[conda] error while loading shared libraries: libnetcdf.so.7: cannot open shared object file: No such file or directory

[Karen Guihou]

Hi all

I am getting the following error when trying to run pynco
(https://github.com/nco/pynco)

*/home/xxx/miniconda2/envs/python_env/bin/ncra: error while loading shared
libraries: libnetcdf.so.7: cannot open shared object file: No such file or
directory*

- I do not have libnetcdf.so.7 in /home/xxx/miniconda2/envs/libs (I
have libnetcdf.so.13)
- I installed pynco by "conda install -c conda-forge pynco". No error
during install.
- I use other libraries linked to libnetcdf and no problem (nco, netcdf4,
numpy, ...)
- Config
platform : linux-64
conda version : 4.3.29
conda is private : False
conda-env version : 4.3.29
conda-build version : not installed
python version : 2.7.14.final.0
requests version : 2.18.4


I have been looking on the web, and a lot of people get this type of error
on Conda, with all sort of libraries. Some people suggest to append a path,
but it is not recommended by the Conda team:
"CAUTION: Setting LD_LIBRARY_PATH on Linux or DYLD_LIBRARY_PATH on macOS
can interfere with this because the dynamic linker short-circuits link
resolution by first looking at LD_LIBRARY_PATH."
(https://conda.io/docs/user-guide/tasks/build-packages/use-shared-libraries.html)

Can anyone clearly explain to me what is the issue, and how to correct it?
Is it a conflict due to a bad install, a path ...? It seem many people
encounter this error and there is not really a clear solution.
Thank you

[Chris Barker]

I am getting the following error when trying to run pynco (
https://github.com/nco/pynco)
*/home/xxx/miniconda2/envs/python_env/bin/ncra: error while loading shared
libraries: libnetcdf.so.7: cannot open shared object file: No such file or
directory*
- I do not have libnetcdf.so.7 in /home/xxx/miniconda2/envs/libs (I
have libnetcdf.so.13)
- I installed pynco by "conda install -c conda-forge pynco". No error
during install.
- I use other libraries linked to libnetcdf and no problem (nco, netcdf4,
numpy, ...)
- Config
platform : linux-64
conda version : 4.3.29
conda is private : False
conda-env version : 4.3.29
conda-build version : not installed
python version : 2.7.14.final.0
requests version : 2.18.4

netcdf in particular can lead to a tangled web ob dependencies, that conda
does not always reasolve. I would try:

create a clean environment to test:

conda create -n pynco python=2

this will create a minimal environment (called pynco) with only python2 and
the stuff conda needs to run.

activate it:

source activate pynco

and now install pynco:

conda install -c conda-forge pynco

That should install pynco and everything it depends on.

see if it works.

If it doesn't file a bug report on conda-forge

If it does, then try installing the other stuff you need and hope they
don't break anything :-)

I recommend putting conda-forge first on your channels path -- once you are
using some conda-forge it's a bit more robust to use it for everyting else
by default.

conda config --add channels conda-forge


Good luck!

-CHB

I have been looking on the web, and a lot of people get this type of error
on Conda, with all sort of libraries. Some people suggest to append a path,
"CAUTION: Setting LD_LIBRARY_PATH on Linux or DYLD_LIBRARY_PATH on macOS
can interfere with this because the dynamic linker short-circuits link
resolution by first looking at LD_LIBRARY_PATH." (
https://conda.io/docs/user-guide/tasks/build-packages/
use-shared-libraries.html)
Can anyone clearly explain to me what is the issue, and how to correct it?
Is it a conflict due to a bad install, a path ...? It seem many people
encounter this error and there is not really a clear solution.
Thank you
--
You received this message because you are subscribed to the Google Groups
"conda - Public" group.
To unsubscribe from this group and stop receiving emails from it, send an
Visit this group at https://groups.google.com/a/continuum.io/group/conda/.
To view this discussion on the web visit https://groups.google.com/a/
continuum.io/d/msgid/conda/81b64b97-a7dc-431c-aab6-
4d14795fba51%40continuum.io
<https://groups.google.com/a/continuum.io/d/msgid/conda/81b64b97-a7dc-431c-aab6-4d14795fba51%40continuum.io?utm_medium=email&utm_source=footer>
.
For more options, visit https://groups.google.com/a/continuum.io/d/optout.

