or wire managed recovery/monitor
commands. Their local fixture tests/build are deferred to the orchestrator.
```sh
zapret-autotune health youtube
```
@@ -60,9 +72,9 @@ Configuration is parsed once per health invocation. The proxy must use `socks5`
with a loopback IP or `localhost`, port 1–65535, and no credentials/path/query.
Video IDs must be 11 characters; durations and the failure threshold must be
positive. Explicit empty values are invalid. Interval, consecutive-failure
threshold and cooldown are validated future recovery policy settings only:
**An injected autonomous scheduling core exists, but no monitor command, actual
recovery callback or live managed target is wired yet**.
threshold and cooldown configure `managed monitor` (the one-shot command still
does not schedule anything). **CLI/entrypoint integration exists; tests/build
and live target validation remain deferred.**
The future managed target is a **brand-new prod-like zapret2-ss container in the
test environment only**, not the existing container. Its stable configurable
@@ -105,9 +117,9 @@ their existing target constraint. Database failures emit no options.
Signed URLs are not printed. Scanner diagnostics are retained in restricted
artifact files rather than relayed to the terminal. No configuration is applied,
no profile is marked active, and no service is restarted. The staged pipeline is
scan/propose → guarded manual managed-test activation → future monitoring;
the deployment definition exists but is not deployed; the monitor scheduling core
is internal and **not wired to a command or actual recovery**. Manual
scan/propose → guarded managed-test activation → dedicated target monitoring;
the deployment definition exists but is not deployed. `managed monitor` wires
the scheduler and recovery; live compatibility remains unverified. Manual
proposal-only recovery is available below.
No-argument `tune youtube` remains frontend-only and unchanged.
@@ -190,8 +202,8 @@ or duplicate completion is rejected without changing the lock, token or cooldown
even after a newer acquisition starts. State survives reopening the database.
The internal `runManagedMonitor` supervisor probes immediately and periodically,
persists observations, and invokes injected recovery only at threshold outside
cooldown. Recovery errors are reported without stopping monitoring; DB, lock and
finalization errors are fatal. Cooldown starts when recovery returns, including
cooldown. Unresolved GGC is reported without stopping monitoring; DB, safety,
runner, lock and finalization errors are fatal. Cooldown starts when recovery returns, including
failure or cancellation. Callbacks must honor cancellation and join their children.
Existing target guards run before a lifetime nonblocking `flock` in the private
target state directory. With that exclusive singleton held, one-shot startup
@@ -201,8 +213,14 @@ A live owner remains busy; legacy or unverifiable owners fail closed. Probe
history is preserved. The lock file is never unlinked.
The scheduler core has fake-time tests for scheduling, reset, cooldown,
cancellation, parallel starts, fatal DB errors, reopen and safe startup
reconciliation. **There is no actual recovery adapter,
monitor CLI/entrypoint wiring or live target yet**; container/Xray are untouched.
reconciliation. Internal injected managed recovery and monitor callbacks now
support saved-first frontend checks, scan fallback only for valid non-pass
evidence, and fresh exact-host GGC recovery with a frontend-only rollback boundary.
Checks remove prior combined reports; missing/malformed reports fail closed.
Scan shortlists must be regular files no older than the scan start's whole second
(production scans use unique temporary paths). Fake-only orchestration tests are
added; verification is deferred. **Monitor CLI/entrypoint wiring exists but no
live target has been deployed**; existing container/Xray are untouched.
Manual managed activation remains separate. Verification of this increment is
deferred to the orchestrator.
@@ -214,8 +232,8 @@ in the same transaction as the existing run. Qualification requires every declar
endpoint to pass exactly 3/3 for both TLS1.2 and TLS1.3, with full success/coverage.
Existing v2 history is unchanged and is not backfilled into independent candidates.
Candidates have no DB active/installed state; managed activation uses its own
durable config journal. Autonomous scheduling is internal only; actual recovery
and live monitoring remain unwired.
durable config journal. Scheduling and recovery are wired only to the dedicated
managed target; live monitoring has not been verified or deployed.
The explicit `tune googlevideo` path adds exact-cache proposals only.
Proposed runs store a SHA-256 fingerprint of validated arguments with separator
whitespace normalized and argument order preserved; repeated proposals retain