Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
  • 14:19
    aledegano labeled #2575
  • 14:19
    aledegano opened #2575
  • May 13 15:31
    seanrmurphy review_requested #2574
  • May 13 15:31
    seanrmurphy opened #2574
  • May 13 15:25

    seanrmurphy on documentation_updates_secret_generator_tool

    docs: minor clarifications to d… (compare)

  • May 13 12:50
    ciyer closed #2568
  • May 13 12:50
    ciyer closed #2531
  • May 13 12:50
    ciyer synchronize #2573
  • May 13 12:50

    ciyer on 000-tests-ui-2_0-next

    test: update the project visibi… fix formatt change description input for te… and 2 more (compare)

  • May 13 12:48
    ciyer opened #2573
  • May 13 12:47

    ciyer on 000-tests-ui-2_0-next

    (compare)

  • May 12 09:54
    lorenzo-cavazzi review_requested #2562
  • May 12 09:47
    lorenzo-cavazzi synchronize #2562
  • May 12 09:47

    lorenzo-cavazzi on 2550-docs-porject-settings

    docs: add changes from review (compare)

  • May 11 14:00

    github-actions[bot] on 0.12.10-2.e711ddd

    (compare)

  • May 11 13:54

    rokroskar on 0.12.10

    (compare)

  • May 11 13:53

    rokroskar on prep-0.12.10

    (compare)

  • May 11 13:53

    rokroskar on master

    chore: prep 0.12.10 (#2572) (compare)

  • May 11 13:53
    rokroskar closed #2572
  • May 11 12:27
    andre-code synchronize #2568
Simon Blanchoud
@pintomollo
I'm just starting with Renku and I'm running into a really basic issue. I'm trying to migrate my local code developed using Conda. I updated environment.yml to include my required packages (I've tried installing them manually too). I can indeed see them there (conda list) but I cannot access them. I mean if I try to import the package in a notebook or try to call the functions directly in the terminal, it always says "Name not defined". Thanks in advance :D
13 replies
Philippe Grosjean
@phgrosjean
Help! My problem is still there, and I really don't understand what is wrong. On 2021-08-25, I build a functional image (registry.renkulab.io/sciviews/svbox-renku:3c32dc5). There, RStudio works fine, and I have `/usr/lib/rstudio-server/bin/rserver --www-port=8787`that runs well, and I see this:

`ls -l /var/run/rstudio-server/revocation-list` that gives
`-rw------- 1 rstudio users 0 Aug 26 19:03 /var/run/rstudio-server/revocation-list`

All docker images I compiled after that have broken RStudio. So, I revert back to same `Dockerfile`, same `install.R`, same everything than the previous working one... and I still got a broken RStudio with the message

"500: Internal Error, could not start rstudio in time"

(this is registry.renkulab.io/sciviews/svbox-renku:d59816e). On a terminal from Jupyterlab, I have:

`/usr/lib/rstudio-server/bin/rserver --www-port=8787
27 Aug 2021 14:06:27 [rserver] ERROR system error 1 (Operation not permitted) [path: /var/run/rstudio-server/revocation-list, description: Could not set revocation file permissions to 600 for file: /var/run/rstudio-server/revocation-list]; ...`
and

`ls -l /var/run/rstudio-server/revocation-list
-rw------- 1 rstudio-server rstudio-server 0 Aug 26 17:01 /var/run/rstudio-server/revocation-list`

