Hi Nate, Thanks for the note! So may I ask whether this is expected? I mean, the len_file_path points to tool-data/..., which is right inside galaxy-dist, and it is natural. I didn't realize that this path is relative to the individual job working dir. Are there any other similar cases? Best, Rui On Mon, Nov 18, 2013 at 10:40 AM, Nate Coraor <nate@bx.psu.edu> wrote:
On Sat, Nov 16, 2013 at 12:27 PM, ruiwang.sz <ruiwang.sz@gmail.com> wrote:
Hi All,
I‘m seeing some weird error messages...I googled but didn't see anything useful:
So, it is during the wigToBigwig conversion:
Dataset generation errors Dataset 47: Wig/BedGraph-to-bigWig on data 43 Tool execution generated the following error message: grep: writing output: Broken pipe grep: writing output: Broken pipe grep: writing output: Broken pipe and many more lines of the same error
The tool produced the following additional output: Couldn't open tool-data/shared/ucsc/chrom/mm10.len , No such file or directory
but it is there:
ls -l tool-data/shared/ucsc/chrom/mm10.len
-rw-rw-r-- 1 bioinfoadmin bioinfoadmin 1405 Oct 9 11:33 tool-data/shared/ucsc/chrom/mm10.len
Hi Rui,
Jobs run in a working directory based on the job id under whatever is configured as job_working_directory in universe_wsgi.ini. Thus, from that directory, tool-data/shared/ ... does not exist. To fix this, set len_file_path in universe_wsgi.ini to an absolute path.
--nate
From the Galaxy server log, I found these:
galaxy.tools WARNING 2013-11-15 17:46:43,200 Failed to resolve
dependency on
'ucsc_tools', ignoring
galaxy.jobs.runners.local DEBUG 2013-11-15 17:46:43,255 (216) executing: grep -v "^track" /home/bioinfoadmin/app/galaxy-dist/database/files/000/dataset_458.dat | wigToBigWig stdin tool-data/shared/ucsc/chrom/mm10.len /home/bioinfoadmin/app/galaxy-dist/database/files/000/dataset_473.dat -blockSize=256 -itemsPerSlot=1024 -clip 2>&1 || echo "Error running wigToBigWig." >&2
and then, setmetadata set the dataset state to ERROR.
I did verify that the result dataset_473.dat has a size 0.
However, when I run the command alone:
grep -v "^track" /home/bioinfoadmin/app/galaxy-dist/database/files/000/dataset_458.dat | wigToBigWig stdin tool-data/shared/ucsc/chrom/mm10.len /home/bioinfoadmin/app/galaxy-dist/database/files/000/dataset_473.dat -blockSize=256 -itemsPerSlot=1024 -clip
it successfully generated the dataset_473.dat.
also, I look into the wig_to_Bigwig.xml in tools/filters, the command is:
grep -v "^track" $input1 | wigToBigWig stdin $chromInfo $out_file1
however, I don't see where this "$chromInfo" was defined, yet it indeed found the
the correct value: tool-data/shared/ucsc/chrom/mm10.len. how does that happen?
it is very puzzling...I'm not sure if anyone has seen this before, please let me know!
Thanks,
Rui
___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: http://lists.bx.psu.edu/
To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/