Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
selten
@selten
Sure
Nothing from it will get me in trouble here
Marius van den Beek
@mvdbeek
Thanks again, I suspect many more people have been running into variants of this
selten
@selten
At least we have a fix now and we can start splitting up large workflows into smaller ones with subworkflows
galaxybot
@galaxybot
[gronke] Two weeks ago I uploaded a paired collection that was around 0.99gb in size. Today, I tried reuploading the same set and I am getting Upload request failed. (0).
[gronke] How do I troubleshoot why this error is happening?
[gronke] It uploaded fine the first time.
Dannon
@dannon
On usegalaxy.org? There was an issue earlier, but I thought we had a temporary fix in place.
galaxybot
@galaxybot
[gronke] In fact ever since I started working with those files the whole server has been slowing down to a crawl. I've restarted it and have upped the cores/memory being used in the VCL instance. How can I fix this?
Dannon
@dannon
Ahh, ok, on your own server.
galaxybot
@galaxybot
[gronke] Dannon, no, on our own instance we are running
[gronke] I have the values in our virtualization config set to 16000 MB RAM, 8 cores, 1000 processor speed, and 10 Mbps Minimum Network Speed
[gronke] wondering what requirements I need or what I need to be doing to make this thing work
Dannon
@dannon
Hard to say, but I'd guess I'd probably start with looking at utilization with the current config
Is it memory, cpu, or network that's the bottleneck? Any particular runaway process you can see, etc?
galaxybot
@galaxybot
[gronke] Is that just looking at the processes and seeing what is using a lot?
[gronke] I have a workflow running, Samtools is currently using 26% CPU and 44% memory
Dannon
@dannon
Yeah, exactly, top/htop/iotop, etc.
galaxybot
@galaxybot
[gronke] that's about it, everything else is negligible
[gronke] I am not sure if that much usage would cause issues in the galaxy server or not
[gronke] but when I was googling that error message (upload request failed) results were saying that it means the server is not responding to requests, so maybe it is too busy with the samtools
Dannon
@dannon
Is this the 'new' multipart chunked upload, do you know? What release are you guys on?
Patricia Rodriguez
@phrd1994_gitlab
How can I add a reference sequence to my 16S workflow?
luciamh
@luciamh
Hello! I've got a issue with a script that I added to my local galaxy and was re-directed here after posting it on galaxy's biostar. Here's the link https://biostar.usegalaxy.org/p/28993/#28996
All the help is welcome :)
Dannon
@dannon
@phrd1994_gitlab That depends on the tool, I guess. Some will use references out of the history, others will operate on 'custom genomes' that you can create in your account
@luciamh is that input actually an integer id?
(from the screenshot it looks like it might be, instead of a regular dataset selector)
Patricia Rodriguez
@phrd1994_gitlab
How can I upload a reference genome from NCBI?
luciamh
@luciamh
@dannon for the ID parameter it's a random int, but for the parameters 'read 1' and 2 is a fastq file
Dannon
@dannon
Ahh, got it! I didn't scroll far enough on that image, I see the reads now.
What does the command portion of the script look like?
^ command portion of the tool.xml, that is.
luciamh
@luciamh

the command:

<command detect_errors="exit_code" interpreter="bash"><![CDATA[
lungcard_report.sh "$patientID" "$sampleID" "$birthday" "$gender" "$labKit" "$sequencer" "$vcf_file" "$idxstats" "$flagstat" "$fastq1" "$fastq2"
]]></command>

Dannon
@dannon
Well, I don't see anything immediately obvious, but what I'd do is open up lungcard_report.sh and dump the params at each step they are passed through, to verify.
On a tool run, if you look at the command line generated (visible on the dataset info interface), does that show the .dat files correctly?
luciamh
@luciamh
on the details of the run everything looks ok
Marius van den Beek
@mvdbeek
+1 to what dannon said, it looks like your commandline parsing may be off ? while you're at it, interpreter="bash" is deprecated, drop that part and use bash '$__tool_directory__/lungcard.sh'. Shouldn't affect the error you have reported though.
luciamh
@luciamh
the .dat files are correct
Marius van den Beek
@mvdbeek
I'd use the command line and run it outside of galaxy
luciamh
@luciamh
thank you @mvdbeek !
on the echo that I added the last two inputs are indeed incorrect, just like the ioerror states
Dannon
@dannon
Ah-ha!
Marius van den Beek
@mvdbeek
It doesn't look like that to me, but I haven't seen the code in lungcard.sh. You're trying to open the file 34564740, while that should be something like dataset_N.dat
that often happens when the arguments are not given in the expected order
luciamh
@luciamh
34564740 isn't a file it's the first random int I give in the first text input, and about the order with the echo it goes as expected
Marius van den Beek
@mvdbeek
can we look at the tool somewhere
?
the error you've posted is from a python script, so even if the echo is right there can still be plenty of things wrong
luciamh
@luciamh
true, but what I'm getting is that the problem is what is being inputted. I don't have the code online so I'm trying to upload it somewhere you may see