28 Jan 2013 22:06
DEXSeq and parallel, orphan processes.
Bin Ye <robin.ye@...>
2013-01-28 21:06:01 GMT
2013-01-28 21:06:01 GMT
Dear list, I've been using DEXSeq as describe in vignette on our SGE cluster which worked as expected, but I noticed that when any command with nCore=N (N > 1) called, there will be N orphan processes left in background. After the command finished, those N processes will be in sleep and hold certain amount of memory, but 0 cpu. These just keep adding up until it crashes the compute node. I've tried to clean-up with multicore command kill(children()) in R, it just return NULL without doing anything. I had to manually kill in linux command line. And I can't find in vignette how to deal with this. Is this an issue with package parallel/multicore? My session info is attached below. Is there a way to clean-up in R after calling those multi-processes commands? It will be really helpful if the vignette can be updated on this issue. Thank you! Bin > sessionInfo() R version 2.15.1 (2012-06-22) Platform: x86_64-redhat-linux-gnu (64-bit) locale:  LC_CTYPE=en_US.iso885915 LC_NUMERIC=C  LC_TIME=en_US.iso885915 LC_COLLATE=en_US.iso885915  LC_MONETARY=en_US.iso885915 LC_MESSAGES=en_US.iso885915  LC_PAPER=C LC_NAME=C  LC_ADDRESS=C LC_TELEPHONE=C  LC_MEASUREMENT=en_US.iso885915 LC_IDENTIFICATION=C attached base packages:  stats graphics grDevices utils datasets methods base other attached packages:  DEXSeq_1.4.0 AnnotationDbi_1.20.3 Biobase_2.18.0  BiocGenerics_0.4.0 RSQLite_0.11.2 DBI_0.2-5 loaded via a namespace (and not attached):  annotate_1.36.0 biomaRt_2.14.0 clusterProfiler_1.6.0  colorspace_1.2-0 DESeq_1.10.1 dichromat_1.2-4  digest_0.5.2 DO.db_2.5.0 DOSE_1.4.0  genefilter_1.40.0 geneplotter_1.36.0 ggplot2_0.9.3  GO.db_2.8.0 GOSemSim_1.16.1 grid_2.15.1  gtable_0.1.2 hwriter_1.3 igraph_0.5.5-4  IRanges_1.16.4 KEGG.db_2.8.0 labeling_0.1  lattice_0.20-10 MASS_7.3-22 munsell_0.4  parallel_2.15.1 plyr_1.8 proto_0.3-9.2  qvalue_1.32.0 RColorBrewer_1.0-5 RCurl_1.95-3  reshape2_1.2.2 scales_0.2.3 splines_2.15.1  statmod_1.4.16 stats4_2.15.1 stringr_0.6.2  survival_2.36-14 tcltk_2.15.1 tools_2.15.1  XML_3.95-0.1 xtable_1.7-0 [[alternative HTML version deleted]] _______________________________________________ Bioconductor mailing list Bioconductor@... https://stat.ethz.ch/mailman/listinfo/bioconductor Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor