[BUG] incompatible protobuf dep version when installing cudf

[pxLi]

Describe the bug
To use conda install cudf=22.08/22.06 will introduce incompatible protobuf version as dependency,

conda install -c conda-forge mamba
mamba install -y -c rapidsai -c rapidsai-nightly -c nvidia -c conda-forge -c defaults cudf=22.08 python=3.8 cudatoolkit=11.2

which gave us protobuf==3.19.4

when we try import cudf, we will get,

>>> import cudf
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/opt/conda/lib/python3.8/site-packages/cudf/__init__.py", line 71, in <module>
    from cudf.io import (
  File "/opt/conda/lib/python3.8/site-packages/cudf/io/__init__.py", line 8, in <module>
    from cudf.io.orc import read_orc, read_orc_metadata, to_orc
  File "/opt/conda/lib/python3.8/site-packages/cudf/io/orc.py", line 14, in <module>
    from cudf.utils.metadata import (  # type: ignore
  File "/opt/conda/lib/python3.8/site-packages/cudf/utils/metadata/orc_column_statistics_pb2.py", line 7, in <module>
    from google.protobuf.internal import builder as _builder

and this also happens to 22.06 intermittently, if we are lucky enough we could get protobuf>=3.20+, but not always

manual run pip install --upgrade protobuf to upgrade protobuf version (3.20+) could resolve this issue.
Seems this is like one issue before #8570

Steps/Code to reproduce bug

conda install -c conda-forge mamba
mamba install -y -c rapidsai -c rapidsai-nightly -c nvidia -c conda-forge -c defaults cudf=22.08 python=3.8 cudatoolkit=11.2

python
import cudf

Expected behavior
import cudf correctly

Environment overview (please complete the following information)
nvidia/cuda:11.2.2-runtime-centos7

[shwina]

Hi @pxLi -- is this installing into a fresh environment or an existing one? Perhaps you could try mamba create rather than mamba install?

[pxLi]

Hi @pxLi -- is this installing into a fresh environment or an existing one? Perhaps you could try mamba create rather than mamba install?

hi, I guess this does not help consistently. ref: #8570 (comment)
Pin the correct version in cudf conda env should help in this case #10996 thx @galipremsagar

[galipremsagar]

resolved in #10996