-
v1.7.15 Stable
Administrator released this
2026-03-07 17:15:15 +01:00 | 363 commits to main since this releaseFix hosterNotAvailable provider cooldown stalling auto-retry
Bug Fixes
-
Fix provider cooldown escalation on hosterNotAvailable errors: The inner unrestrict error handler was still calling
recordProviderFailure()for hosterNotAvailable errors, causing provider-level cooldown to escalate (up to 180 seconds). This made items stall at "Unrestrict-Fehler, Retry X/inf" indefinitely while manual reset ("Zurücksetzen") worked instantly. The root cause was that hosterNotAvailable is a hoster-side issue, not a provider failure, so it should not penalize the provider. -
Dedicated hosterNotAvailable retry path: Added a specialized error handler that:
- Clears
item.providerfor fresh provider selection on retry (mimicking manual reset behavior) - Does NOT call
recordProviderFailure()— avoids provider cooldown escalation - Caps backoff at 30 seconds (5s base with 1.5x growth, max 30s) instead of the previous 5s to 120s+ escalation
- Properly cleans up partial downloads before retry
- Clears
-
Inner catch no longer penalizes providers for hoster issues: The catch block that records provider failures during unrestrict now explicitly excludes hosterNotAvailable errors, preventing the failure counter from growing and triggering shelve/cooldown mechanisms
Downloads
-