#galaxyproject

/

      • guerler joined the channel
      • bmpvieira joined the channel
      • guerler has quit
      • guerler joined the channel
      • guerler has quit
      • guerler joined the channel
      • guerler has quit
      • guerler joined the channel
      • trondth joined the channel
      • trondth has quit
      • trondth joined the channel
      • g2roboto
      • [bgruening] look at the date :(
      • [nsoranzo] Ah, GCC2017
      • [bgruening] :( They invited us personally ... but this date is really not good.
      • [nsoranzo] It's a shame
      • [bgruening] I will try to send Yasset from BioContainers or my PI.
      • guerler has quit
      • guerler joined the channel
      • guerler has quit
      • [MingChen0919] good morning, everyone. I have a galaxy instance on jetstream. but it doesn't run any jobs. And i found this troubleshooting page: https://galaxyproject.org/cloud/jetstream/troub.... but i can't find the script `fix_hostname.sh`.
      • natefoo
        @enis: ping ^^
      • eskimojoe joined the channel
      • eskimojoe
        Quick question. We used to be able to upload large files to our galaxy instance by directly linking to a file on the server it was running on. Has this feature been removed?
      • g2roboto
        [martenson] eskimojoe: this is still supported, it uses data libraries import
      • eskimojoe
        Awesome, thanks!
      • cganote joined the channel
      • g2roboto
        [jcirish] Hello! I've just joined, I'm stuck at an error when trying to run a custom tool I've built for galaxy. Interestingly, a couple years ago I successfully built a few simple custom tools, however I seem to have forgotten some important step (although following the custom tool building tutorial also results in the error). The error message seems to indicate that the Cheetah template system cannot "find" the name I
      • guerler joined the channel
      • [mvdbeek] https://gist.github.com/ is widely used
      • [jcirish] Thanks! I'll put something together
      • guerler has quit
      • [enis] @MingChen0919 The script is in /etc/init.d (I updated the docs to include that). Let me know if running the script does not get the jobs running.
      • [jcirish] Gist containing my error stacktrace and the tool script, config, and tool_conf files: https://gist.github.com/jcirish/fa7ee1681af5b7f...
      • [mvdbeek] hmm there is not `outputsam1` in your tool xml, but the error says you somehow referenced `outputsam1`, are you sure the error corresponds to the current version of your tool? Maybe this is the error for a previosu version of your tool?
      • [martenson] @jcirish how old is that Galaxy? Maybe you need to restart it to reload the tool if you changed it.
      • [jcirish] Sorry for the confusion: I put a toy version in the gist, the true script has "$outputsam1" as an output parameter.
      • [martenson] Can you put the full version to gist?
      • [jcirish] Crap, the stack trace is for the actual script; I'll load up the info for that.
      • [jcirish] My bad.
      • dave_b
        humans can be rubber ducks sometimes, too
      • g2roboto
        [jcirish] @Martin I've tried both the older Galaxy I was using a while ago, and the newest (I think) version pulled from GitHub a few days ago. Same error in both versions.
      • [mvdbeek] also `<command interpreter="bash”>` is deprecated, if you want to run a script that lives next to your tool, you would typicially use `<command>$__tool_directory__/myscript.sh</command>`. But that’s not causing the error you have posted
      • [jcirish] Good to know, thank you.
      • [mvdbeek] you have a typo, ouput should be outputs
      • [jcirish] Sheesh! I'm sorry guys!
      • [mvdbeek] I also recommend planemo for testing your tools, this is he output of `planemo lint`:
      • ```
      • Applying linter xml_order... CHECK
      • .. INFO: Unknown tag [ouputs] encountered, this may result in a warning in the future.
      • Applying linter tests... WARNING
      • g2roboto has quit
      • g2roboto joined the channel
      • [jcirish] This one was driving me crazy as my older tools were working fine. I'll check out planemo. Thanks again.
