Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Theodore Aptekarev
    @piiq
    Hi! I am looking for a way to make SlicerJupyter kernel (made using xeus) available in a Google Cloud managed Jupyter Lab instance (they call them AI Notebooks).
    Can someone advice if i can use xeus-python kernel from a docker container?
    Wolf Vollprecht
    @wolfv_:matrix.org
    [m]
    @piiq: can you use conda packages there?
    4 replies
    Nicholas Devenish
    @ndevenish
    I want to get the mamba list but for the environment that would be installed by mamba install, because I want to get the resolved package list for a different platform. Is there a way to get this, and does it mean digging into the mamba API?
    Wolf Vollprecht
    @wolfv
    @ndevenish you can just use --dry-run
    and to change the platform you can use CONDA_SUBDIR=win-64 mamba create -n blabla mypackage --dry-run
    to get a proper list is ... probably not easily doable right now
    but you can get JSON output with --json and then parse that further
    however, note that there are some virtual packages dependent on the platform you are on, and that might change the resolution. conda-lock has implemetned these things properly: https://github.com/conda-incubator/conda-lock
    Nicholas Devenish
    @ndevenish
    @wolfv thanks, I tried --json but unfortunately none of the variants I tried gave useful information - CONDA_SUBDIR (instead of manual channels and --override-channels) and --dry-run instead of --download-only seem to have done the trick, thanks! (for building bundled releases we "pin" everything in the conda dependency list and not all our platforms are available in our CI)
    Wolf Vollprecht
    @wolfv
    yeah, conda-lock would be ideal for that.
    Nicholas Devenish
    @ndevenish
    yes, this looks to cover the cases we're trying to do. So you both solved and eliminated the need for my question :)
    Dave Hirschfeld
    @dhirschfeld
    Just checking - with a clean mamba install do you still have to use conda init to initialise your shells?
    i.e. there's no mamba equivalent (and I shouldn't be using micromamba)?
    Wolf Vollprecht
    @wolfv
    you can use mamba init these days and it will be better
    Dave Hirschfeld
    @dhirschfeld
    Thanks @wolfv - will give that a go!
    mcmaclachlan
    @mcmaclachlan
    Hi! voila is not rendering the ipywidgets in my Jupyterlab notebook when I run voila on apache web server. Any special considerations when deploying voila on a server that I might be missing? I'm seeing failure to excute/load js and css files errors in the javascript console.
    David Brochart
    @davidbrochart
    Hi @mcmaclachlan, maybe that helps: https://voila.readthedocs.io/en/stable/deploy.html#running-voila-on-a-private-server
    mcmaclachlan
    @mcmaclachlan
    Thanks @davidbrochart, I've followed all the steps here and I've deployed the same notebook on mybinder with all widgets showing successfully, and everything works fine when I run voila as my own user as well. The problems show up when running voila with apache
    Wolf Vollprecht
    @wolfv
    Do you have anything in the Apache logs?
    mcmaclachlan
    @mcmaclachlan
    @wolfv nothing relating to voila that I can see.
    Wolf Vollprecht
    @wolfv
    maybe it's a problem with websocket connections through the apache server?
    you might need mod_proxy_wstunnel ?
    Anything in the javascript console on the webpage?
    https://stackoverflow.com/questions/17334319/setting-up-a-websocket-on-apache
    mcmaclachlan
    @mcmaclachlan
    @wolfv These are the errors in the javascript console:[Error] Refused to execute https://studmed.uio.no/voila/static/require.min.js as script because "X-Content-Type-Options: nosniff" was given and its Content-Type is not a script MIME type.
    [Error] Failed to load resource: too many HTTP redirects (require.min.js, line 0)
    [Error] Failed to load resource: too many HTTP redirects (index.css, line 0)
    [Error] Failed to load resource: too many HTTP redirects (theme-light.css, line 0)
    [Error] ReferenceError: Can't find variable: requirejs
    Global Code (voila:77470)
    Wolf Vollprecht
    @wolfv
    looks like your apache config is wrong?!
    the content type of require.min.js is wrong?
    mcmaclachlan
    @mcmaclachlan
    @wolfv the require.min.js is a voila static file?
    Wolf Vollprecht
    @wolfv
    yes, it should return a js file with content type application/javascript or somethign along those lines
    mcmaclachlan
    @mcmaclachlan
    @wolfv sorry I'm very new to this, where would the error be? what would I be looking for?
    Wolf Vollprecht
    @wolfv
    If i visit the page it redirects to https://studmed.uio.no/voila/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/files/static/require.min.js
    and then says: ERR_TOO_MANY_REDIRECTS
    you've got to check the apache config
    mcmaclachlan
    @mcmaclachlan
    @wolfv what am I looking for in the config?
    Wolf Vollprecht
    @wolfv
    no idea
    would also have to google
    mcmaclachlan
    @mcmaclachlan
    @wolfv :) thanks for your help and for the mod_proxy_tunnel tip, I'll look into that as well. The rest of the notebook does render though https://studmed.uio.no/voila
    Wolf Vollprecht
    @wolfv
    yeah you just ahve some problem with all the static files
    Adam Scislowicz
    @adam-scislowicz
    I notice conda-build and conda are not very active. Is there a good place to go for help with a conda-build issue. I think it is just some environment interaction when conda calls cmake vs when I call cmake by way of pip. e.g. Its my first time making a conda package. I am using both conda build and conda mambabuild and I'm having a few issues, I think due to interactions with CMAKE and the sysroot used by conda-build. ... Also if anyone wants a quick project I can pay well for the work to get the package building. Note: the issue is just as likely in my CMakeLists file as it might be in my conda build artifacts. I apologize if this is the wrong channel to mention this.
    2 replies
    Adam Scislowicz
    @adam-scislowicz
    The conda-build and conda channels here rather, not the source code repositories.
    Wolf Vollprecht
    @wolfv
    @adam-scislowicz i don't completely understand your problem from this message :)
    do you have a public build script anywhere?
    Adam Scislowicz
    @adam-scislowicz
    @wolfv I don't, it is a private code base. I can whip up a really simple contract for paid access to work with it. A simpler summary, when I build my python package using pip (and thus setup.py) it triggers CMake internally and links correctly. However when I attempt build using conda, which populates a sysroot with the specified versioned components and then executed setup.py it fails to link to the pthread library (but it does detect it), even though I do see that present in the conda sysroot used by the build. I assume I'm not properly detecting the environment within CMake, or not properly excluding the host environment, etc.
    Wolf Vollprecht
    @wolfv
    @adam-scislowicz that might be a problem with Centos 6. Are you using the conda compilers?
    you can install compilers from conda-forge for a quick way to get them
    Adam Scislowicz
    @adam-scislowicz
    @wolfv I'm using debian 11 currently and I do specify {{ compiler('cxx') }} and the same for c in the build requirements section of meta.yaml. However I wasn't aware of the compilers package. Is that something else?
    Wolf Vollprecht
    @wolfv
    you could just add -pthread to the LDFLAGS or CXXFLAGS?
    e.g. export LDFLAGS="$LDFLAGS -pthread may work
    Adam Scislowicz
    @adam-scislowicz
    @wolfv indeed. I may be able to resolve it by forcing that instead of letting cmake detect the pthread method.
    Wolf Vollprecht
    @wolfv
    also you can get verbose build output with -DCMAKE_VERBOSE_MAKEFILE=ON