So the `preserve_python_environment = legacy_and_local` setting is causing this for you.
Now normally deeptools is not a local tool, but I understand that because of https://github.com/galaxyproject/galaxy/issues/4591,
you have manipulated the shed_tool_conf.xml file, and it is now being considered as a local tool.
In any case this setting should be set to legacy_and_local only if you experience difficulties when it is not set,
that is why `legacy` is the default.


On 3 October 2017 at 10:30, julie dubois <dubjulie@gmail.com> wrote:
When I run computeMatrix, I have on ly this in the log file :

192.168.000.000 - - [03/oct./2017:10:16:07 +0200] "POST
/galaxy/api/tools/deeptools_compute_matrix/build HTTP/1.1" 200 -
"http://192.168.100.000:8080/galaxy?job_id=7e6a9fe53e29fc38&__identifer=tmmenbexqep"
"Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:37.0) Gecko/20100101
Firefox/37.0"

 On the job working directory I have this in tool_script.sh :
 #!/bin/bash

# The following block can be used by the job system
# to ensure this script is runnable before actually attempting
# to run it.
if [ -n "$ABC_TEST_JOB_SCRIPT_INTEGRITY_XYZ" ]; then
    exit 42
fi
export PATH=$PATH:'/data/galaxy-dist/tool_dependency/_conda/envs/__deeptools@2.5.1/bin'
; computeMatrix --version >
/data/galaxy-dist/database/tmp/GALAXY_VERSION_STRING_38413 2>&1; ln -f
-s "/data/galaxy-dist/database/files/080/dataset_80817.dat"
"MACS2narrow_SamToBam_Bowtie_mapped_readsTrim__0.bw" &&    ln -f -s
"/data/galaxy-dist/database/files/080/dataset_80593.dat"
"MACS2narrow_SamToBam_Bowtie_mapped_readsTrim__0.bed" &&
computeMatrix  scale-regions --regionsFileName
'MACS2narrow_SamToBam_Bowtie_mapped_readsTrim__0.bed'  --scoreFileName
'MACS2narrow_SamToBam_Bowtie_mapped_readsTrim__0.bw'  --outFileName
'/data/galaxy-dist/database/files/080/dataset_80995.dat'
--numberOfProcessors "${GALAXY_SLOTS:-4}"   --regionBodyLength 500

It seems to export the good path for executable. So I don't understand.

In the galaxy_38413.sh :
#!/bin/bash



# The following block can be used by the job system
# to ensure this script is runnable before actually attempting
# to run it.
if [ -n "$ABC_TEST_JOB_SCRIPT_INTEGRITY_XYZ" ]; then
    exit 42
fi

GALAXY_SLOTS="1"; export GALAXY_SLOTS;
export GALAXY_SLOTS
PRESERVE_GALAXY_ENVIRONMENT="True"
GALAXY_LIB="/data/galaxy-dist/lib"
if [ "$GALAXY_LIB" != "None" -a "$PRESERVE_GALAXY_ENVIRONMENT" = "True" ]; then
    if [ -n "$PYTHONPATH" ]; then
        PYTHONPATH="$GALAXY_LIB:$PYTHONPATH"
    else
        PYTHONPATH="$GALAXY_LIB"
    fi
    export PYTHONPATH
fi

GALAXY_VIRTUAL_ENV="/data/galaxy-dist/.venv"
if [ "$GALAXY_VIRTUAL_ENV" != "None" -a -z "$VIRTUAL_ENV" \
     -a -f "$GALAXY_VIRTUAL_ENV/bin/activate" -a
"$PRESERVE_GALAXY_ENVIRONMENT" = "True" ]; then
    . "$GALAXY_VIRTUAL_ENV/bin/activate"
fi
echo "$GALAXY_SLOTS" >
'/data/galaxy-dist/database/job_working_directory/038/38413/__instrument_core_galaxy_slots'
date +"%s" > /data/galaxy-dist/database/job_working_directory/038/38413/__instrument_core_epoch_start
cd /data/galaxy-dist/database/job_working_directory/038/38413
rm -rf working; mkdir -p working; cd working;
/data/galaxy-dist/database/job_working_directory/038/38413/tool_script.sh;
return_code=$?; cd
'/data/galaxy-dist/database/job_working_directory/038/38413';
if [ "$GALAXY_LIB" != "None" ]; then
    if [ -n "$PYTHONPATH" ]; then
        PYTHONPATH="$GALAXY_LIB:$PYTHONPATH"
    else
        PYTHONPATH="$GALAXY_LIB"
    fi
    export PYTHONPATH
fi
if [ "$GALAXY_VIRTUAL_ENV" != "None" -a -z "$VIRTUAL_ENV"      -a -f
"$GALAXY_VIRTUAL_ENV/bin/activate" ]; then
    . "$GALAXY_VIRTUAL_ENV/bin/activate"
fi
GALAXY_PYTHON=`command -v python`
export PATH=$PATH:'/data/galaxy-dist/tool_dependency/_conda/envs/__samtools@1.3.1/bin'
; python "/data/galaxy-dist/database/job_working_directory/038/38413/set_metadata_c9AkjG.py"
"/data/galaxy-dist/database/job_working_directory/038/38413/registry.xml"
"/data/galaxy-dist/database/job_working_directory/038/38413/working/galaxy.json"
"/data/galaxy-dist/database/job_working_directory/038/38413/metadata_in_HistoryDatasetAssociation_72822_Eqsgna,/data/galaxy-dist/database/job_working_directory/038/38413/metadata_kwds_HistoryDatasetAssociation_72822_3VQ3be,/data/galaxy-dist/database/job_working_directory/038/38413/metadata_out_HistoryDatasetAssociation_72822_TQg8Pq,/data/galaxy-dist/database/job_working_directory/038/38413/metadata_results_HistoryDatasetAssociation_72822_ctopJa,/data/galaxy-dist/database/files/080/dataset_80995.dat,/data/galaxy-dist/database/job_working_directory/038/38413/metadata_override_HistoryDatasetAssociation_72822_Izp4CZ"
5242880; sh -c "exit $return_code"
echo $? > /data/galaxy-dist/database/job_working_directory/038/38413/galaxy_38413.ec
date +"%s" > /data/galaxy-dist/database/job_working_directory/038/38413/__instrument_core_epoch_end


