Closed wesm closed 7 years ago
ocefpaf
commented
7 years ago So the build 3 with pinned jemalloc is there, but the dependency resolver isn't doing the right thing, or I may not understand how the dependency pinning works
That is quite painful. Anyways, I move all jemalloc 5.0.0 to a "broken" label. Let me know if that works for you.
wesm
commented
7 years ago @ocefpaf thank you! seems to be back working now
$ conda create -n arrow-conda-test python=3.6 pyarrow
Fetching package metadata ...............
Solving package specifications: .
Package plan for installation in environment /home/wesm/anaconda3/envs/arrow-conda-test:
The following NEW packages will be INSTALLED:
arrow-cpp: 0.4.1-np113py36_3 conda-forge
ca-certificates: 2017.4.17-0 conda-forge
certifi: 2017.4.17-py36_0 conda-forge
jemalloc: 4.4.0-0 conda-forge
mkl: 2017.0.1-0 defaults
ncurses: 5.9-10 conda-forge
numpy: 1.13.0-py36_0 defaults
openssl: 1.0.2k-0 conda-forge
pandas: 0.20.2-np113py36_1 conda-forge
parquet-cpp: 1.2.0.pre-0 conda-forge
pip: 9.0.1-py36_0 conda-forge
pyarrow: 0.4.1-np113py36_0 conda-forge
python: 3.6.1-3 conda-forge
python-dateutil: 2.6.0-py36_0 conda-forge
pytz: 2017.2-py36_0 conda-forge
readline: 6.2-0 conda-forge
setuptools: 33.1.1-py36_0 conda-forge
six: 1.10.0-py36_1 conda-forge
sqlite: 3.13.0-1 conda-forge
tk: 8.5.19-1 conda-forge
wheel: 0.29.0-py36_0 conda-forge
xz: 5.2.2-0 conda-forge
zlib: 1.2.11-0 conda-forge
Proceed ([y]/n)? nNo problem.
@jakirkham @ocefpaf these builds are broken due to a glibc incompatibility. We are discussing with the jemalloc developers how to resolve this (https://github.com/jemalloc/jemalloc/issues/937), but in the meantime, despite our efforts to pin the jemalloc version at 4.4.0 for some reason conda still wants to install 5.0.0
in contrast
So the build 3 with pinned jemalloc is there, but the dependency resolver isn't doing the right thing, or I may not understand how the dependency pinning works