Error in fmesher.read(prefix, "manifold") on Mac

[Renato Morais]

Hi all,

I'm trying to run INLA from R 4.1.0 on a Mac

R version 4.1.0 (2021-05-18)

Platform: x86_64-apple-darwin17.0 (64-bit)

Running under: macOS Mojave 10.14.6

But when trying to build my 2d mesh, I get the following error (worth noting that the error persists in fresh sessions):

pts <- matrix(c(0,3,2,1,0,0,1,2,0,0,1,0), ncol=3)

mtest <- inla.mesh.create(loc=pts, tv=matrix(as.integer(c(2,3,3,4,1,1)), ncol=3))

/Users/jc340172/Library/R/4.1/library/INLA/bin/mac/64bit/fmesher.run: line 42: /Users/jc340172/Library/R/4.1/library/INLA/bin/mac/64bit/fmesher: No such file or directory

/Users/jc340172/Library/R/4.1/library/INLA/bin/mac/64bit/fmesher.run: line 42: exec: /Users/jc340172/Library/R/4.1/library/INLA/bin/mac/64bit/fmesher: cannot execute: No such file or directory

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

  File '/var/folders/jb/9571fckx7jd9c0djq6gxg9xw5d4mw5/T//Rtmpf78zir/fmesher6f467978641.manifold' does not exist.

I've seen this error message coming up on a series of past messages in this list.

Any quick, obvious, fixes? I appreciate the help.

Cheers,

Renato

[Renato Morais]

Maybe worth complementing that INLA has been installed today from source, via devtools and GitHub, given that I was not being successful on downloading via the INLA repository

[Helpdesk]

A quick fix is to run with R CRAN version R-4.0.5 and reinstall INLA.

this seems to be some kind of 'setup issue' where the path is wrong.
if you do

searchpaths()

where is INLA installed ?

> --
> 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 view this discussion on the web, visit
> https://groups.google.com/d/msgid/r-inla-discussion-group/4092aab4-1fdd-477a-9b68-9dd6f3871792n%40googlegroups.com
> .

--
Håvard Rue
he...@r-inla.org

[Helpdesk]

locate the binary and then set variables as follows

KW-18302:~ rueh$ locate inla.mkl.run
/Library/Frameworks/R.framework/Versions/4.0/Resources/library/INLA/bin/
mac/64bit/inla.mkl.run

KW-18302:~ rueh$ locate fmesher.run
/Library/Frameworks/R.framework/Versions/4.0/Resources/library/INLA/bin/
mac/64bit/fmesher.run


kw-16659:~ rueh$ R
R version 4.0.4 (2021-02-15) -- "Lost Library Book"
Copyright (C) 2021 The R Foundation for Statistical Computing
Platform: x86_64-apple-darwin17.0 (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

Natural language support but running in an English locale

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

Loading required package: Matrix
Loading required package: foreach
Loading required package: parallel
Loading required package: sp
This is INLA_21.05.02 built 2021-05-03 11:17:14 UTC.
- See www.r-inla.org/contact-us for how to get help.
- To enable PARDISO sparse library; see inla.pardiso()
> library(INLA)
>
inla.setOption(inla.call="/Library/Frameworks/R.framework/Versions/4.0/R
esources/library/INLA/bin/mac/64bit/inla.mkl.run")
>
inla.setOption(fmesher.call="/Library/Frameworks/R.framework/Versions/4.
0/Resources/library/INLA/bin/mac/64bit/fmesher.run")

> --
> 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 view this discussion on the web, visit

> https://groups.google.com/d/msgid/r-inla-discussion-group/19d67839-f5e6-4fc2-bdcb-7b41fd7d9a97n%40googlegroups.com

[Finn Lindgren]

If you installed it directly from the github repository you won’t have the compiled binaries, as they need to be built separately. The inla repository package versions contain the binaries. What was the problem when trying to install from those?

Finn

On 24 May 2021, at 03:31, Renato Morais <renatomor...@gmail.com> wrote:

Maybe worth complementing that INLA has been installed today from source, via devtools and GitHub, given that I was not being successful on downloading via the INLA repository

--

[Renato Morais]

Hi Finn and Havard,

Thank you for your responses.

