Cannot install torch 1.10.2 with poetry 1.4.1 on python:3.8-slim linux/amd64 docker image

[malanEvans]

• Poetry 1.4.1:
• When building on python:3.8-slim linux/amd64 docker base image:
• pyproject.toml (Dockerfile is attached as well):

• I am on the latest stable Poetry version, installed using a recommended method.
• I have searched the issues of this repo and believe that this is not a duplicate. [()](Poetry install succeeds on macOS but fails in Docker image with error "*.whl not found in known hashes" #7099) looks similar, but not the same.
• I have consulted the FAQ and blog for any relevant entries or release notes.
• If an exception occurs when executing a command, I executed it again in debug mode (-vvv option) and have included the output below.

Issue

When the attached Dockerfile is used to build an image with following command,
docker build .
from the directory with pyproject.toml file, 'RUN poetry install' step fails with the following error,

` 
  _WheelFileValidationError
 
  ["In /root/.cache/pypoetry/artifacts/c8/6e/03/5c824b6564a6d8451e0fdfc8843110e61a5342f7aa62dcd8fc0b05a83e/torch-1.10.2-cp39-cp39-manylinux1_x86_64.whl, hash / size of torch-1.10.2.dist-info/METADATA didn't match RECORD"]
 
  at /opt/venv/lib/python3.9/site-packages/installer/sources.py:289 in validate_record
      285│                         f"In {self._zipfile.filename}, hash / size of {item.filename} didn't match RECORD"
      286│                     )
      287│
      288│         if issues:
    → 289│             raise _WheelFileValidationError(issues)
      290│
      291│     def get_contents(self) -> Iterator[WheelContentElement]:
      292│         """Sequential access to all contents of the wheel (including dist-info files).
      293│
` 

this works well when poetry==1.4.0 is used.
This works well even with poetry==1.4.1 if docker image is built on arm64v8/python base image

[radoering]

Duplicate, see #7686

[admackin]

Thanks @radoering – we were skeptical but it looks like this is correct. So you're saying that incredibly widely-used python ML package Pytorch has had broken intel binaries in version 1.10.2 this whole time? It is not entirely clear what error to report to Pytorch maintainers, since it's been working with pip and poetry (and other packaging tools presumably) up until now.

[dimbleby]

the error message says exactly what's wrong with the wheel and it sounds as though you have verified it yourself, this should be an easy bug report

[github-actions]

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.