Re: [galaxy-dev] SLURM configuration problem

[Marius van den Beek]

Hello Leonor,

the log you've sent indicates that you're picking up pulsar from
/usr/local/lib.
That should not happen if you're running galaxy in a virtualenv.

Apart from that you did not mention if you able to submit slurm jobs from
the command line.
That is a prerequisite for launching jobs through galaxy.

Could you post the full startup logs and job_conf.xml file somewhere?

Best,
Marius

On 31 March 2017 at 12:38, Leonor Palmeira  wrote:

> Dear all,
>
> we are struggling with the basics in our Galaxy/SLURM configuration.
>
> - Galaxy is installed on a virtual machine that is physically
> independent from our cluster, but on a shared filesystem that is also
> mounted on the Cluster
>
> - Our Cluster is running SLURM and has 'slurm-drmaa' (Poznan version)
> installed. The shared filesystem is mounted on the same mount point as
> the VM, so their /paths are identical
>
> What do we need so that the Galaxy VM is able to submit jobs to the
> Cluster?
> Currently, running ".run.sh" from the VM as root leads to the $SGE_ROOT
> error I posted in my previous email and that ends like this :
>
>   File
> "/usr/local/lib/python2.7/dist-packages/pulsar/managers/
> util/drmaa/__init__.py",
> line 49, in __init__
> DrmaaSession.session.initialize()
>   File "/usr/local/lib/python2.7/dist-packages/drmaa/session.py", line
> 257, in initialize
> py_drmaa_init(contactString)
>   File "/usr/local/lib/python2.7/dist-packages/drmaa/wrappers.py", line
> 73, in py_drmaa_init
> return _lib.drmaa_init(contact, error_buffer, sizeof(error_buffer))
>   File "/usr/local/lib/python2.7/dist-packages/drmaa/errors.py", line
> 151, in error_check
> raise _ERRORS[code - 1](error_string)
> InternalException: code 1: Please set the environment variable SGE_ROOT.
>
> Any help would be greatly appreciated.
>
> Best
> Leonor
>
> Leonor Palmeira | PhD
> Associate Scientist
> Department of Human Genetics
> CHU de Liège | Domaine Universitaire du Sart-Tilman
> 4000 Liège | BELGIQUE
> Tél: +32-4-366.91.41
> Fax: +32-4-366.72.61
> e-mail: lpalme...@chu.ulg.ac.be
>
> On 02/20/2017 03:57 PM, Marius van den Beek wrote:
> > It doesn't hurt to try this, but I don't think that will solve the
> problem.
> >
> > Just to be sure, the basics are working? You can submit jobs via sbatch?
> > How did you compile/install slurm-drmaa ?
> >
> > Also it looks like drmaa-python is being used from /usr/local/... .
> > Are you running galaxy in a virtualenv?
> > It's strongly recommended to do that.
> > Starting galaxy through run.sh will handle the creation and installation
> > of all necessary dependencies for you.
> > Finally it looks like you're loading pulsar from /usr/local ... this is
> > a bit messy.
> > Please try getting the cluster submission to work using run.sh first.
> >
> >
> > On 20 February 2017 at 15:24, Leonor Palmeira  > > wrote:
> >
> > Hi Marius,
> >
> > yes, we are using the one from Poznan. Should we give it a try with
> the
> > fork?
> >
> > Best
> > Leonor
> >
> > Leonor Palmeira | PhD
> > Associate Scientist
> > Department of Human Genetics
> > CHU de Liège | Domaine Universitaire du Sart-Tilman
> > 4000 Liège | BELGIQUE
> > Tél: +32-4-366.91.41 
> > Fax: +32-4-366.72.61 
> > e-mail: lpalme...@chu.ulg.ac.be 
> >
> > On 02/20/2017 03:13 PM, Marius van den Beek wrote:
> > > Hi Leonor,
> > >
> > > Are you sure that you are using a drmaa library that is compatible
> with
> > > slurm?
> > > This http://apps.man.poznan.pl/trac/slurm-drmaa
> >  should work, IIRC,
> > > or alternatively you can use Nate Coraor's fork
> > > here https://github.com/natefoo/slurm-drmaa
> > .
> > >
> > > Best,
> > > Marius
> > >
> > > On 20 February 2017 at 15:06, Leonor Palmeira <
> lpalme...@chu.ulg.ac.be 
> > > >>
> > wrote:
> > >
> > > Hi,
> > >
> > > we modified our configuration as Marius suggested, but we
> > still get the
> > > following error. This is an error we had just before, and we
> > were trying
> > > to fix it by specifying an $SGE_ROOT variable.
> > >
> > > I don't know why this error pops up, as we are trying to use
> > SLURM, not
> > > SGE...
> > >
> > > galaxy.jobs.runners.state_handler_factory DEBUG 2017-02-20
> > 14:58:59,768
> > > Loaded 'failure' state handler from module
> > > galaxy.jobs.runners.state_handlers.resubmit
> > > pulsar.managers.util.drmaa DEBUG 2017-02-20 14:58:59,807
> > > Initializing DRMAA
> > > session from thread MainThread
> 

Re: [galaxy-dev] SLURM configuration problem

[Leonor Palmeira]

Dear all,

we are struggling with the basics in our Galaxy/SLURM configuration.