Finn, when trying to download the binaries, the download stops at 30 MB and I get the following error:

trying URL 'https://inla.r-inla-download.org/R/stable/bin/macosx/contrib/4.1/INLA_21.02.23.tgz'

Content type 'application/x-tar' length 185894595 bytes (177.3 MB)

========

downloaded 30.3 MB

Error in download.file(url, destfile, method, mode = "wb", ...) : 

  download from 'https://inla.r-inla-download.org/R/stable/bin/macosx/contrib/4.1/INLA_21.02.23.tgz' failed

In addition: Warning messages:

1: In download.file(url, destfile, method, mode = "wb", ...) :

  downloaded length 31801344 != reported length 185894595

2: In download.file(url, destfile, method, mode = "wb", ...) :

  URL 'https://inla.r-inla-download.org/R/stable/bin/macosx/contrib/4.1/INLA_21.02.23.tgz': Timeout of 60 seconds was reached

Warning in download.packages(pkgs, destdir = tmpd, available = available,  :

  download of package ‘INLA’ failed

Havard,

searchpaths() shows that INLA in my current version is installed in "/Users/jc340172/Library/R/4.1/library/" alongside all non-base packages, which are in "/Library/Frameworks/R.framework/Versions/4.1/Resources/library".

However,

C02MX057F8J8:~ jc340172$ locate inla.mkl.run

/Library/Frameworks/R.framework/Versions/4.1/Resources/library/INLA/bin/linux/64bit/inla.mkl.run

/Library/Frameworks/R.framework/Versions/4.1/Resources/library/INLA/bin/mac/64bit/inla.mkl.run

C02MX057F8J8:~ jc340172$ locate fmesher.run

/Library/Frameworks/R.framework/Versions/4.1/Resources/library/INLA/bin/linux/32bit/fmesher.run

/Library/Frameworks/R.framework/Versions/4.1/Resources/library/INLA/bin/linux/64bit/fmesher.run

/Library/Frameworks/R.framework/Versions/4.1/Resources/library/INLA/bin/mac/32bit/fmesher.run

/Library/Frameworks/R.framework/Versions/4.1/Resources/library/INLA/bin/mac/64bit/fmesher.run

C02MX057F8J8:~ jc340172$

Yet, going to /Library/Frameworks/R.framework/Versions/4.1/Resources/library I can't see INLA there. I can't load INLA from R running in the Terminal either.

Is this a matter of messed up paths then?

Thanks again, cheers,

Renato 

--

Renato Morais | ARC Postdoctoral Research Associate

Research Hub for Coral Reef Ecosystem Functions

College of Science and Engineering

ARC Centre of Excellence for Coral Reef Studies

James Cook University

Townsville, Queensland, Australia

https://www.reeffunctionhub.org

http://thebellwoodreeffishlab.com

 

-- 

[Finn Lindgren]



As said in the warning messages, you hit the default 60 seconds download timeout set by R itself. You can increase that with

  options(timeout=600)

for example, for 10 min timeout instead.

Finn

On 24 May 2021, at 08:18, Renato Morais <renatomor...@gmail.com> wrote:



[Renato Morais]

Thank you,

I knew it was something stupid, but had no idea what. Plus, R installed INLA in a different library because I didn't have activated admin rights. Re-installing in Frameworks and downloading INLA properly solved the issue.

Thanks again,

Cheers

[Kay Yan]

Hi Harvard,

I encountered the same error. When I tried "locate inla.mkl.run " and "locate fmesher.run ", it returns that "WARNING: The locate database (/var/db/locate.database) does not exist."

I tried reinstall the package by ```install.packages("INLA",repos=c(getOption("repos"),INLA="https://inla.r-inla-download.org/R/stable"), dep=TRUE)```. Loading the library(INLA) seems to be fine, as it returns "Loading required package" etc. But the error persists. 

Thanks in advance.

[Helpdesk]

for locate, you might need to do

updatedb

as root, and maybe enable a service so its updated regularly.

try using the testing version (replace /stable with /testing) and use R-
4.1

[Kay Yan]

Thanks for the prompt response. 

Emm the question might be stupid.. how do I `updatedb`? In the R Terminal? 

I tried reinstall R CRAN and tried install INLA package again. It failed to install and showed error: 

Error in download.file(url, destfile, method, mode = "wb", ...) : 

  download from 'https://inla.r-inla-download.org/R/testing/bin/macosx/contrib/4.1/INLA_21.07.10-1.tgz' failed

./configure: line 3769: Darwin: command not found

[INLA help]

updatedb is a system command

Increase timeout in ‘options’ ?