--
Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R [phone] voice
7600 Sand Point Way NE [phone] fax
Seattle, WA 98115 [phone] main reception

***@noaa.gov

[Paul Hobson]

i'm going to add to Chris' suggestion a bit.

In my experience with GIS-y libraries (GDAL, fiona, geopandas), it's best
to just install *everything* (yes, really everything) from conda-forge. So
something to try if his procedure doesn't work would be:

conda create --name=pyncotest python=2 pynco --channel=conda-forge

Then activate and test that env.

Let's us know how it works out.
-Paul

Post by Chris Barker

I am getting the following error when trying to run pynco (
https://github.com/nco/pynco)
*/home/xxx/miniconda2/envs/python_env/bin/ncra: error while loading
shared libraries: libnetcdf.so.7: cannot open shared object file: No such
file or directory*
- I do not have libnetcdf.so.7 in /home/xxx/miniconda2/envs/libs (I
have libnetcdf.so.13)
- I installed pynco by "conda install -c conda-forge pynco". No error
during install.
- I use other libraries linked to libnetcdf and no problem (nco, netcdf4,
numpy, ...)
- Config
platform : linux-64
conda version : 4.3.29
conda is private : False
conda-env version : 4.3.29
conda-build version : not installed
python version : 2.7.14.final.0
requests version : 2.18.4

netcdf in particular can lead to a tangled web ob dependencies, that conda
conda create -n pynco python=2
this will create a minimal environment (called pynco) with only python2
and the stuff conda needs to run.
source activate pynco
conda install -c conda-forge pynco
That should install pynco and everything it depends on.
see if it works.
If it doesn't file a bug report on conda-forge
If it does, then try installing the other stuff you need and hope they
don't break anything :-)
I recommend putting conda-forge first on your channels path -- once you
are using some conda-forge it's a bit more robust to use it for everyting
else by default.
conda config --add channels conda-forge
Good luck!
-CHB

I have been looking on the web, and a lot of people get this type of
error on Conda, with all sort of libraries. Some people suggest to append a
"CAUTION: Setting LD_LIBRARY_PATH on Linux or DYLD_LIBRARY_PATH on macOS
can interfere with this because the dynamic linker short-circuits link
resolution by first looking at LD_LIBRARY_PATH." (
https://conda.io/docs/user-guide/tasks/build-packages/use-
shared-libraries.html)
Can anyone clearly explain to me what is the issue, and how to correct
it? Is it a conflict due to a bad install, a path ...? It seem many people
encounter this error and there is not really a clear solution.
Thank you
--
You received this message because you are subscribed to the Google Groups
"conda - Public" group.
To unsubscribe from this group and stop receiving emails from it, send an
Visit this group at https://groups.google.com/a/continuum.io/group/conda/
.
To view this discussion on the web visit https://groups.google.com/a/co
ntinuum.io/d/msgid/conda/81b64b97-a7dc-431c-aab6-4d14795fba5
1%40continuum.io
<https://groups.google.com/a/continuum.io/d/msgid/conda/81b64b97-a7dc-431c-aab6-4d14795fba51%40continuum.io?utm_medium=email&utm_source=footer>
.
For more options, visit https://groups.google.com/a/continuum.io/d/optout
.

--
Christopher Barker, Ph.D.
Oceanographer
Emergency Response Division
NOAA/NOS/OR&R [phone] voice
7600 Sand Point Way NE
<https://maps.google.com/?q=7600+Sand+Point+Way+NE&entry=gmail&source=g>
[phone] fax
Seattle, WA 98115 [phone] main reception
--
You received this message because you are subscribed to the Google Groups
"conda - Public" group.
To unsubscribe from this group and stop receiving emails from it, send an
Visit this group at https://groups.google.com/a/continuum.io/group/conda/.
To view this discussion on the web visit https://groups.google.com/a/
continuum.io/d/msgid/conda/CALGmxELrQDbEL1DyO5Sqmv0V%
3DTdZfHb9JdLHGFhAygr1bzn5OQ%40mail.gmail.com
<https://groups.google.com/a/continuum.io/d/msgid/conda/CALGmxELrQDbEL1DyO5Sqmv0V%3DTdZfHb9JdLHGFhAygr1bzn5OQ%40mail.gmail.com?utm_medium=email&utm_source=footer>
.
For more options, visit https://groups.google.com/a/continuum.io/d/optout.

