FS#62298 - [python-tensorflow-opt-cuda] missing shared library - cuda

Attached to Project: Community Packages
Opened by John Tukey (John_Tukey) - Wednesday, 10 April 2019, 13:33 GMT
Last edited by Konstantin Gizdov (kgizdov) - Thursday, 11 April 2019, 16:51 GMT
Task Type Bug Report
Category Packages
Status Closed
Assigned To Sven-Hendrik Haase (Svenstaro)
Konstantin Gizdov (kgizdov)
Architecture All
Severity Low
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 2
Private No

Details

Description:
After upgrading to 1.13.1-4, I receive this cuda related error:

F tensorflow/stream_executor/lib/statusor.cc:34] Attempting to fetch value instead of handling error Failed precondition: could not dlopen DSO: libcublas.so.10.1; dlerror: libcublas.so.10.1: cannot open shared object file: No such file or directory


Additional info:
* package version: 1.13.1-4

$ pwd
/opt/cuda/targets/x86_64-linux/lib
$ ls -lrt libcublas.so.10.1
lrwxrwxrwx 1 root root 23 Mar 26 23:16 libcublas.so.10.1 -> libcublas.so.10.1.0.105
This task depends upon

Closed by  Konstantin Gizdov (kgizdov)
Thursday, 11 April 2019, 16:51 GMT
Reason for closing:  Fixed
Additional comments about closing:  glibc-2.28-6
cuda-10.1.105-8
Comment by Konstantin Gizdov (kgizdov) - Wednesday, 10 April 2019, 18:41 GMT
Could you please provide the steps to reproduce? Thanks.
Comment by Konstantin Gizdov (kgizdov) - Thursday, 11 April 2019, 10:08 GMT Comment by Konstantin Gizdov (kgizdov) - Thursday, 11 April 2019, 12:59 GMT
Could you try glibc-2.28-6 in [testing] and cuda-10.1.105-8 in [community-testing] and see if it works?

Loading...