python environment on SGE cluster nodes
Hi, We're close to having a working Galaxy job_conf.xml to submit galaxy jobs to an SGE cluster. At the moment a submitted job is failing with: python: error while loading shared libraries: libpython2.7.so.1.0: cannot open shared object file: No such file or directory We'd like to solve this by sourcing the dev tools for all our jobs: source /opt/rh/python27 enable Please could you advise on how to proceed? regards David -- David Simpson Systems Engineer MRC Centre for Neuropsychiatric Genetics and Genomics Cardiff University School of Medicine Hadyn Ellis Building Maindy Road Cardiff CF24 4HQ ------------------ phone 44 29206 88426 mail SimpsonD4 at Cardiff.ac.uk
Managed to get working by using <env file="/opt/rh/python27/enable" /> On Mon, 2017-05-08 at 10:52 +0100, David Simpson wrote: Hi, We're close to having a working Galaxy job_conf.xml to submit galaxy jobs to an SGE cluster. At the moment a submitted job is failing with: python: error while loading shared libraries: libpython2.7.so.1.0: cannot open shared object file: No such file or directory We'd like to solve this by sourcing the dev tools for all our jobs: source /opt/rh/python27 enable Please could you advise on how to proceed? regards David -- David Simpson Systems Engineer MRC Centre for Neuropsychiatric Genetics and Genomics Cardiff University School of Medicine Hadyn Ellis Building Maindy Road Cardiff CF24 4HQ ------------------ phone 44 29206 88426 mail SimpsonD4 at Cardiff.ac.uk -- David Simpson Systems Engineer MRC Centre for Neuropsychiatric Genetics and Genomics Cardiff University School of Medicine Hadyn Ellis Building Maindy Road Cardiff CF24 4HQ ------------------ phone 44 29206 88426 mail SimpsonD4 at Cardiff.ac.uk
Now a bit further along (running a basic workflow on the cluster) and encountering this problem (see bottom). This is how the job_conf.xml is setting up: <destination id="sge_default" runner="sge"> <env id="SGE_ROOT">/opt/gridengine</env> <env file="/opt/rh/python27/enable" /> <param id="nativeSpecification">-V -l h_vmem=2000M</param> </destination> It's an issue with the environment on the compute nodes, but not quite sure how to solve it. Traceback (most recent call last): File "/c8000xd2/illumina-cnv- galaxy/galaxy/database/jobs_directory/000/711/set_metadata_wAx_ru.py", line 1, in <module> from galaxy_ext.metadata.set_metadata import set_metadata; set_metadata() File "/c8000xd2/illumina-cnv- galaxy/galaxy/lib/galaxy_ext/metadata/set_metadata.py", line 25, in <module> import galaxy.model.mapping # need to load this before we unpickle, in order to setup properties assigned by the mappers File "/c8000xd2/illumina-cnv- galaxy/galaxy/lib/galaxy/model/__init__.py", line 21, in <module> from sqlalchemy import (and_, func, join, not_, or_, select, true, type_coerce, ImportError: cannot import name true On Mon, 2017-05-08 at 14:24 +0100, David Simpson wrote:
Managed to get working by using <env file="/opt/rh/python27/enable" />
On Mon, 2017-05-08 at 10:52 +0100, David Simpson wrote: Hi,
We're close to having a working Galaxy job_conf.xml to submit galaxy jobs to an SGE cluster.
At the moment a submitted job is failing with:
python: error while loading shared libraries: libpython2.7.so.1.0: cannot open shared object file: No such file or directory
We'd like to solve this by sourcing the dev tools for all our jobs:
source /opt/rh/python27 enable
Please could you advise on how to proceed?
regards David
-- David Simpson Systems Engineer MRC Centre for Neuropsychiatric Genetics and Genomics Cardiff University School of Medicine Hadyn Ellis Building Maindy Road Cardiff CF24 4HQ ------------------ phone 44 29206 88426 mail SimpsonD4 at Cardiff.ac.uk -- David Simpson Systems Engineer MRC Centre for Neuropsychiatric Genetics and Genomics Cardiff University School of Medicine Hadyn Ellis Building Maindy Road Cardiff CF24 4HQ
phone 44 29206 88426 mail SimpsonD4 at Cardiff.ac.uk
Problem solved by using the Galaxy VENV environment instead of RH Dev tools python27 - changed within the job_conf.xml: <env file="/opt/rh/python27/enable" /> changed to <env file="/some/dir/galaxy/.venv/bin/activate" /> As the RH Dev tools python27 has an older sqlalchemy ('0.7.9') vs the VENV ('1.0.15') ... causing the previous error On Tue, 2017-05-09 at 11:30 +0100, David Simpson wrote:
Now a bit further along (running a basic workflow on the cluster) and encountering this problem (see bottom).
This is how the job_conf.xml is setting up:
<destination id="sge_default" runner="sge"> <env id="SGE_ROOT">/opt/gridengine</env> <env file="/opt/rh/python27/enable" /> <param id="nativeSpecification">-V -l h_vmem=2000M</param> </destination>
It's an issue with the environment on the compute nodes, but not quite sure how to solve it.
Traceback (most recent call last): File "/c8000xd2/illumina-cnv- galaxy/galaxy/database/jobs_directory/000/711/set_metadata_wAx_ru.py" , line 1, in <module> from galaxy_ext.metadata.set_metadata import set_metadata; set_metadata() File "/c8000xd2/illumina-cnv- galaxy/galaxy/lib/galaxy_ext/metadata/set_metadata.py", line 25, in <module> import galaxy.model.mapping # need to load this before we unpickle, in order to setup properties assigned by the mappers File "/c8000xd2/illumina-cnv- galaxy/galaxy/lib/galaxy/model/__init__.py", line 21, in <module> from sqlalchemy import (and_, func, join, not_, or_, select, true, type_coerce, ImportError: cannot import name true
On Mon, 2017-05-08 at 14:24 +0100, David Simpson wrote:
Managed to get working by using <env file="/opt/rh/python27/enable" />
On Mon, 2017-05-08 at 10:52 +0100, David Simpson wrote: Hi,
We're close to having a working Galaxy job_conf.xml to submit galaxy jobs to an SGE cluster.
At the moment a submitted job is failing with:
python: error while loading shared libraries: libpython2.7.so.1.0: cannot open shared object file: No such file or directory
We'd like to solve this by sourcing the dev tools for all our jobs:
source /opt/rh/python27 enable
Please could you advise on how to proceed?
regards David
-- David Simpson Systems Engineer MRC Centre for Neuropsychiatric Genetics and Genomics Cardiff University School of Medicine Hadyn Ellis Building Maindy Road Cardiff CF24 4HQ ------------------ phone 44 29206 88426 mail SimpsonD4 at Cardiff.ac.uk -- David Simpson Systems Engineer MRC Centre for Neuropsychiatric Genetics and Genomics Cardiff University School of Medicine Hadyn Ellis Building Maindy Road Cardiff CF24 4HQ
phone 44 29206 88426 mail SimpsonD4 at Cardiff.ac.uk
participants (1)
-
David Simpson