Hello, I'm trying to run the CLIPseq pipeline on a...
# configs
l
Hello, I'm trying to run the CLIPseq pipeline on a torque cluster. I've a config file that I specified in my command line. I've also added the -w option to specify the working directory. The first jobs are running correctly (fastqc, cutadapt...) but the job get_software_versions return an error and I can't solve the problem. I tried to add a line scratch = 'path' in the process part of my config file but it doesn't change a thing neither setting all TMP variable that I could think of... Do you have any idea please ? I've been stuck for days :'(
j
Unfortunately I can't look properly myself as I have a sick kid - but if yo ucould send your
.nextflow.log
file this would be very helpful
l
Thanks for your answer and I hope you kid will get better soon. I'll sent the requested file as soon as I find it.
j
It should be whereever you execute teh command
(and it's a hidden file, just to be clear 🙂
ls -lha
should help)
(sorry if you already know this, but often people miss that)
l
ha yes ! I wasn't looking at the right place 🙂
Here, it seems to be the last one.
j
I have a spare 2 minutes, but fortunately I've found the most likely issue - looking at the command you have:
-profile singularity, cluster
There shouldn't be a space between the two profiles
And note you should specify to use singularity within the config, it's not necessary to specify that on the command line as well
l
Thanks. It seems worse right now but you were right that the cluster profile wasn't taken into account as now it says : " Unknown configuration profile : 'cluster'".
j
Him need to send config now
Hm*
And how are you specifying the config?
l
It's a very basic config file (first time I use a nextflow pipeline). I specify the config file with the option -c /data/users/clabbe/PATH/testPipelineNFCore/testClipAlina/curie.config I didn't change the nextflow config file : /data/users/clabbe/PATH/testPipelineNFCore/clipseq-1.0.0
j
Still with sick kid unfortunately and I need more than a phone screen to look at it :(
Hey @Li Las Back at work now! So I think the error/log/configs are out of sync now 😅
So please try again a fresh run and then send all the following files associated with THAT run: 1. The
nextflow.log
2. The command you use to execute 3. The error message 4. Any/all config files specified in that run
l
Hey @James Fellows Yates, thanks. Sorry for the delay. I was out of the office last week. I test that right now.
Here's the result :
If I try this with the cluster config, it doesn't work at all : nextflow run clipseq-1.0.0 -profile singularity,cluster --input /data/users/clabbe/Projets/2024_CLIPseq_BRAFidentificationRNABind_Vagner/testPipelineNFCore/testClipAlina/designFileTestClipNFCore.txt --genome GRCh37 --smrna_org human --peakcaller pureclip -c /data/users/clabbe/Projets/2024_CLIPseq_BRAFidentificationRNABind_Vagner/testPipelineNFCore/testClipAlina/curie.config -w /data/users/clabbe/Projets/2024_CLIPseq_BRAFidentificationRNABind_Vagner/testPipelineNFCore/workingDir N E X T F L O W ~ version 21.10.6 Launching
clipseq-1.0.0/main.nf
[jovial_hopper] - revision: e28c82e82e Unknown configuration profile: 'cluster'
Then i tried without and I get that : nextflow run clipseq-1.0.0 -profile singularity --input /data/users/clabbe/Projets/2024_CLIPseq_BRAFidentificationRNABind_Vagner/testPipelineNFCore/testClipAlina/designFileTestClipNFCore.txt --genome GRCh37 --smrna_org human --peakcaller pureclip -c /data/users/clabbe/Projets/2024_CLIPseq_BRAFidentificationRNABind_Vagner/testPipelineNFCore/testClipAlina/curie.config -w /data/users/clabbe/Projets/2024_CLIPseq_BRAFidentificationRNABind_Vagner/testPipelineNFCore/workingDir