I was trying to do the tutorial for DIA Analysis using OpenSwathWorkflow
After reaching the point of converting the raw files to mzml i get this message:
CAN SUDO 1
/data/jwd06/main/087/057/87057919/tool_script.sh: line 10: wine64: command not found
Any ideas?
p.s. i am using the web browser galaxy platform
Welcome @K_A
Which server are you working at? UseGalaxy.eu or somewhere else? This is a dependency issue in the Docker image. Maybe it needs to be updated. We can help if this is happening at a public server. But if this is happening at your own server, you’ll need to install the package.
Please let us know and we can follow up more! Thanks! 
Xref → Wine64 not found (Galaxy docker) (public server) and Trouble running msconvert in prod galaxy:wine64_anyuser: command not found (private server)
it happens in the public platform please check the attached image. Sometimes it works, other times with same settings i get that error. I go one file at a time just to be sure.
Hi @K_A
Ok, thanks for sharing the screenshot! There might be a problem with a few of the cluster nodes at the EU server. Would you please send in a bug report from the error dataset to the administrators?
Bug icon → copy/paste the URL to the topic here into the comments → submit
They’ll see it and hopefully be able to make the correction. Let’s also ping them here as another notificaion. ping @wm75
Then, to get you going again, are you able to rerun this and get a sucessful result?
ok i submitted the bug
Yes slowly but steady I am proceeding . For now, looks like sample3.raw had some issues. Let’s see
commandline from error report:
ln -s ‘/data/dnb11/galaxy_db/files/3/4/4/dataset_344bc408-9564-4802-96b9-79597b10a53a.dat’ ‘Sample3.raw’ && CAN_SUDO=$(sudo -n -l > /dev/null 2> /dev/null || echo $?) && echo “CAN SUDO $CAN_SUDO” && if [ -z “$CAN_SUDO” ]; then uid=`id -u` && gid=`id -g` && WINE=“wine64_anyuser”; else WINE=“wine64” && export WINEPREFIX=$(mktemp -d) && (cp -a /wineprefix64/* $WINEPREFIX 2> /dev/null || true); fi && $WINE msconvert ‘Sample3.raw’ --outdir outputs --mzML --simAsSpectra --runIndexSet 0 --stripLocationFromSourceFiles --filter “peakPicking vendor msLevel=1-” --filter “demultiplex massError=10.0PPM nnlsMaxIter=50 nnlsEps=1e-10 noWeighting=false demuxBlockExtra=0.0 variableFill=false noSumNormalize=false optimization=none interpolateRT=true minWindowSize=0.2” --64 --zlib --outfile ‘Sample3’ && if [ -z “$CAN_SUDO” ]; then sudo chown -R $uid:$gid ‘./’; fi && mv ‘outputs/Sample3.mzML’ ‘/data/jwd07/main/087/119/87119698/outputs/dataset_0685b6af-638e-4a1f-9a37-f8372cca75c0.dat’
I think I half understand what’s going on, and we will try to debug this over the day.
Not exactly an easy situation, so please give us some time to do this properly.
Thanks for reporting the issue,
Wolfgang
Thanks for your time. I don’t think i can process some files, i will just proceed with less in order to complete the training 
Tentatively declaring this solved.
Neither me, nor @K_A have seen failing msconvert jobs anymore on Galaxy Europe after we applied what we think is the fix for the issue.
Explanation of what we think was going on: the tool was not running containerized on all our job destinations (we still do not know exactly why), but is bound to fail when not containerized. We are now enforcing containerized execution and have not seen failing jobs since.
@K_A feel free to come back and report future runs failing with the original error if they should occur.
