Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Nicola Soranzo
    @nsoranzo:matrix.org
    [m]
    and you can drop all symlinks for the output files
    martenson
    @martenson:matrix.org
    [m]
    Does anyone have an idea why my registry.xml file in job working directory has content duplicated 11 times? (resulting in invalid xml)
    martenson
    @martenson:matrix.org
    [m]
    this is after update to 22.01
    martenson
    @martenson:matrix.org
    [m]
    From another bucket, can you downgrade from alembic to sqlalchemy migrate?
    Marius
    @mvdbeek:matrix.org
    [m]
    Race condition between handlers ? Can't really think of any other scenario
    What kind of job setup and handler assignment are you running ?
    sergey
    @ic4f:matrix.org
    [m]
    martenson: shouldn't be a problem. Going from 22.01 to dev (or 22.05) doesn't delete the migrate_version table. However, I would first run the db downgrade under alembic (because sqlalchemy migrate doesn't know about anything that happened under alembic), then checkout whatever pre-alembic branch you need, and you should be good to go.
    if something doesn't work as expected, pls let me know
    martenson
    @martenson:matrix.org
    [m]
    @mvdbeek: that's my dev instance :D
    should have default gunicorn and jobs sent to docker I think
    although it affects upload jobs too, and those are run locally
    martenson
    @martenson:matrix.org
    [m]
    sergey: that's great news
    martenson
    @martenson:matrix.org
    [m]
    confirmed working, thanks!
    Admin training always had a couple of folks confused with version migration, I fear this is going to enlarge it a bit. :D
    admins of new instances should have easier life though
    sergey
    @ic4f:matrix.org
    [m]
    yeah, it might. Maybe I should be at the workshop to help out in case it gets confusing
    martenson
    @martenson:matrix.org
    [m]
    the readme document in the module source code is helpful, thanks for that
    sergey
    @ic4f:matrix.org
    [m]
    it's outdated - i'm working on a new version: galaxyproject/galaxy#13710
    martenson
    @martenson:matrix.org
    [m]
    Maybe a note that downgrade to pre-alembic is possible would fit
    sergey
    @ic4f:matrix.org
    [m]
    yep, for sure - already added to my notes/draft to do :)
    martenson
    @martenson:matrix.org
    [m]
    But you know what John would say.
    sergey
    @ic4f:matrix.org
    [m]
    "it'll also change next year anyway" ? ;-)
    martenson
    @martenson:matrix.org
    [m]
    "I hate Galaxy, rumble rumble, but this should go to https://docs.galaxyproject.org/."
    sergey
    @ic4f:matrix.org
    [m]
    yeah, that too. And I agree. So this readme is a "staging version" of sorts
    bgruening
    @bgruening:matrix.org
    [m]
    tried to answer, thanks Jennifer Hillman-Jackson
    Nikolay Vazov
    @vazovn:matrix.org
    [m]
    Hi, which is the correct tag for the OIDC_ENDPOINT in the file oidc_backends_config.xml: <url>, <api_url>or can I skip it as long as I have it in the backend python file as _OIDCENDPOINT
    Some providers in https://galaxyproject.org/authnz/config/oidc/ do not have it at all.
    Actually in the doc https://galaxyproject.org/authnz/config/oidc/ <url>appears as an option, and in the oidc_backends_config.xml it is replaced, in some providers, as <api_url>.
    Martin Wolstencroft
    @martinwolst:matrix.org
    [m]
    To my limited understanding it is generally similar but provider-specific for some tags
    Nikolay Vazov
    @vazovn:matrix.org
    [m]
    All right, I will check with the provider here. I was wondering if it was needed at all. Since <url>only appears in the example in the doc https://galaxyproject.org/authnz/config/oidc/#oidc-configuration-options-for-identity-providers but it is not explained in the text. All other tags are explained in detail. Maybe, if it is not required, we shall remove it from the doc?
    Martin Wolstencroft
    @martinwolst:matrix.org
    [m]
    There are a few things that could be clearer e.g. the redirect_uri is included, but you still seem to have to provide that callback address to the identity provider separately for it to work (my experience with Azure)
    Nikolay Vazov
    @vazovn:matrix.org
    [m]
    Yes, you certainly need to provide the <redirect_url> to the IdP. My question was more about the opposite link: how your client will connect to the IdP ...
    Nikolay Vazov
    @vazovn:matrix.org
    [m]
    OK, I can confirm now that it worked without the <url> tag in the oidc_backends_config.xml, it seems that the OIDC_ENDPOINT variable is enough in the backend python file. I am using our own IdP and the backend (very basic) is based on OpenIdConnectAuth class.
    Nicola Soranzo
    @nsoranzo:matrix.org
    [m]
    Glad it's solved! If documentation can be improved, pull requests are always welcome, you are the domain experts here!
    pvanheus
    @pvanheus
    as I recall there's a setting to stop tool dependencies being installed with conda when installing with ephemeris - can someone please remind me what it is?
    3 replies
    Cristóbal Gallardo
    @gallardoalba:matrix.org
    [m]
    Hi! There's a question in Galaxy Help about Ansible, could somebody have a look? https://help.galaxyproject.org/t/galaxy-installation-via-ansible/7890 Thanks!
    1 reply
    Marius
    @mvdbeek:matrix.org
    [m]
    You can also get more details on the 400 response in your browser's javascript console
    Maiken Pedersen
    @maikenp