—

Haavard Rue

HelpDesk 

help@r-inla. org

To view this discussion on the web, visit https://groups.google.com/d/msgid/r-inla-discussion-group/011a61e0-bd48-4bcd-adcc-4415ad8f2b6dn%40googlegroups.com.

[Kay Yan]

Hi, I updatedb by "sudo /usr/libexec/locate.updatedb" (it worked) and increased timeout. A new error popped up: 

Error in download.file(url, destfile, method, mode = "wb", ...) : 

  download from 'https://inla.r-inla-download.org/R/testing/bin/macosx/contrib/4.1/INLA_21.07.10-1.tgz' failed

Warning in install.packages :

  download of package ‘INLA’ failed

[Kay Yan]

It worked after following this: 

I clicked the link "https://inla.r-inla-download.org/R/testing/bin/macosx/contrib/4.1/INLA_21.07.10-1.tgz", downloaded the .tgz and installed the package from local by "install.packages(path_to_file, repos = NULL, type="source")". 

Thanks!!! 

[Erin Bryce]

Hi, I'm having a similar problem/error message when I run "inla.mesh.2d". Any help would be greatly appreciated. I'm using R 4.1.1 and downloaded INLA as described above and a simple inla() works fine. The error looks like this:

mesh <- inla.mesh.2d(loc= coords, 

+                      max.edge = c(2, 5),

+                      cutoff   = 0.001)

fmesher.cc(1066) 1D data not implemented.

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

  File '/var/folders/30/d2pcg3md7f9dcnb7vh7tsmpc0000gn/T//RtmpgQgQN6/fmesher30175578c4fc.manifold' does not exist. 

 

[John M. Aiken]

having installed using:

install.packages("INLA",repos=c(getOption("repos"),INLA="https://inla.r-inla-download.org/R/stable"), dep=TRUE)

then i change the options like this:

inla.setOption(inla.call="/fp/homes01/u01/ec-johnmai/R/x86_64-pc-linux-gnu-library/4.3/INLA/bin/mac/64bit/inla.mkl.run")

inla.setOption(fmesher.call="//fp/homes01/u01/ec-johnmai/R/x86_64-pc-linux-gnu-library/4.3/INLA/bin/mac/64bit/fmesher.run")

I am having the same error message. I tried setting these options like this, but its still not finding this mesh file

and it returns this message which states the manifold file doesn't exist. 

sh: 1: //fp/homes01/u01/ec-johnmai/R/x86_64-pc-linux-gnu-library/4.3/INLA/bin/mac/64bit/fmesher.run: not found Error in fmesher.read(prefix, "manifold") : File '/tmp/RtmpdxC9D2/fmesher140aba30e33ead.manifold' does not exist. In addition: Warning message: In system(paste(shQuote(fmesher.call), all.args, shQuote(prefix)), : error in running command

You can see here my version of R

platform x86_64-pc-linux-gnu arch x86_64 os linux-gnu system x86_64, linux-gnu status major 4 minor 3.0 year 2023 month 04 day 21 svn rev 84292 language R version.string R version 4.3.0 (2023-04-21) nickname Already Tomorrow

[John M. Aiken]

attempting to install from the package locally seems to have broken everything.

> install.packages("/fp/homes01/u01/ec-johnmai/repos/tblakes/INLA_21.07.10-1.tgz", repos = NULL, type="source") Installing package into ‘/fp/homes01/u01/ec-johnmai/R/x86_64-pc-linux-gnu-library/4.3’ (as ‘lib’ is unspecified) * installing *binary* package ‘INLA’ ... * DONE (INLA)

[John M. Aiken]

I tried this here

https://groups.google.com/g/r-inla-discussion-group/c/qPrEJrTSvN4

But actually it leads me to believe that maybe the issue is there is not a build prepared for Arch Linux.

[Helpdesk (Haavard Rue)]

please follow the instructions on r-inla.org on how to install R-INLA.
Use the testing version as you have R-4.3.

you're installing an old package which is the issue

On Wed, 2023-06-14 at 07:39 -0700, John M. Aiken wrote:
> I tried this here
>
> https://groups.google.com/g/r-inla-discussion-group/c/qPrEJrTSvN4
>
> But actually it leads me to believe that maybe the issue is there is
> not a build prepared for Arch Linux.
> On Wednesday, June 14, 2023 at 4:18:31 PM UTC+2 John M. Aiken wrote:
> > attempting to install from the package locally seems to have broken
> > everything.
> >
> > > install.packages("/fp/homes01/u01/ec-
> > > johnmai/repos/tblakes/INLA_21.07.10-1.tgz", repos = NULL,
> > > type="source")Installing package into ‘/fp/homes01/u01/ec-

