by

Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Marius van den Beek
    @mvdbeek
    I guess that’s the main reason
    On the other hand simple stuff like subclassing a datatype is actually pretty nice and won’t cause any harm
    Nuwan Goonasekera
    @nuwang
    @mvdbeek So what do you recommend for this? Should we consider this as needing to be fixed at the Galaxy level, or fixed at the tool level? Should the tool author submit the datatype to Galaxy through a PR perhaps?
    Marius van den Beek
    @mvdbeek
    The PR is ready to go, is there something else that is missing ?
    Nuwan Goonasekera
    @nuwang
    You mentioned though that the other handlers won’t sync?
    Marius van den Beek
    @mvdbeek
    I see. No, we can fix that too
    But I wouldn’t backport that to 20.01
    Nuwan Goonasekera
    @nuwang
    Sounds good. I think a future-facing fix sounds fine too..
    Marius van den Beek
    @mvdbeek
    We can probably get that into 20.05, which should be right around the corner
    Nuwan Goonasekera
    @nuwang
    Ok great… We’ll probably need to build a container for that pretty much immediately anyway
    Pablo Moreno
    @pcm32
    @nuwang @almahmoud is there documentation somewhere that I could follow to make our single cell Galaxy setup deployable with CloudMan (assuming that that is already using the current helm charts, which I presume is the case). Thanks!
    Alexandru Mahmoud
    @almahmoud
    I don't think there is documentation, but can manually add it as an extra appliance in CloudLaunch if that's what you're looking for
    Pablo Moreno
    @pcm32
    and how do you do that?
    Alexandru Mahmoud
    @almahmoud
    You'd need admin access to CloudLaunch
    Is it just values changes to the galaxy chart?
    Nuwan Goonasekera
    @nuwang
    Alexandru Mahmoud
    @almahmoud
    Like what is that "single cell Galaxy setup" vs default Galaxy, just to understand how involved it would be
    Nuwan Goonasekera
    @nuwang
    This defines all the apps listed on CloudLaunch. You can copy/paste the GVL template, customise the containers etc. and you should be good to go?
    Pablo Moreno
    @pcm32
    mostly yes, it should be values in the helm chart. Well, it has the selection of tools and the layout for them that we want Alex. Also, I’m not sure if our user’s toolshed tools (ebi-gxa) are being added into the “default Galaxy”. But also includes branding, our dynamic destinations (which should have resubmissions working by now), etc.
    Is this something that would be within the scope of what CloudMan/CloudLaunch offers to its users?
    I’d rather have a UI based deployment like the one I presume we would get from cloudman rather than sending our users to our cli based setup scripts that deploy to GCP or AWS.
    thanks guys!
    Pablo Moreno
    @pcm32
    So on that yaml @nuwang, is any of those an example of a helm based deployment? to understand how it maps to that.
    (I searched for helm, but probably found a lot of things that are not Galaxy)
    Alexandru Mahmoud
    @almahmoud
    If you look here: https://github.com/galaxyproject/cloudlaunch-registry/blob/master/app-registry.yaml#L2152 It's the config that is being sent from CloudLaunch to CloudMan which then gets used by Galaxy
    You could PR a change to the values there, then I can take the values from there and make a separate app, unless you want to add a whole app entry
    Nuwan Goonasekera
    @nuwang
    That makes sense I think. You can then just add it as a GVL app through an install template
    Nuwan Goonasekera
    @nuwang
    @mvdbeek Thought of pinging you about this since @zoeangel asked about this fix. First PR seems to have been merged: galaxyproject/galaxy#9835 Does the handler syncing issue remain?
    Marius van den Beek
    @mvdbeek
    Yes, I don’t have time to work on this.
    Nuwan Goonasekera
    @nuwang
    Ok. I guess the alternative fix is to ask that the tool be updated and a PR made for the datatype?
    Marius van den Beek
    @mvdbeek
    Just restart Galaxy seems way easier :)
    It’s not a hard fix, you’d need to figure out how to reload the datatypes registry and send a control task if you install a new datatype
    Nuwan Goonasekera
    @nuwang
    I’ll take a look if I can find some time, but at least in the immediate future that seems unlikely. A restart is a reasonable enough fix for now, especially since the whole mechanism is deprecated?
    Marius van den Beek
    @mvdbeek
    yep, that’s my thinking
    Nuwan Goonasekera
    @nuwang
    Ok. So just a matter of rebuilding the image with 20.05. @zoeangel We’ll ping you once this is done.
    Kiran Telukunta
    @telukir_gitlab
    I would like to test GVL on the Openstack and right now it does not work totally and there are issues in Cloudbridge (which is abstraction layer over different IaaS) which is related to security i.e sharing of keys to the cloudlaunch .. just wanted to confirm my understanding with the readings in chat and cloudbridge github
    Alexandru Mahmoud
    @almahmoud
    What types of errors related to sharing of keys to cloud launch? Also, what keys are you referring to?
    We have it working with OpenStack on Jetstream and Openstack on Nectar, so it would be likely from either a specific limitation of your cloud, or something related to your configuration
    Alexandru Mahmoud
    @almahmoud
    I think the plan was to dedicate a day during cofest to get up and running for multiple people on their custom openstacks, let's work on it together, it should be doable unless there's something very different about your cloud
    Kiran Telukunta
    @telukir_gitlab
    ok .. then I was looking old bugs of openstack .. its great if it is working .. yes, I was looking for cofest ..
    I have not yet tried in openstack but would like to try .. as you were saying I need to give openstack configuration .. please let me know which information is required for cloudlaunch
    Alexandru Mahmoud
    @almahmoud
    Where were you looking out of curiosity? If there are outdated issues on CB that imply things don't work we can/should close them
    The info that we need is Auth URL, region name & ID, and zone name & ID as applicable
    If there are different zones for different resources (eg: we have zone-r1, zone-r2 for compute and nova for storage/networkin) we need that info to make a mapping
    Alexandru Mahmoud
    @almahmoud
    image.png
    For reference/example from our use
    Essentially if you provide that information we can try a launch same as we do for our OS clouds, after which we can see if any problems arise from your specific configuration and address them as they come up
    If you're having problems finding any, lmk and I can help you locate them