Skip to content

Conversation

@HosseinKaviani-H
Copy link
Contributor

Changes

  1. Added missing hyperactor import (line 28)

    • from monarch.tools.components import hyperactor
    • Required for hyperactor.host_mesh() call in get_allocator()
  2. Changed SLURM transport to TcpWithHostname (line 131)

    • Changed from ChannelTransport.Tcp to ChannelTransport.TcpWithHostname
    • Fixes connectivity issues in SLURM multi-node environments where hostname resolution is required

@meta-cla meta-cla bot added the CLA Signed This label is managed by the Meta Open Source bot. label Nov 17, 2025
Co-authored-by: Felipe Mello <fmellomascarenhas@gmail.com>
@HosseinKaviani-H HosseinKaviani-H merged commit 6177da8 into meta-pytorch:main Nov 18, 2025
10 checks passed
mreso pushed a commit to mreso/torchforge that referenced this pull request Nov 19, 2025
…nsport (meta-pytorch#582)

Co-authored-by: Hossein Kavianihamedani <hosseinkh@fb.com>
Co-authored-by: Allen Wang <9057208+allenwang28@users.noreply.github.com>
Co-authored-by: Felipe Mello <fmellomascarenhas@gmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CLA Signed This label is managed by the Meta Open Source bot.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants