Inappropriate ioctl for device

[nick-youngblut]

The log:

This is phyloFlash v3.4.2

[20:33:07] Using dbhome
           '/workspaces/genome_assembly/tmp/work/25/0e1e09f4cc53ded5f987a83f5db393/PHYLOFLASH_DB'
[20:33:07] working on library FINAL
[20:33:07] Forward reads read1.fq
[20:33:07] Reverse reads read2.fq
[20:33:07] Automatic read length detection
[20:33:07] running subcommand:
           readlength.sh readlength reads=10000 bin=1 in=read1.fq
           in2=read2.fq >FINAL.readlength.out 2>FINAL.readlength.err
[20:33:07] Reads are not all the same length, using longest read length
           found...
[20:33:07] Auto-detected read length: 151
[20:33:07] Running "everything" except EMIRGE- overrides other command line
           options
[20:33:07] Current operating system linux
[20:33:07] Checking for required tools.
[20:33:07] Using barrnap found at
           "/opt/conda/lib/phyloFlash/barrnap-HGV/bin/barrnap_HGV".
[20:33:07] Using reformat found at "/opt/conda/bin/reformat.sh".
[20:33:07] Using bbmap found at "/opt/conda/bin/bbmap.sh".
[20:33:07] Using sed found at "/opt/conda/bin/sed".
[20:33:07] Using fastaFromBed found at "/opt/conda/bin/fastaFromBed".
[20:33:07] Using mafft found at "/opt/conda/bin/mafft".
[20:33:07] Using cat found at "/bin/cat".
[20:33:07] Using spades found at "/opt/conda/bin/spades.py".
[20:33:07] Using vsearch found at "/opt/conda/bin/vsearch".
[20:33:07] Using grep found at "/bin/grep".
[20:33:07] Using readlength found at "/opt/conda/bin/readlength.sh".
[20:33:07] Using plotscript_SVG found at
           "/opt/conda/lib/phyloFlash/phyloFlash_plotscript_svg.pl".
[20:33:07] Using nhmmer found at
           "/opt/conda/lib/phyloFlash/barrnap-HGV/binaries/linux/nhmmer".
[20:33:07] Using awk found at "/usr/bin/awk".
[20:33:07] All required tools found.
[20:33:07] filtering reads with SSU db using minimum identity of 70%
[20:33:07] running subcommand:
           /opt/conda/bin/bbmap.sh fast=t minidentity=0.7 -Xmx20g reads=-1
           threads=6 po=f outputunmapped=f
           path=/workspaces/genome_assembly/tmp/work/25/0e1e09f4cc53ded5f987a83f5db393/PHYLOFLASH_DB
           out=FINAL.bbmap.sam outm=FINAL.read1.fq.SSU.1.fq noheader=t
           ambiguous=all build=1 in=read1.fq
           bhist=FINAL.basecompositionhistogram ihist=FINAL.inserthistogram
           idhist=FINAL.idhistogram scafstats=FINAL.hitstats overwrite=t
           outm2=FINAL.read1.fq.SSU.2.fq pairlen=1200 in2=read2.fq
           2>FINAL.bbmap.out
[20:34:20] done...
[20:34:20] Reading SAM file FINAL.bbmap.sam into memory
[20:34:20] Writing fixed SAM file to FINAL.read1.fq.SSU.sam
[20:34:20] Done
[20:34:20] Total read segments processed: 2399728
[20:34:20] insert size median: 0
[20:34:20] insert size std deviation: 0
[20:34:20] Summarizing taxonomy from mapping hits to SILVA database
[20:34:20] done...
[20:34:20] Forward read segments mapping: 5176
[20:34:20] Reverse read segments mapping: 5176
[20:34:20] Reporting mapping statistics for paired end input
[20:34:20] Total read pairs with at least one segment mapping: 5176
[20:34:20] => both segments mapping to same reference: 0
[20:34:20] => both segments mapping to different references: 5176
[20:34:20] Read segments where next segment unmapped: 0
[20:34:20] mapping rate: 0.431%
[20:34:20] subsampling SSU reads and running nhmmer to check coverage
           evenness across gene
[20:34:20] running subcommand:
           /opt/conda/bin/reformat.sh in=FINAL.read1.fq.SSU.1.fq
           out=FINAL.readsf.subsample.fasta srt=10000 ow=t
           in2=FINAL.read1.fq.SSU.2.fq 2>FINAL.reformat.out
[20:34:21] running subcommand:
           /opt/conda/lib/phyloFlash/barrnap-HGV/binaries/linux/nhmmer
           --cpu 6 --tblout FINAL.nhmmer.tblout
           /opt/conda/lib/phyloFlash/barrnap-HGV/db/ssu/ssu_ABE.hmm
           FINAL.readsf.subsample.fasta >/dev/null
[20:34:46] Fewer than 50 maps to eukaryotic SSU HMM model, skip plotting
           histogram...
[20:34:46] done
[20:34:46] creating phylotypes with SPAdes
[20:34:46] kmers for SPAdes are 99,111,127
[20:34:46] running subcommand:
           /opt/conda/bin/spades.py -o FINAL.spades -t 6 -m 20 -k
           99,111,127 -1 FINAL.read1.fq.SSU.1.fq -2 FINAL.read1.fq.SSU.2.fq
           >FINAL.spades.out 2>&1
[20:34:48] done...
[20:34:48] getting SSU phylotypes and their coverages...
[20:34:48] no phylotypes assembled with SPAdes
[20:34:48] possible causes include low or uneven coverage of SSU sequences
[20:34:48] writing final files...
[20:34:48] exporting results to csv
[20:34:48] generating graphics for report in SVG format
[20:34:48] Plotting mapping ID histogram
[20:34:48] running subcommand:
           /opt/conda/lib/phyloFlash/phyloFlash_plotscript_svg.pl --hist
           FINAL.idhistogram  --title="Mapping identity (%)" 
           >FINAL.plotscript.out 2>&1
[20:34:48] Plotting piechart of mapping ratios
[20:34:48] running subcommand:
           /opt/conda/lib/phyloFlash/phyloFlash_plotscript_svg.pl -pie
           FINAL.mapratio.csv -title="0.431 % pairs mapped"
           >FINAL.plotscript.out 2>&1
[20:34:48] Plotting histogram of insert sizes
[20:34:48] running subcommand:
           /opt/conda/lib/phyloFlash/phyloFlash_plotscript_svg.pl --hist 
           FINAL.inserthistogram  --title="Insert size (bp)" 
           >FINAL.plotscript.out 2>&1
[20:34:48] FATAL: Tool execution failed!.
           Error was 'Inappropriate ioctl for device' and return code
           '6400'
           Check log file FINAL.plotscript.out
           Check error log file &1
           Aborting.

It appears similar to https://github.com/HRGV/phyloFlash/issues/142.

I don't want any of the svg plots. @kbseah can you just add an option to turn off all plotting?

[kbseah]

As noted in issue #180, convenience switches -everything and -almosteverything override other options.

The SVG plots are produced for the HTML formatted report. If HTML output is turned off with -nohtml, plots will not be produced.

The appropriate options for output configuration (section 2.5 here) would therefore be: -nohtml -nozip -log.

[HRGV]

Sidenote: The read mapping patterns look weird, no pairs map on the same reference. Are these normal shotgun metagenomic reads? Looks like spades can't assemble anything, so you might also want to add -skip_spades and -skip_emirge to avoid the assembly steps that might break with such exotic data.