Hacktoberfest 2026:メンテナが10月に向けて印を付けた、オープンで初心者向けの issue。 Hacktoberfest の issue を見る

Catalyst upload tempfile is deleted during request in #1642 integration

オープン
#1,795 コメント 0 件 リアクション 0 件 担当者 0 名 GitHub で見る

メンテナーはふだん 1 日以内に返信

まだ誰も着手していません。

評価

難易度
4/5
見積もり時間
3〜5日
初心者へのやさしさ
15/100
issue の種類
バグ
明瞭さ
おおむね明確
活発さ
活発
技術スタック
java, perl
領域
backend

調査の方向性

Start with t/aggregate/live_engine_request_uploads.t, the test that fails with 'Upload temp file was deleted' in the integration run. The issue says the work depends on #1642 landing first, so wait for its final commit before reproducing. Done means that test passes on that commit and the lost ownership edge for the File::Temp object is covered by a focused regression test.

索引モデルが issue の本文から書いたものです。

説明

Summary

The #1642 cleanup changes appear to introduce a Catalyst upload lifetime regression. Please triage after #1642 is merged or its implementation settles; this report records integration evidence and avoids duplicating the cleanup implementation.

Reproduction and evidence

In a disposable integration checkout merging PerlOnJava3 PR #1777 (47e9bfb78f1ac790143838fafcf92f852147bc2b) with the #1642 work in PerlOnJava5 PR #1746 (46d9200e5553bc645fe38b4ee341433aff8ae958), the Catalyst release acceptance target reached 3,819/3,820 passing tests and failed:

  • t/aggregate/live_engine_request_uploads.t: Upload temp file was deleted.
  • During global destruction, the runtime also reported DESTROY created new reference to dead object 'File::Temp'.

This failure occurred only in the integration run; the PerlOnJava3 #1777 branch's earlier Catalyst acceptance run passed 3,733 tests. The integration build was b124844c9 (test-only merge commit); neither review branch was changed. Because this is a tempfile lifetime/destructor failure, it may have a different root cause from the Mojolicious request and WebSocket failures; it is tracked separately for that reason.

Expected behavior

A Catalyst upload's temporary file remains available for the documented request lifetime and is cleaned up only after its final owner is gone. Global destruction must not resurrect or dereference an already-dead File::Temp object.

Requested follow-up

After #1642 lands, rerun the named test against its final commit, identify which ownership edge is lost, and add focused permanent regression coverage before changing lifecycle cleanup behavior. The run had no output for 600 seconds while waiting on DBIx::Class's _dbictest_global.lock, so it was stopped by the target soft timeout; no conclusion about DBIx::Class regressions is claimed here.

主要言語
Perl
スター
64
フォーク
7
平均マージ
4時間 57分
マージ済み PR(30日)
203

環境構築

はじめの一歩

  1. issue を最後まで読み、次にプロジェクトのコントリビューションガイドを読みます。
  2. 着手することを issue にコメントします — 二人が同じ作業をするのを防げます。
  3. リポジトリをフォークし、ブランチを切って変更します。
  4. issue 番号を参照したプルリクエストを送ります。

fglock/PerlOnJava のほかの issue

fglock/PerlOnJava の issue をすべて見る

似ている issue

Perl の issue をもっと見る

新しい issue をメールで受け取る

初心者向けの GitHub issue を短くまとめたダイジェスト。