May be my galaxy.ini in not well configured for conda.
For the moment concerning conda in galaxy.ini, I have these lines uncommented :

preserve_python_environment = legacy_and_local
dependency_resolvers_config_file = config/dependency_resolvers_conf.xml



2017-10-03 9:53 GMT+02:00 Marius van den Beek <m.vandenbeek@gmail.com>:
> Also upon re-reading I see that your dependency is being picked up from
> /usr/local/bin,
> which indicates that this isn't using conda at all.
> A small excerpt from the logs when you try to run the tool may be helpful
> to understand what is going on there.
>
> Best,
> Marius
>
> On 3 October 2017 at 09:29, Marius van den Beek <m.vandenbeek@gmail.com>
> wrote:
>>
>> Hi Julie,
>>
>> any chance you are setting PYTHONPATH before starting galaxy's job
>> handlers (a common place would the supervisor config, in case you're using
>> that)?
>> This will interfere with all python modules that are installed via conda.
>>
>> Best,
>> Marius
>>
>> On 3 October 2017 at 09:13, julie dubois <dubjulie@gmail.com> wrote:
>>>
>>> Dear Greg,
>>> I read your discussion with @bgruening and I think it's not exactly
>>> the same. Because in my case, The executable IS present in my
>>> galaxy-dist/tool_dependency/_conda/envs/__deeptools@2.5.1/bin .
>>> The problem is far away in the execution when the tool (computeMatrix
>>> in my example) try to load the python module deeptools.computeMatrix.
>>> So is there a solution to discover where is the problem ?
>>>
>>> Thanks
>>>
>>> 2017-10-02 15:16 GMT+02:00 Von Kuster, Greg <ghv2@psu.edu>:
>>> > This looks like the issue I’ve recently seen as well - I just discussed
>>> > this
>>> > with @bgruening on gitter - https://gitter.im/galaxy-iuc/iuc.  I’m
>>> > using a
>>> > work-around of creating the __deeptools conda env manually.  In my
>>> > case, I’m
>>> > not installing the whole suite, just the bamCoverage tool from the TS -
>>> >
>>> > https://toolshed.g2.bx.psu.edu/view/bgruening/deeptools_bam_coverage/5d11599b8a7d.
>>> > My work-around is to create the conda env manually:
>>> >
>>> > $conda create -n __deeptools@2.5.1 deep tools=2.5.1
>>> >
>>> > This env is created in the conda environment outside of Galaxy, so I
>>> > just
>>> > copy it to the conda environment I’ve configured for Galaxy.  This
>>> > works,
>>> > but it seems there is some issue with the installation from the TS, so
>>> > perhaps a more ling term fix is needed.
>>> >
>>> >
>>> > On Oct 2, 2017, at 8:59 AM, julie dubois <dubjulie@gmail.com> wrote:
>>> >
>>> > Hi dev team,
>>> >
>>> > I have a very big problem since I updated to galaxy 17.05 with
>>> > deeptools from toolshed.
>>> >
>>> > Due to a bug in galaxy 17.05 with toolshed installed tools (reported
>>> > in this issue https://github.com/galaxyproject/galaxy/issues/4591), I
>>> > re-installed deeptools to try to resolve it.
>>> > Not resolved really but it's another problem. I temporarly bypassed it.
>>> >
>>> > Here is my problem : I've installed deeptools from toolshed and when
>>> > we try to use "compute matrix" for example, we have this error message
>>> > :
>>> >
>>> > Fatal error: Exit code 1 ()
>>> > Traceback (most recent call last):
>>> >  File "/usr/local/bin/computeMatrix", line 4, in <module>
>>> >    from deeptools.computeMatrix import main
>>> > ImportError: No module named deeptools.computeMatrix
>>> >
>>> > It seems that wrapper are installed but not the tool "deeptools".
>>> >
>>> > For more information, here it's the way for the installation :
>>> > search and install via toolshed
>>> > Install "suite_deeptools" (version 2.5.1.1.0 : 3d68b716965a)
>>> > Info in Manage tool dependencies : version=2.5.1, resolver=Conda,
>>> > green check symbol in the line with a list of all deeptools.
>>> >
>>> > Galaxy 17.05, commit cfabe37 (21 sept 2017)
>>> >
>>> > I've tried to uninstall/reinstall dependencies with conda but without
>>> > success.
>>> >
>>> > Have you any idea to install the deeptools please ?
>>> >
>>> > Thanks.
>>> > Julie
>>> > ___________________________________________________________
>>> > Please keep all replies on the list by using "reply all"
>>> > in your mail client.  To manage your subscriptions to this
>>> > and other Galaxy lists, please use the interface at:
>>> >  https://lists.galaxyproject.org/
>>> >
>>> > To search Galaxy mailing lists use the unified search at:
>>> >  http://galaxyproject.org/search/
>>> >
>>> >
>>> ___________________________________________________________
>>> Please keep all replies on the list by using "reply all"
>>> in your mail client.  To manage your subscriptions to this
>>> and other Galaxy lists, please use the interface at:
>>>   https://lists.galaxyproject.org/
>>>
>>> To search Galaxy mailing lists use the unified search at:
>>>   http://galaxyproject.org/search/
>>
>>
>