[Paul Hobson]

Michael and Ray,

Thanks for the updated info. That's great to know.

-Paul

Installing from conda-forge is certainly historically the better option
historically, but I have had feedback from Filipe Fernandes that he trusts
our rebuilt stack more now, thanks to its reliance on new conda-build 3
features such as run_exports. I don't think pynco in particular is on
defaults yet.
I second Ray's suggestion to not mix packages. Binary incompatibility is
a major problem with the GIS stack especially. We're working with
conda-forge to achieve binary compatibility, but it will take some time.
There are several technical and social issues to resolve.
Whichever you choose to go with for now, keep in mind that what you choose
affects where you should file your issues and support requests. If you go
with conda-forge, you should ask questions on their gitter channel,
https://gitter.im/conda-forge/conda-forge.github.io, or on the relevant
feedstock repo, such as https://github.com/conda-forge/pynco-feedstock.
If you go with defaults, please file issues on https://github.com/
continuumio/anaconda-issues

AD has a completely a completely modernised and compatible gdal stack now
so feel free to try either. Just avoid mixing conda-forge and AD in the
same environment.

Post by Paul Hobson
i'm going to add to Chris' suggestion a bit.
In my experience with GIS-y libraries (GDAL, fiona, geopandas), it's
best to just install *everything* (yes, really everything) from
conda create --name=pyncotest python=2 pynco --channel=conda-forge
Then activate and test that env.
Let's us know how it works out.
-Paul

Post by Chris Barker

I am getting the following error when trying to run pynco (
https://github.com/nco/pynco)
*/home/xxx/miniconda2/envs/python_env/bin/ncra: error while loading
shared libraries: libnetcdf.so.7: cannot open shared object file: No such
file or directory*
- I do not have libnetcdf.so.7 in /home/xxx/miniconda2/envs/libs (I
have libnetcdf.so.13)
- I installed pynco by "conda install -c conda-forge pynco". No error
during install.
- I use other libraries linked to libnetcdf and no problem (nco,
netcdf4, numpy, ...)
- Config
platform : linux-64
conda version : 4.3.29
conda is private : False
conda-env version : 4.3.29
conda-build version : not installed
python version : 2.7.14.final.0
requests version : 2.18.4

netcdf in particular can lead to a tangled web ob dependencies, that
conda create -n pynco python=2
this will create a minimal environment (called pynco) with only python2
and the stuff conda needs to run.
source activate pynco
conda install -c conda-forge pynco
That should install pynco and everything it depends on.
see if it works.
If it doesn't file a bug report on conda-forge
If it does, then try installing the other stuff you need and hope they
don't break anything :-)
I recommend putting conda-forge first on your channels path -- once you
are using some conda-forge it's a bit more robust to use it for everyting
else by default.
conda config --add channels conda-forge
Good luck!
-CHB

I have been looking on the web, and a lot of people get this type of
error on Conda, with all sort of libraries. Some people suggest to append a
"CAUTION: Setting LD_LIBRARY_PATH on Linux or DYLD_LIBRARY_PATH on
macOS can interfere with this because the dynamic linker short-circuits
link resolution by first looking at LD_LIBRARY_PATH." (
https://conda.io/docs/user-guide/tasks/build-packages/use-
shared-libraries.html)
Can anyone clearly explain to me what is the issue, and how to correct
it? Is it a conflict due to a bad install, a path ...? It seem many people
encounter this error and there is not really a clear solution.
Thank you
--
You received this message because you are subscribed to the Google
Groups "conda - Public" group.
To unsubscribe from this group and stop receiving emails from it, send
Visit this group at https://groups.google.com/a/co
ntinuum.io/group/conda/.
To view this discussion on the web visit
https://groups.google.com/a/continuum.io/d/msgid/conda/81b64
b97-a7dc-431c-aab6-4d14795fba51%40continuum.io
<https://groups.google.com/a/continuum.io/d/msgid/conda/81b64b97-a7dc-431c-aab6-4d14795fba51%40continuum.io?utm_medium=email&utm_source=footer>
.
For more options, visit https://groups.google.com/a/co
ntinuum.io/d/optout.