    Hi! I am now trying to run a notebook for ML that needs to use the gpus on the host machine. I have installed nvidia container toolkit, and can run docker successfully and see the gpus in the container with: docker run --rm --gpus all nvidia/cuda:11.0-base nvidia-smi

    Can I make galaxy add the --gpus all option for the docker run command it uses? I have naively tried <param id="docker_run">--gpus all</param> in the job_conf.xml file of the destination that is used. But that is not rendered. I will in the meantime look deeper into the documentation to see if I find the solution myself.

    2 replies
    lhg2
    @lhg2
    Hello galaxy admins - I have deployed Galaxy on Postgresql-9.6 without issues, but wanted to know if anyone uses latest version out there? Or if there are any issues with any version I should know about after 9.6? I am building another instance and have the opportunity to deploy a more current version of postgresql.
    ic4f
    @ic4f:matrix.org
    [m]
    I believe main (usegalaxy.org ) is 12.7. Also, I run 13.6 locally as a dev instance (not in production though)
    lhg2
    @lhg2
    ok thanks much!
    lhg2
    @lhg2
    hello all - i just deployed a new instance of Galaxy, installed postgresql, but getting that psycopg2 error (ModuleNotFoundError: No module named 'psycopg2'). This is a RHEL8 server and I've installed python2-psycopg2 and python3-psycopg2. Any suggestions?
    nsoranzo
    @nsoranzo:matrix.org
    [m]
    No need for the python2 one.
    psycopg2 needs to be installed inside Galaxy's virtual environment.
    lhg2
    @lhg2
    ok thanks much.
    lhg2
    @lhg2
    One more please - once that dependency is satisfied (assuming with pip) then i need to run create_db.sh or manage_db.sh upgrade ? This is a new deploy with postgresql freshly installed and empty db created.
    Nicola Soranzo
    @nsoranzo:matrix.org
    [m]
    That shouldn't be necessary, Galaxy initialise the database tables if the database is empty.
    lhg2
    @lhg2
    @nsoranzo:matrix.org (still getting used to deployment step). Thanks!
    triasteran
    @triasteran
    Hi everyone. Is there any way to configure CVMFS reference data to be seen only for a certain tool? E.g. I have bowtie wrapper for genome and for transcriptome, I want reference built-in genomes from CVMFS to be accessed only for genome wrapper, but for transcriptome I want to use my own data
    2 replies