Obviously, RStudio is started with the wrong (default) "rstudio-server" user, but it should be "rstudio". How two compilations of Docker images with same data could give two different results?
25 replies
Simon Blanchoud
@pintomollo
Hi, I have observed a really strange behavior with conda packages in Renku when editing the environment.yml file, namely that the package trinity (in bioconda) takes forever to install itself (I have tried to install them manually and it works well). I have tried it on a totally fresh project, and installing this package only. It always take more than 1h to build, while separately the dependencies of trinity take less than 5 min to build. Looking at the logs of the build in GitLab did not provide much information as it is stuck at the "Solving environment: ...working...". Is there a way to get some more details about what is causing this delay?
5 replies
Simon Blanchoud
@pintomollo
Hi, I am forking my project to several students and one of them has a seemingly endless stalling saying "Identifying project number". I have checked in Github and the project seems to be forked properly and the docker image was build successfully. Link to the session is here: https://renkulab.io/projects/18027263/maristem-omics/sessions
6 replies
Simon Blanchoud
@pintomollo
Hi there, yet another issue here. My last build do not complete properly, while the only thing I think I have changed was a minor spelling mistake in one of my notebook. The error is a bit surprising as it relates to pip, which I do not use at least explicitly, and to the renku package itself.
Screenshot_2021-09-07_16-35-03.png
8 replies
Olivier Levitt
@olevitt
Hi ! I'm trying to deploy renku on our (on-premise) kubernetes cluster. I started by running make-values.sh providing my gitlab url & application credentials. Then I installed renku using helm providing the generated values. Every pod runs fine except for the renku-gateway-auth one which fails with "Obtaining public key from https://<redacted>/auth/realms/Renku" which leads me to believe that something is wrong with the keycloak configuration (note that I have not done any specific configuration except for the make-values.sh). Browsing https://<redacted>/auth/realms/Renku gets me a keycloak page with "We are sorry...
Page not found".
Any tips on what to configure for keycloak ?
11 replies
Simon Blanchoud
@pintomollo
Hi, I'm trying to run some R code inside my Jupyter notebook, currently with little success. I've decided to use rpy2 (installed through conda) and I have installed the necessary R packages through install.R. Both work fine, but apparently separately. If I run :

%run -m rpy2.situation

rpy2 version:
3.4.5
Python version:
3.9.5 | packaged by conda-forge | (default, Jun 19 2021, 00:32:32)
[GCC 9.3.0]
Looking for R's HOME:
Environment variable R_HOME: /usr/local/lib/R
Calling R RHOME: /opt/conda/lib/R
Environment variable R_LIBS_USER: None
Warning: The environment variable R_HOME differs from the default R in the PATH.
R's additions to LD_LIBRARY_PATH:
/usr/local/lib/R/lib
R version:
In the PATH: R version 4.1.0 (2021-05-18) -- "Camp Pontanezen"
Loading R library from rpy2: OK
Additional directories to load R packages from:
None
C extension compilation:
include:
['/usr/local/lib/R/include']
libraries:
['R', 'pcre2-8', 'lzma', 'bz2', 'z', 'rt', 'dl', 'm', 'icuuc', 'icui18n']
library_dirs:
['/usr/local/lib', '/usr/local/lib/R/lib']
extra_compile_args:
[]
extra_link_args:
['-Wl,--export-dynamic', '-fopenmp']

I believe this shows that there are two separate installations. I can run the rpy2 one by changing the R_HOME variable (os.environ['R_HOME'] = '/opt/conda/lib/R') but I then do not have access to the packages installed by R. So my question is, should I just add the R folder to my library folder to use the packages installed by install.R, or is there a more direct way to use the exact same R from the start? I have tried to set the home to the conda folder inside install.R, but it tells me that the folder is not writable.

Thank

13 replies
Simon Blanchoud
@pintomollo
Screenshot_2021-09-14_18-38-58.png
Hi again :D
I seem to have corrupted a bit my project somehow. It does not prevent it from working (or so that I can notice), but at every commit, I get a lot of errors like these above.
3 replies
Eric Pohl
@pohleric
Hello, I have a question on how students can fork a certain environment. I did the following: 1) Create an environment with a docker file X1, 2) have a group of students A (last year's students) fork this environment X1, 3) forked the environment X1 myself and make adjustments for a new course X2, 4) have my new students B (this year's course) fork X2.
The problem is now that students who attended the first course cannot properly fork X2. They end up getting X1 even though they choose to fork X2. This is probably a gitter question. Does anyone know what to do so students A (now in group B) can fork X2?
5 replies
Antonio Pérez
@aperezvelasco

Hi there! I have tried to clone a repository from a challenge names "s2s-ai-challenge" and it gives me an authentication error when using the renku clone command:
renku clone https://renkulab.io/gitlab/antonioperezvelasco/s2s-ai-challenge-predictia.git

I have tried to use my e-mail and my username with the corresponden password, but I'm not able to solve the error

1 reply
Eric Pohl
@pohleric

I have two out of 38 students who told me they could not save their work. They would get the attached message when they try to git push the tracked, and committed changes. Any help is appreciated:

Locking support detected on remote "origin". Consider enabling it with: $ git config lfs.https://renku.unifr.ch/gitlab/<student_name>/sp21-sgg-00272.git/info/lfs.locksverify true To https://renku.unifr.ch/gitlab/<student_name>/sp21-sgg-00272.git ! [rejected] HEAD -> master (fetch first) error: failed to push some refs to 'https://renku.unifr.ch/gitlab/<student_name>/sp21-sgg-00272.git' hint: Updates were rejected because the remote contains work that you do hint: not have locally. This is usually caused by another repository pushing hint: to the same ref. You may want to first integrate the remote changes hint: (e.g., 'git pull ...') before pushing again. hint: See the 'Note about fast-forwards' in 'git push --help' for details.