--
Christopher Barker, Ph.D.
Oceanographer
Emergency Response Division
NOAA/NOS/OR&R [phone] voice
7600 Sand Point Way NE
<https://maps.google.com/?q=7600+Sand+Point+Way+NE&entry=gmail&source=g>
[phone] fax
Seattle, WA 98115 [phone] main reception
--
You received this message because you are subscribed to the Google
Groups "conda - Public" group.
To unsubscribe from this group and stop receiving emails from it, send
Visit this group at https://groups.google.com/a/co
ntinuum.io/group/conda/.
To view this discussion on the web visit https://groups.google.com/a/co
ntinuum.io/d/msgid/conda/CALGmxELrQDbEL1DyO5Sqmv0V%3DTdZfHb9
JdLHGFhAygr1bzn5OQ%40mail.gmail.com
<https://groups.google.com/a/continuum.io/d/msgid/conda/CALGmxELrQDbEL1DyO5Sqmv0V%3DTdZfHb9JdLHGFhAygr1bzn5OQ%40mail.gmail.com?utm_medium=email&utm_source=footer>
.
For more options, visit https://groups.google.com/a/co
ntinuum.io/d/optout.

--
You received this message because you are subscribed to the Google
Groups "conda - Public" group.
To unsubscribe from this group and stop receiving emails from it, send
Visit this group at https://groups.google.com/a/co
ntinuum.io/group/conda/.
To view this discussion on the web visit https://groups.google.com/a/co
ntinuum.io/d/msgid/conda/CADT3MEAdP-bVA3Jh%2Bum9DWaH5DBUpZZ%
2By7oC3Ut1FjqZ0gT7FQ%40mail.gmail.com
<https://groups.google.com/a/continuum.io/d/msgid/conda/CADT3MEAdP-bVA3Jh%2Bum9DWaH5DBUpZZ%2By7oC3Ut1FjqZ0gT7FQ%40mail.gmail.com?utm_medium=email&utm_source=footer>
.
For more options, visit https://groups.google.com/a/co
ntinuum.io/d/optout.

--
You received this message because you are subscribed to the Google Groups
"conda - Public" group.
To unsubscribe from this group and stop receiving emails from it, send an
Visit this group at https://groups.google.com/a/continuum.io/group/conda/
.
To view this discussion on the web visit https://groups.google.com/a/co
ntinuum.io/d/msgid/conda/CAF5kcVVPiv%2BE1F3mXv%2BqkZi3YGWTqa
PE9%3Dsfm8xzX1hyD5SPMA%40mail.gmail.com
<https://groups.google.com/a/continuum.io/d/msgid/conda/CAF5kcVVPiv%2BE1F3mXv%2BqkZi3YGWTqaPE9%3Dsfm8xzX1hyD5SPMA%40mail.gmail.com?utm_medium=email&utm_source=footer>
.
For more options, visit https://groups.google.com/a/continuum.io/d/optout
.

[Karen Guihou]

Hi all

Thanks for your answers!
I have reinstalled my whole environnement using only conda-forge, and it
seems indeed much better.

Cheers

Post by Karen Guihou
Hi all
I am getting the following error when trying to run pynco (
https://github.com/nco/pynco)
*/home/xxx/miniconda2/envs/python_env/bin/ncra: error while loading shared
libraries: libnetcdf.so.7: cannot open shared object file: No such file or
directory*
- I do not have libnetcdf.so.7 in /home/xxx/miniconda2/envs/libs (I
have libnetcdf.so.13)
- I installed pynco by "conda install -c conda-forge pynco". No error
during install.
- I use other libraries linked to libnetcdf and no problem (nco, netcdf4,
numpy, ...)
- Config
platform : linux-64
conda version : 4.3.29
conda is private : False
conda-env version : 4.3.29
conda-build version : not installed
python version : 2.7.14.final.0
requests version : 2.18.4
I have been looking on the web, and a lot of people get this type of error
on Conda, with all sort of libraries. Some people suggest to append a path,
"CAUTION: Setting LD_LIBRARY_PATH on Linux or DYLD_LIBRARY_PATH on macOS
can interfere with this because the dynamic linker short-circuits link
resolution by first looking at LD_LIBRARY_PATH." (
https://conda.io/docs/user-guide/tasks/build-packages/use-shared-libraries.html
)
Can anyone clearly explain to me what is the issue, and how to correct it?
Is it a conflict due to a bad install, a path ...? It seem many people
encounter this error and there is not really a clear solution.
Thank you