I'm running R version 3.0.2 and (to my knowledge) the most recent version of bioconductor on two different servers. On one of them, when I execute
source("http://bioconductor.org/biocLite.R")
biocLite("deepSNV")
I get some warnings, but can load the deepSNV library and run deepSNV. On the other server, the same sequence of commands gives the following error messages in spite of being the same version of R running the same version/source of bioconductor:
ld: cannot find -lrdmacm
make: *** [VariantAnnotation.so] Error 1
ERROR: compilation failed for package ‘VariantAnnotation’
* removing ‘/home1/<myaccount>/R/x86_64-unknown-linux-gnu-library/3.0/VariantAnnotation’
ERROR: dependencies ‘VGAM’, ‘VariantAnnotation’ are not available for package ‘deepSNV’
* removing ‘/home1/<myaccount>/R/x86_64-unknown-linux-gnu-library/3.0/deepSNV’
an attempt to correct this by biocLite("VariantAnnotation") gives:
make: *** [VariantAnnotation.so] Error 1
ERROR: compilation failed for package ‘VariantAnnotation’
* removing ‘/home1/<myaccount>/R/x86_64-unknown-linux-gnu-library/3.0/VariantAnnotation’
Do you have any idea as to why I'm experiencing this error, and how to best correct it? Note that I do not have root or mod/admin privileges, so anything I load or install is local on both servers.
Yes, I am using a cluster.
What I find frustrating is that while both of the servers are running R 3.0.2, I am able to load and run deepSNV on one of them but not the other.
sessionInfo() gives the following specifics for both servers (as I mentioned, they are configured identically, except that to run R 3.0.2 on the "problem" server, I need to load it via an intel module first):
> sessionInfo()
R version 3.0.2 (2013-09-25)
Platform: x86_64-unknown-linux-gnu (64-bit)
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
[3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
[5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
[7] LC_PAPER=en_US.UTF-8 LC_NAME=C
[9] LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
attached base packages:
[1] stats graphics grDevices utils datasets methods base
This sounds like a configuration issue and I'm not sure I can be of help. Have you asked your IT support to install R on the 'problem' server so you don't have to use a module? It would also be good to update R if possible since the version is 2 years old.
Maybe someone else on the list has experience with loading via modules on a cluster and will chime in.
Valerie
You're on the right track to 'drill down' to the underlying problem. So it looks like deepSNV fails to install because VariantAnnotation fails to install. Then you tried to install just VariantAnnotation (good), but it's not really clear from the output that you posted why it failed to install, just that 'compliation failed'. What comes *before* that message? If drilling down the problem is captured by the message about
ld: cannot find -lrdmacmthen this is quite weird -- there are no direct dependencies on that (C) library in VariantAnnotation, so it is likely something fairly deep and possibly not related to R, as Val said, about the configuration of that particular machine; more complete output of the attempt to install VariantAnnotation might help to understand at what point that library is being required.