Problems Using INLA on server

[Juan Ossa-Moreno]

Hello

I have used INLA for a while but started using it on a server this week and have not been able to run my model because found some issues creating the mesh.

I tried to run a simple example, to double check if my original code was wrong but I still get this:

 mesh=inla.mesh.create.helper(points=as.matrix(unique(coordinates[,list(UTMX,UTMY)])),max.edge=c(25000,100000))

/home/uqjossam/.R/library/INLA/bin/linux/64bit/fmesher: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.15' not found (required by /home/uqjossam/.R/library/INLA/bin/linux/64bit/fmesher)

/home/uqjossam/.R/library/INLA/bin/linux/64bit/fmesher: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by /home/uqjossam/.R/library/INLA/bin/linux/64bit/fmesher)

/home/uqjossam/.R/library/INLA/bin/linux/64bit/fmesher: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by /home/uqjossam/.R/library/INLA/bin/linux/64bit/libgsl.so.19)

Error in fmesher.read(prefix, "manifold") :

  File '/scratch/30275/Rtmp25iUCc/fmesher48db412c71e2.manifold' does not exist.

I would say the code is not the problem as I run the same code with the same input data on my machine and it worked fine.

I tried using another function to create the Mesh, but got something similar:

 mesh=inla.mesh.2d(loc=as.matrix(unique(coordinates[,list(UTMX,UTMY)])),max.edge=c(25000,100000))

/home/uqjossam/.R/library/INLA/bin/linux/64bit/fmesher: /usr/lib64/libstdc++.so.6: version `GLIBCXX_3.4.15' not found (required by /home/uqjossam/.R/library/INLA/bin/linux/64bit/fmesher)

/home/uqjossam/.R/library/INLA/bin/linux/64bit/fmesher: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by /home/uqjossam/.R/library/INLA/bin/linux/64bit/fmesher)

/home/uqjossam/.R/library/INLA/bin/linux/64bit/fmesher: /lib64/libc.so.6: version `GLIBC_2.14' not found (required by /home/uqjossam/.R/library/INLA/bin/linux/64bit/libgsl.so.19)

Error in fmesher.read(prefix, "manifold") :

  File '/scratch/30275/Rtmp25iUCc/fmesher48db69c68bb.manifold' does not exist.

I remove the package and installed it again, but the problem persist. Anyone has an idea why this may happen and how can I solve it? Could the R version has something to do with this?

Info from the server is as follows:

sessionInfo()

R version 3.3.0 (2016-05-03)

Platform: x86_64-redhat-linux-gnu (64-bit)

Running under: CentOS release 6.8 (Final)

Info from my machine

sessionInfo()

R version 3.2.3 (2015-12-10)

Platform: x86_64-w64-mingw32/x64 (64-bit)

Running under: Windows 7 x64 (build 7601) Service Pack 1

Thanks a lot in advance

Juan

[Håvard Rue]

Hi,

the system libraries on your server are to old, if you upgrade the
server to a newer linux version, it should be ok. 

meanwhile, you can use the static versions which are also available, 

library(INLA)
INLA:::inla.dynload.workaround()

and then proceed as normal. 

let me know if this does not work.

H

--
Håvard Rue
Department of Mathematical Sciences
Norwegian University of Science and Technology
N-7491 Trondheim, Norway
Voice: +47-7359-3533 URL : http://www.math.ntnu.no/~hrue
Mobile: +47-9260-0021 Email: havar...@math.ntnu.no

R-INLA: www.r-inla.org

[Juan Ossa-Moreno]

Hi 

The static version did work. Thanks a lot.

Regards

J

[Penny Hancock]

Hi, I am getting the fmesher error as well, although I have run the command INLA:::inla.dynload.workaround().  This used to fix the problem but it seems that this fix is no longer working on our server.  However the program runs fine on my local machine.  The server error says:

Error in fmesher.read(prefix, "s") : File '/tmp/RtmprVS8qy/fmesher24663f60817e.s' does not exist.

I am running INLA on a HPC server with R version 3.3.2 loaded.  If you have any suggestions for what to do I would much appreciate them!

Thanks,

Penny Hancock

[Helpdesk]

On Wed, 2017-11-08 at 08:27 -0800, Penny Hancock wrote:
> Hi, I am getting the fmesher error as well, although I have run the
> command INLA:::inla.dynload.workaround(). This used to fix the
> problem but it seems that this fix is no longer working on our
> server. However the program runs fine on my local machine. The
> server error says:
> Error in fmesher.read(prefix, "s") : File
> '/tmp/RtmprVS8qy/fmesher24663f60817e.s' does not exist.

sure that you're not using a mesh that is created in another R-session? there are temporary files
that does not survive at quit-and-restart R

H
--
Håvard Rue
Helpdesk
he...@r-inla.org

[Finn Lindgren]

The temporary files issue affected spde objects in early versions, but that issue was fixed some time ago.

This error message usually indicates that fmesher did not run, or crashed. If it runs on one machine but not another, it’s more likely a build/linking issue, or an issue with long running R sessions. In the latter case, restarting R should solve it.

Finn

> --
> You received this message because you are subscribed to the Google Groups "R-inla discussion group" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to r-inla-discussion...@googlegroups.com.
> To post to this group, send an email to r-inla-disc...@googlegroups.com.
> Visit this group at https://groups.google.com/group/r-inla-discussion-group.
> For more options, visit https://groups.google.com/d/optout.

[Zhe Sha]

Hi

I have the same problem after upgrading INLA to the latest version yesterday and even with the inla.dynload.workaround() I cannot run basic examples.

The error message I got without INLA:::inla.dynload.workaround() 

/home/R/x86_64-redhat-linux-gnu-library/3.5/INLA/bin/linux/64bit/inla: /lib64/libm.so.6: version `GLIBC_2.23' not found (required by /home/R/x86_64-redhat-linux-gnu-library/3.5/INLA/bin/linux/64bit/inla)

/home/R/x86_64-redhat-linux-gnu-library/3.5/INLA/bin/linux/64bit/inla: /lib64/libm.so.6: version `GLIBC_2.23' not found (required by /home/R/x86_64-redhat-linux-gnu-library/3.5/INLA/bin/linux/64bit/libRmath.so.1)

The error message I got after INLA:::inla.dynload.workaround() 

sh: line 1: 136611 Segmentation fault      (core dumped) '/home/R/x86_64-redhat-linux-gnu-library/3.5/INLA/bin/linux/64bit/inla.static' -b -s -v '/tmp/RtmpPjf8i6/file2133077072679/Model.ini' > '/tmp/RtmpPjf8i6/file2133077072679/Logfile.txt'

Is this caused by the libraries on my server? If so what I should ask the admin to upgrade?

Thank you!

Zhe

[INLA help]

Try the centos 7 build on inla.r-inla-download.org

--
Håvard Rue
Helpdesk
he...@r-inla.org

--

You received this message because you are subscribed to the Google Groups "R-inla discussion group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to r-inla-discussion...@googlegroups.com.

To post to this group, send email to r-inla-disc...@googlegroups.com.

[Alleggs M.]

Hi,
I've had the same issue. Downloading centos 7 solved it.

I ran something like:
wget "https://www.math.ntnu.no/inla/Linux-builds/CentOS 20Linux-7 20(Core)/Version_18.07.12/64bit.tgz"
tar xvf 64bit.tgz
#rename
mv 64bit.tgz /home/.../R_libs/INLA/bin/linux/

May be you need to remove/rename the old 64bit folder.

Best
Alex