Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Nuwan Goonasekera
    @nuwang
    @pcm32 Yes, the chart should be up to date. WRT to installing tools, assuming the tool you want is not in CVMFS, could you use an extraInitContainer or an extraInitScript? https://github.com/galaxyproject/galaxy-helm/blob/5908795139d439f75e969efb9e8eda2a275e4910/galaxy/values-cvmfs.yaml#L55
    By default, all of the cvmfs tools will be available
    Pablo Moreno
    @pcm32
    great, thanks!
    Nuwan Goonasekera
    @nuwang
    @VJalili Is the “Send to cloud” tool available in 19.05 too, or only 19.09? Also, it needs an authnz_id. Is it still necessary to manually visit the api endpoint to get that? ( http://localhost:8080/api/cloud/authz) or is there a more direct way?
    Vahid
    @VJalili
    @nuwang I guess 19.09. There is Authnz UI that may be helpful: galaxyproject/galaxy#7665
    jashwantraj92
    @jashwantraj92
    @nuwang I fixed the statefulset error. But now i get a new error
    Error: validation failed: unable to recognize "": no matches for kind "Ingress" in version "networking.k8s.io/v1beta1"
    can you help me with this.?
    sorry i couldnt get back to you sooner, as I was held up with some other work
    never mind. I fixed the issue. The name should be changed to extensions/v1beta1 in the ingress.yaml file.
    jashwantraj92
    @jashwantraj92
    the helm service for galaxy is now up. But my pods are failing to create due to an error "pod has unbound immediate PersistentVolumeClaims". I need use my default storage as a k8s storage plugin i guess.
    jashwantraj92
    @jashwantraj92
    though i create a persistent volume I still get the same error. can someone help me with this.?
    Nuwan Goonasekera
    @nuwang
    The persisent volume must support the “ReadWriteMany” access mode. Try installing the NFS provisioner helm chart and set NFS as the default storage class and it should work.
    jashwantraj92
    @jashwantraj92
    I manually created all persistentvolumeclaims and that error is fixed
    now all the pods are crashing due to insufficient login credentials to postgres. should i provide the credentials apriori in some config file.?
    Nuwan Goonasekera
    @nuwang
    use helm install —set postgresql.galaxyDatabasePassword=password <repo>/galaxy. You’ll have to perform a fresh install
    jashwantraj92
    @jashwantraj92
    i did del --purge for galaxy and again installed freshly
    i still get the same error
    Nuwan Goonasekera
    @nuwang
    What chart repo are you using? Try this:
         - name: cloudve
           url: https://raw.githubusercontent.com/CloudVE/helm-charts/master/
    jashwantraj92
    @jashwantraj92
    i use the same chart repo
    then again did "helm install --set postgresql.galaxyDatabasePassword=password --name galaxy"
    but i get the same issue
    jashwantraj92
    @jashwantraj92
    what is the password set by galaxysecretkey
    becuase my postgress password is being acquired from "galaxy.createGalaxyDbSecret"
    Nuwan Goonasekera
    @nuwang
    it should be the same… inspect the secret value to make sure
    Alexandru Mahmoud
    @almahmoud
    @jashwantraj92 If you're specifying --name galaxy some of the resources from the previous install might not have been deleted and could be reused (the database PV in particular perhaps)
    jashwantraj92
    @jashwantraj92
    I made sure all pods are deleted before installing again
    anyway let me double check
    Alexandru Mahmoud
    @almahmoud
    not pods, PVs/PVCs
    jashwantraj92
    @jashwantraj92
    I have problem with that too. The PVCs do not get automatically created, each time i need to create it manually. Though I installed NFS as default storage
    Alexandru Mahmoud
    @almahmoud
    The PVCs or PVs ?
    The claims should be part of the chart
    The PVs are the ones that should be created automatically based on the class
    Make sure, however, that the NFS storage request on the actual volume is bigger than the galaxy requests on NFS if that makes sense
    Alexandru Mahmoud
    @almahmoud
    Can we get shared data into a GVL easily through admin panel?
    @nuwang @afgane
    Nuwan Goonasekera
    @nuwang
    You mean the galaxy admin panel?
    Alexandru Mahmoud
    @almahmoud
    Yes
    Nuwan Goonasekera
    @nuwang
    data libraries?
    Alexandru Mahmoud
    @almahmoud
    Yes, like Alex O was wondering if there is any easy way to get libraries installed to get the shared data for tutorials from zenodo
    Nuwan Goonasekera
    @nuwang
    so without pasting multiple urls in the upload dialog, just have the data already available in a library?
    Alexandru Mahmoud
    @almahmoud
    Yeh, if it's possible to import a whole library
    Enis Afgan
    @afgane
    @almahmoud, I don’t think so. There is no concept of Export / Import for a library. It would need to be a scripted approach (e.g., via BioBlend) for a dynamic solution at startup, or we create the library before baking it into the image so every launch would have it.
    Dannon
    @dannon
    I thought we had some bootstrapping code that'd do it programmatically at some point, specifically for workshop prep and the like?
    Enis Afgan
    @afgane
    I know we talked about it a bunch but, if we had the code, I’m guessing it would need to be updated...
    Dannon
    @dannon
    Hrmm. Yeah, maybe it never made it into an actual repo :)
    John Chilton
    @jmchilton
    I'm sure it isn't exactly what you're looking for but ephemeris has a starting point https://github.com/galaxyproject/ephemeris/blob/master/src/ephemeris/setup_data_libraries.py - I think most of the training material is distributed with files compatible with this script (and the Upload 2.0 API which can create whole libraries at once)
    Enis Afgan
    @afgane
    This will be easier than direct BioBlend. Cool. Thanks.
    Nuwan Goonasekera
    @nuwang
    It looks like AWS has an instance types API now? There’s a new tab in the console.