# PR stuck since a month

**URL:** <https://dev-discuss.pytorch.org/t/pr-stuck-since-a-month/3403>\
**Category:** Uncategorized\
**Created:** [June 19, 2026, 2:03am UTC](https://dev-discuss.pytorch.org/t/pr-stuck-since-a-month/3403 "2026-06-19T02:03:54Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![praful-srinivasan-02](https://yyz2.discourse-cdn.com/flex036/user_avatar/dev-discuss.pytorch.org/praful-srinivasan-02/32/3279_2.png) [@praful-srinivasan-02](https://dev-discuss.pytorch.org/u/praful-srinivasan-02)\
**Post date:** [June 19, 2026, 2:03am UTC](https://dev-discuss.pytorch.org/t/pr-stuck-since-a-month/3403/1 "2026-06-19T02:03:54Z")

</div>

PR #184888 approved and green for a month, needs a land/merge.

---

<div class="post-metadata">

**Author:** ![aorenste](https://yyz2.discourse-cdn.com/flex036/user_avatar/dev-discuss.pytorch.org/aorenste/32/3225_2.png) [@aorenste](https://dev-discuss.pytorch.org/u/aorenste)\
**Post date:** [June 23, 2026, 8:23pm UTC](https://dev-discuss.pytorch.org/t/pr-stuck-since-a-month/3403/2 "2026-06-23T20:23:12Z")

</div>

Hi @praful-srinivasan-02 I just approved/requested a merge. For the future generally once a PR is approved merging is self-service using pytorchbot: [Bot commands · pytorch/pytorch Wiki · GitHub](https://github.com/pytorch/pytorch/wiki/Bot-commands?fbclid=IwAR3vEBs4wC-64tr7xvWCRmEyld3LIbw4fjBUvNj4MoywOCjKaX098CdaszQ)

---

<div class="post-metadata">

**Author:** ![praful-srinivasan-02](https://yyz2.discourse-cdn.com/flex036/user_avatar/dev-discuss.pytorch.org/praful-srinivasan-02/32/3279_2.png) [@praful-srinivasan-02](https://dev-discuss.pytorch.org/u/praful-srinivasan-02)\
**Post date:** [July 7, 2026, 8:25am UTC](https://dev-discuss.pytorch.org/t/pr-stuck-since-a-month/3403/3 "2026-07-07T08:25:07Z")

</div>

Hey @aorenste, the checks were passing earlier, but on repeated CI attempts I’m still getting flaky failures. The failures seem unrelated to my changes (for example, one was a transient SSL/package download failure). Could you please help rerun the failed jobs or take a look?