3 replies
Tasko Olevski
@olevski

Hello everyone, as you may have noticed by the recent notifications, we are planning to release a new version of Renku on renkulab.io tomorrow afternoon at 16:00 CET. This new version of Renku brings significant changes and improvements but it also requires us to shut down all active user sessions before doing the upgrade. Therefore please make sure you shut down any active sessions you may have on renkulab.io by 16:00 CET tomorrow (November 4th). If you leave any sessions running when we start the update process any data or code in them that has not been saved will be lost.

Please let me know if you have any questions or concerns. The update process should be finished within a few hours at most. We do not anticipate that similar interruptions will occur any time in the near future.

2 replies
Pamela
@pameladelgado

Dear renkulab users,

the Renkulab 0.10.0 rollout went fine and you can now enjoy working with Renku again :tada:

Also, in case you were not aware: there was a small outage this morning on renkulab.io, see the post-mortem, if you had open sessions it's possible that unsaved work was kept in an auto-saved branch on your project. See this post for more details on how to potentially recover unsaved work and how the auto-save feature works.

Geert van Geest
@GeertvanGeest

Hi! I'm rather new to renku, and I'm trying to build an image based on this environment.yml: https://renkulab.io/gitlab/geert.vangeest/ngs-introduction-training/-/blob/master/environment.yml . However, during building it gives the following error:

UnavailableInvalidChannel: The channel is not accessible or is invalid.
  channel name: bioconda
  channel url: https://conda.anaconda.org/bioconda
  error code: 404
You will need to adjust your conda configuration to proceed.
Use `conda config --show channels` to view your configuration's current state,
and use `conda config --show-sources` to view config file locations.

e.g. adding conda config --add channels bioconda to the Dockerfile did not help. How should I proceed?

9 replies
markrobinsonuzh
@markrobinsonuzh
Screenshot 2021-11-09 at 15.01.08.png
Hi, I'm trying to run a (fairly standard?) instance, but it's been waiting this point (screenshot above) for a long time. I ran one earlier this morning and it was fine. The project is here: https://renkulab.io/projects/mark.robinson/sta426hs2021-r .. Any ideas?
31 replies
Pol del Aguila Pla
@poldap
Hi. Noob question here. I wanted to try to get a bit more resources for a session and now I am stuck waiting. I'd rather reduce the resources and work now than wait, but I can not launch a new one or modify the one that is waiting to start.
project: https://renkulab.io/projects/pol.delaguilapla/lp-stability-examples (nothing to remark, just started it)
3 replies
ansonrel
@ansonrel
Hi, I'm trying to start a new session on one of my projects (https://renkulab.io/projects/omnibenchmark/omni_data/omni-batch-processed) but it has been pending for >6 hours. Strangely, a colleague could create a session on this project without any issue. Also, I couldn't get any logs (No logs available for this pod yet.). Do you know what is happening ?
2 replies
Raphaël
@r-fndv

Hi there, I did not use one of my Renku project (https://renkulab.io/projects/rdefondeville/mvPot_tutorial) for while and I am requested to migrate it. When I click on the warming sign, it is recommended to launch a session, follow some instruction (page is not found) and use "renku migrate".

I am unfortunately not able to start any session and the page of instruction being not available does not help me figuring out how I can do this migration. Could you please help me updating the project? Thanks in advance!