> > > johnmai/R/x86_64-pc-linux-gnu-library/4.3’ (as ‘lib’ is
> > > unspecified)* installing *binary* package ‘INLA’ ...* DONE (INLA)

[John M. Aiken]

I have followed the instructions however they do not seem to work. I still get the manifold error.

below you can see where i download using the testing option,  and it still creates the manifold error

Restarting R session... > library(INLA) Error in library(INLA) : there is no package called ‘INLA’ > install.packages("INLA", repos = c(getOption("repos"), INLA = "https://inla.r-inla-download.org/R/testing"), dep = TRUE) Installing package into ‘/fp/homes01/u01/ec-johnmai/R/x86_64-pc-linux-gnu-library/4.3’ (as ‘lib’ is unspecified) Warning in install.packages : dependencies ‘Rgraphviz’, ‘graph’ are not available trying URL 'https://inla.r-inla-download.org/R/testing/src/contrib/INLA_23.06.15.tar.gz' Content type 'application/x-gzip' length 75739352 bytes (72.2 MB) ================================================== downloaded 72.2 MB * installing *binary* package ‘INLA’ ... * DONE (INLA) The downloaded source packages are in ‘/tmp/Rtmpv9kzDH/downloaded_packages’ > source("~/repos/tblakes/tblakes_4dm.R") Loading required package: Matrix Loading required package: sp This is INLA_23.06.15 built 2023-06-15 06:01:53 UTC. - See www.r-inla.org/contact-us for how to get help. Error in fmesher.read(prefix, "manifold") : File '/tmp/Rtmpv9kzDH/fmesher19dc966a5216ba.manifold' does not exist.

[Finn Lindgren]

Hi John,

would be easier to test with basic calls to inla() or fmesher code to
rule out any issues with the input data, and it might give clearer
error messages.

fit <- inla(y~1, data = data.frame(y = rnorm(10)), family="normal",
verbose=TRUE)

and

mesh <- inla.mesh.2d(cbind(0,0), offset = 1, max.edge = 2)

In your case, you might also check if you can build the experimental
fmesher package, with
remotes::install_github("inlabru-org/fmesher")
and try
fmesher::fmesher_rcdt(list(),loc=cbind(0,0))
It doesn't have the functionality you need yet, but might help
identify a build problem.

Finn

> To view this discussion on the web, visit https://groups.google.com/d/msgid/r-inla-discussion-group/70657d9e-d18b-482e-8df6-bdb17e87f0e2n%40googlegroups.com.



--
Finn Lindgren
email: finn.l...@gmail.com

[John M. Aiken]

Well, in fact, these simple examples work.

I was able to build fmesher. And the fit and mesh (shown below) simple examples both work.

So it's unclear to me now why they are not working in my codes:

> source("~/repos/tblakes/tblakes_4dm.R") [1] "i never know what is going on" Error in fmesher.read(prefix, "manifold") : File '/tmp/Rtmpv9kzDH/fmesher19dc963396e3ff.manifold' does not exist.