      • [martenson] @jcirish Good luck! Let us know if you encounter any problems.
      • [jcirish] Will do.
      • [nekrut] @bgruening
      • [nekrut] ``` docker run -d -p 8080:80 bgruening/galaxy-chip-seq-trainingUnable to find image 'bgruening/galaxy-chip-seq-training:latest' locally
      • docker: Error response from daemon: repository bgruening/galaxy-chip-seq-training not found: does not exist or no pull access.
      • See 'docker run --help'.
      • ```
      • [nekrut] i'm I doing smth wrong?
      • [nekrut] Hmmm...I actually do not see `bgruening/galaxy-chip-seq-training` in the docker hub
      • [nsoranzo] Some of these docker image still have to be created, AFAIK
      • [jcirish] Up and running! I feel silly about the simple typo, yet I doubt I would have found it any time soon without the help of this board. Thank you!!
      • [mvdbeek] cool!
      • [mvdbeek] we’ve all been there
      • [MingChen0919] Thanks @afgane! It works!
      • [martenson] Tool to convert shed_tool_conf.xml to yml? ping @afgane
      • [martenson] for use in docker image building, or is it just straight converion?
      • [martenson] conversion*
      • guerler joined the channel
      • [bgruening] @nekrut I need to create them. It will not take long, just need time
      • [bgruening] when do you need it?
      • [nekrut] I'll try myself
      • [nekrut] what is the passwd for admin@galaxy.org
      • [nekrut] in stable image
      • [blankclemens] `admin`
      • [nekrut] tx!
      • [nekrut] what do these correspond to on the host system:
      • [nekrut] ```# Mark folders as imported from the host.
      • VOLUME ["/export/", "/data/", "/var/lib/docker"]```
      • [bgruening] you can leave this out
      • [bgruening] @nekrut try to design it after this one: https://github.com/bgruening/galaxy-rna-seq
      • [nekrut] so `bgruening/galaxy-ngs-preprocessing:17.01` is better than stable?
      • [nekrut] more tools
      • [bgruening] better?
      • [bgruening] They all contain different set of tools
      • [bgruening] but yes, for chip-seq it is probably a good idea to base it on the ngs-preprocessing
      • [bgruening] the rna-seq on also is based on this
      • [nekrut] ok. what I want to create is a docker as close to main as possible
      • [bgruening] uha
      • [bgruening] I have done this locally, but this one is huge
      • [bgruening] you really want to have nearly all tools?
      • [bgruening] I think for the training it is better to keep specialized
      • [nekrut] perhaps
      • [bgruening] This was a try from 2015: https://github.com/bgruening/docker-usegalaxy
      • [bgruening] I have a better one locally somewhere
      • [blankclemens] @nekrut If that means containing as much as possible tools that main also has, you might find https://github.com/blankclemens/tools/blob/mast... interesting. A kind of a docker "clone" can be done in a few steps this way, but I don't think it is useful: For the Freiburg instance that was a 16GB Docker image.
      • [nekrut] I see. thanks. a.
      • [nekrut] does bgruening/galaxy-ngs-preprocessing:17.01 contain any reference data
      • [bgruening] No.
      • [bgruening] but the data-managers
      • [nekrut] so it needs to be configured beforehand
      • [bgruening] yes, or during the startup
      • [bgruening] we can include them into the training material by default
      • [nekrut] during startup?
      • [bgruening] I'm just worried about the size
      • [bgruening] We could trigger the data-managers during startup I suppose
      • [bgruening] My plan was to mount the cvmfs at some point for the training
      • [nekrut] so it can grab from there directly?
      • [bgruening] That was my plan, it's not yet there.
      • [nekrut] that would be awesome
      • [bgruening] :( I will get to it
      • [nekrut] so if I need a data manager I can just install it from tools hed?
      • [bgruening] yes, or add it into the tool.yaml and it will be build into the container
      • guerler has quit
      • [martenson] restarting Main
      • guerler joined the channel