2 replies
Jay Beard
@Jaybeard_gitlab
Hi, my new session has been stuck starting for a few days. @ableuler I think I could use some help. (https://renkulab.io/projects/floorvandenheuvel/conso2/sessions)
3 replies
csulkowskibfs
@csulkowskibfs
Hi. I did not realize that enabling git-lfs pull during session creation in repos with very large datasets could get it stuck. Would really appreciate some help removing this hanged instance (https://health.renkulab.datascience.ch/projects/fso_dscc/astra). Thanks in advance!
3 replies
Rok Roškar
@rokroskar
Hi everyone, in case you don't follow our discourse and/or twitter accounts, we have another Renku Tea Talk session today at 15:00 CET, which will focus on everything new and exciting in the recently-released Renku CLI 1.0. You can find more details here: https://renku.discourse.group/t/renku-talks-2021-2022/469/5
James Runnalls
@JamesRunnalls
Hi! I am trying to set up Matlab and I am having some problems with the Docker file. I have directly cloned https://renkulab.io/projects/learn-renku/working-with-matlab and pushed it to a personal repository however, the build is failing (https://renkulab.io/gitlab/james.runnalls/matlab-test/-/jobs/344781). Does anyone know how to fix this? Thanks!
8 replies
Waldo Rivier
@waldorivier_gitlab
Hello Team, is there a way to upgrade jupyterlab to 3.0 ?
I try to upgrade it but I got following error
image.png
thanks for your help. Waldo.
Rok Roškar
@rokroskar
Hi @waldorivier_gitlab - it looks like you are using a very old base image in this project. You should be able to update your projects to use a newer base image - the latest one is renku/renkulab-py:3.9-0.10.3 for python and it already includes jupyterlab 3.0. The base image is specified on the ARG RENKU_BASE_IMAGEline in the Dockerfile in your repository - so right now you probably have something like ARG RENKU_BASE_IMAGE=renku/renkulab-py:3.7-renku0.10.4-0.6.3 and it should read ARG RENKU_BASE_IMAGE=renku/renkulab-py:3.9-0.10.3. Note that this will also change your python version.
2 replies
Noe Thalheim
@noeleont
Hei everyone I'm not sure if this is the right channel but I'm always getting the following error when trying to login via eduid: We are sorry... invalidFederatedIdentityActionMessage I did a quick search and the following came up: https://stackoverflow.com/a/63249672. Is this only happening for me?
Best regards
Noe
2 replies
Waldo Rivier
@waldorivier_gitlab
image.png
@waldorivier
Hi all, I got error above building docker image. Thanks for your help. Cordialement. Waldo
Eugenio Naselli
@eunaselli:staffchat.ethz.ch
[m]
Hello 👋
I have one question about k8s compatibility
what's the correct version for 0.12 ?
is Kubernetes v1.22.5 fine?
2 replies
marine-manche
@marine-manche

Hello,
I have a problem with my Renku project. I'm working with heavy datasets and I don't manage to unzip heavy .bz2 files to pickle them. When I am running my code, the kernel automatically restarts after a few minutes.
I am working with an 8G memory and 64G storage session and the size of the datasets I don't manage to download are about 500MB when they are zipped.
My code works for lighter datasets (below 500MB) so I think it's a size problem. Is there a way to open a session with more resources?

Thanks in advance for your help!

3 replies
Noe Thalheim
@noeleont
Hello everyone
Has someone experience with sync user forks with a upstream repo? I tried using the following approach inside the post-init.sh script, but since the container has no network at this stage this doesn’t work. Currently I’m testing a cron based approach but this doesn’t seem to be the right direction.
4 replies
Waldo Rivier
@waldorivier_gitlab
Hello everyone, could somebody help me for the problem I have submitted below as 22.02.2022 ? thanks a lot for replying and have a nice day. Waldo.
Tasko Olevski
@olevski

Hello everyone, could somebody help me for the problem I have submitted below as 22.02.2022 ? thanks a lot for replying and have a nice day. Waldo.

Hi Waldo you are referring to your image not building in one of your projects? Have you tried simply re-running the CI job? It seems to me like this could be an intermittent problem related to maybe low resources or problems with our Gitlab deployment.

7 replies
ansonrel
@ansonrel
Hello, I think that new Renku projects with R/Bioconductor templates have a problem with the installation of Renku: bash: renku: command not found. I have tested new project with Python/ Minimal Renku and it seems ok. Not sure, but I think it's not only for v1.1.1 but also for prior versions. Does it also happen with other people .. ?
3 replies
Eugenio Naselli
@eunaselli:staffchat.ethz.ch
[m]
hi @rokroskar
I've opened today SwissDataScienceCenter/renku#2523
i think it would be a nice improvement for all kind of deploys
Rok Roškar
@rokroskar
great! Does this mean you'll be submitting a PR :)
Eugenio Naselli
@eunaselli:staffchat.ethz.ch
[m]
It's more of a design choice on your side for this