fix(library): widen heavy library-fetch timeout to 30s (issue #41) - #90
Open
ndandan wants to merge 2 commits into
Open
fix(library): widen heavy library-fetch timeout to 30s (issue #41)#90ndandan wants to merge 2 commits into
ndandan wants to merge 2 commits into
Conversation
) The full-library endpoints (/api/v3/series, /api/v3/movie) are fetched via the shared get(), which capped CURLOPT_TIMEOUT at 4s (Sonarr) / 8s (Radarr). On a large or busy instance the response body doesn't finish in that window — libcurl reports "Operation timed out after 4002 ms with 0 bytes received" (TCP connected, no body yet), which trips the circuit breaker and surfaces the service as "unreachable/disconnected". Give only the two heavy full-library calls a 30s budget; connect timeout stays 4s so a genuinely-down service still trips the breaker fast. Radarr get() gains an optional $timeout param (default 8s) so all other call sites are unchanged. The library route already allows set_time_limit(120) and the MediaLibraryCache means only the cold fetch pays this cost. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
… scope the 30s budget to library routes Review follow-ups to the Shoshuo#41 fix: - Root cause: get()/multiGet() ran the same markDown branch for a read timeout after a successful connect as for a connection failure, so a fetch that outlived ANY budget still marked the whole instance down and short-circuited every caller for the breaker window — the 30s bump only moved that threshold. The two are now distinguished via curl_errno + CURLINFO_CONNECT_TIME: slow-payload timeouts fail the call (and still short-circuit the remainder of the same request) but never open the cross-request breaker; connect-level failures behave as before. - getRawMovies() had been left on the 8s default while its Sonarr twin was widened — the raw sweep paths kept reproducing Shoshuo#41 on Radarr. - The 30s budget was baked into getMovies()/getSeries(), so dashboard widgets, quick-look membership checks and the admin statistics loop all inherited it (worst case 30s x N instances in one request). The full- library fetchers now take a per-call timeout defaulting to the old 8s/4s; the library/collections/filter/search-index routes pass the new public LIBRARY_TIMEOUT (30) explicitly. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The full-library endpoints (/api/v3/series, /api/v3/movie) are fetched
via the shared get(), which capped CURLOPT_TIMEOUT at 4s (Sonarr) / 8s
(Radarr). On a large or busy instance the response body doesn't finish
in that window — libcurl reports "Operation timed out after 4002 ms with
0 bytes received" (TCP connected, no body yet), which trips the circuit
breaker and surfaces the service as "unreachable/disconnected".
Give only the two heavy full-library calls a 30s budget; connect timeout
stays 4s so a genuinely-down service still trips the breaker fast. Radarr
get() gains an optional $timeout param (default 8s) so all other call
sites are unchanged. The library route already allows set_time_limit(120)
and the MediaLibraryCache means only the cold fetch pays this cost.
Co-Authored-By: Claude Opus 4.8 noreply@anthropic.com