- Galaxy is installed on a virtual machine that is physically
independent from our cluster, but on a shared filesystem that is also
mounted on the Cluster

- Our Cluster is running SLURM and has 'slurm-drmaa' (Poznan version)
installed. The shared filesystem is mounted on the same mount point as
the VM, so their /paths are identical

What do we need so that the Galaxy VM is able to submit jobs to the Cluster?
Currently, running ".run.sh" from the VM as root leads to the $SGE_ROOT
error I posted in my previous email and that ends like this :

  File
"/usr/local/lib/python2.7/dist-packages/pulsar/managers/util/drmaa/__init__.py",
line 49, in __init__
DrmaaSession.session.initialize()
  File "/usr/local/lib/python2.7/dist-packages/drmaa/session.py", line
257, in initialize
py_drmaa_init(contactString)
  File "/usr/local/lib/python2.7/dist-packages/drmaa/wrappers.py", line
73, in py_drmaa_init
return _lib.drmaa_init(contact, error_buffer, sizeof(error_buffer))
  File "/usr/local/lib/python2.7/dist-packages/drmaa/errors.py", line
151, in error_check
raise _ERRORS[code - 1](error_string)
InternalException: code 1: Please set the environment variable SGE_ROOT.

Any help would be greatly appreciated.

Best
Leonor

Leonor Palmeira | PhD
Associate Scientist
Department of Human Genetics
CHU de Liège | Domaine Universitaire du Sart-Tilman
4000 Liège | BELGIQUE
Tél: +32-4-366.91.41
Fax: +32-4-366.72.61
e-mail: lpalme...@chu.ulg.ac.be

On 02/20/2017 03:57 PM, Marius van den Beek wrote:
> It doesn't hurt to try this, but I don't think that will solve the problem.
> 
> Just to be sure, the basics are working? You can submit jobs via sbatch?
> How did you compile/install slurm-drmaa ?
> 
> Also it looks like drmaa-python is being used from /usr/local/... .
> Are you running galaxy in a virtualenv?
> It's strongly recommended to do that.
> Starting galaxy through run.sh will handle the creation and installation
> of all necessary dependencies for you.
> Finally it looks like you're loading pulsar from /usr/local ... this is
> a bit messy.
> Please try getting the cluster submission to work using run.sh first.
> 
> 
> On 20 February 2017 at 15:24, Leonor Palmeira  > wrote:
> 
> Hi Marius,
> 
> yes, we are using the one from Poznan. Should we give it a try with the
> fork?
> 
> Best
> Leonor
> 
> Leonor Palmeira | PhD
> Associate Scientist
> Department of Human Genetics
> CHU de Liège | Domaine Universitaire du Sart-Tilman
> 4000 Liège | BELGIQUE
> Tél: +32-4-366.91.41 
> Fax: +32-4-366.72.61 
> e-mail: lpalme...@chu.ulg.ac.be 
> 
> On 02/20/2017 03:13 PM, Marius van den Beek wrote:
> > Hi Leonor,
> >
> > Are you sure that you are using a drmaa library that is compatible with
> > slurm?
> > This http://apps.man.poznan.pl/trac/slurm-drmaa
>  should work, IIRC,
> > or alternatively you can use Nate Coraor's fork
> > here https://github.com/natefoo/slurm-drmaa
> .
> >
> > Best,
> > Marius
> >
> > On 20 February 2017 at 15:06, Leonor Palmeira  
> > >>
> wrote:
> >
> > Hi,
> >
> > we modified our configuration as Marius suggested, but we
> still get the
> > following error. This is an error we had just before, and we
> were trying
> > to fix it by specifying an $SGE_ROOT variable.
> >
> > I don't know why this error pops up, as we are trying to use
> SLURM, not
> > SGE...
> >
> > galaxy.jobs.runners.state_handler_factory DEBUG 2017-02-20
> 14:58:59,768
> > Loaded 'failure' state handler from module
> > galaxy.jobs.runners.state_handlers.resubmit
> > pulsar.managers.util.drmaa DEBUG 2017-02-20 14:58:59,807
> > Initializing DRMAA
> > session from thread MainThread
> > Traceback (most recent call last):
> >   File
> > "/home/mass/GAL/APP/galaxy/lib/galaxy/webapps/galaxy/buildapp.py",
> > line 55, in paste_app_factory
> > app = galaxy.app.UniverseApplication( global_conf=global_conf,
> > **kwargs )
> >   File "/home/mass/GAL/APP/galaxy/lib/galaxy/app.py", line 170, in
> > __init__
> > self.job_manager = manager.JobManager( self )
> >   File "/home/mass/GAL/APP/galaxy/lib/galaxy/jobs/manager.py",
> line
> > 23, in
> > __init__
> > self.job_handler = handler.JobHandler( app )
> >   File