Where the data works,

> fmesher::fmesher_rcdt(list(),loc=cbind(0,0)) $idx [,1] [1,] -1 $s [,1] [,2] [,3] [1,] 0 0 0 [2,] 0 0 0 [3,] 0 0 0 [4,] 0 0 0 [5,] 0 0 0 [6,] 0 0 0 [7,] 0 0 0 [8,] 0 0 0 [9,] 0 0 0 $segm.bnd.grp [,1] [1,] 0 [2,] 0 [3,] 0 [4,] 0 [5,] 0 [6,] 0 [7,] 0 [8,] 0 $segm.bnd.idx [,1] [,2] [1,] 1 2 [2,] 2 3 [3,] 3 4 [4,] 4 5 [5,] 5 6 [6,] 6 7 [7,] 7 8 [8,] 8 1 $segm.int.grp [,1] $segm.int.idx [,1] [,2] $tt [,1] [,2] [,3] [1,] -1 1 -1 [2,] -1 2 0 [3,] -1 3 1 [4,] -1 4 2 [5,] -1 5 3 [6,] -1 -1 4 $tti [,1] [,2] [,3] [1,] -1 2 -1 [2,] -1 2 1 [3,] -1 2 1 [4,] -1 2 1 [5,] -1 2 1 [6,] -1 -1 1 $tv [,1] [,2] [,3] [1,] 1 2 3 [2,] 1 3 4 [3,] 1 4 5 [4,] 1 5 6 [5,] 1 6 7 [6,] 1 7 8 $vt [,1] [1,] -1 [2,] 0 [3,] 0 [4,] 0 [5,] 1 [6,] 2 [7,] 3 [8,] 4 [9,] 5 $vv $vv$i [1] 1 1 1 1 1 1 1 2 2 3 3 3 4 4 4 5 5 5 6 6 6 7 7 7 8 8 $vv$j [1] 2 3 4 5 6 7 8 1 3 1 2 4 1 3 5 1 4 6 1 5 7 1 6 8 1 7 $vv$x [1] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 $vv$dims [1] 9 9 $manifold [1] "R2" > > mesh <- inla.mesh.2d(cbind(0,0), offset = 1, max.edge = 2) > > mesh $meta $meta$call inla.mesh.create(loc = loc, boundary = boundary1, interior = interior1, extend = FALSE, refine = list(min.angle = min.angle[1], max.edge = max.edge[1], max.edge.extra = max.edge[1], max.n.strict = max.n.strict[1], max.n = max.n[1]), cutoff = cutoff, plot.delay = plot.delay, crs = crs) $meta$fmesher.args [1] "--input=input.s --cutoff=1e-12 --boundary=input.segm.bnd.idx --boundarygrp=input.segm.bnd.grp --rcdt=21,2,2 --max_n0=-1 --max_n1=-1 " $meta$time [,1] [,2] [,3] [,4] [,5] pre 0 0 0 0 0 fmesher 0 0 0 0 0 post 0 0 0 0 0 object 0 0 0 0 0 total 0 0 0 0 0 $meta$prefix [1] "/tmp/Rtmpv9kzDH/fmesher19dc9650f6bb3b." $meta$is.refined [1] TRUE $manifold [1] "R2" $n [1] 9 $loc [,1] [,2] [,3] [1,] -0.4142136 -1.0000000 0 [2,] 0.4142136 -1.0000000 0 [3,] 1.0000000 -0.4142136 0 [4,] 1.0000000 0.4142136 0 [5,] 0.4142136 1.0000000 0 [6,] -0.4142136 1.0000000 0 [7,] -1.0000000 0.4142136 0 [8,] -1.0000000 -0.4142136 0 [9,] 0.0000000 0.0000000 0 $graph $graph$tv [,1] [,2] [,3] [1,] 1 2 9 [2,] 9 5 6 [3,] 3 9 2 [4,] 9 4 5 [5,] 1 9 8 [6,] 7 8 9 [7,] 6 7 9 [8,] 4 9 3 $graph$vt [,1] [1,] 1 [2,] 1 [3,] 3 [4,] 4 [5,] 2 [6,] 2 [7,] 6 [8,] 5 [9,] 1 $graph$tt [,1] [,2] [,3] [1,] 3 5 NA [2,] NA 7 4 [3,] 1 NA 8 [4,] NA 2 8 [5,] 6 NA 1 [6,] 5 7 NA [7,] 6 2 NA [8,] 3 NA 4 $graph$tti [,1] [,2] [,3] [1,] 1 3 NA [2,] NA 2 2 [3,] 1 NA 1 [4,] NA 3 3 [5,] 1 NA 2 [6,] 1 1 NA [7,] 2 2 NA [8,] 3 NA 3 $graph$vv 9 x 9 sparse Matrix of class "dgTMatrix" [1,] . 1 . . . . . 1 1 [2,] 1 . 1 . . . . . 1 [3,] . 1 . 1 . . . . 1 [4,] . . 1 . 1 . . . 1 [5,] . . . 1 . 1 . . 1 [6,] . . . . 1 . 1 . 1 [7,] . . . . . 1 . 1 1 [8,] 1 . . . . . 1 . 1 [9,] 1 1 1 1 1 1 1 1 . $segm $segm$bnd $loc NULL $idx [,1] [,2] [1,] 1 2 [2,] 2 3 [3,] 3 4 [4,] 4 5 [5,] 5 6 [6,] 6 7 [7,] 7 8 [8,] 8 1 $grp [,1] [1,] 0 [2,] 0 [3,] 0 [4,] 0 [5,] 0 [6,] 0 [7,] 0 [8,] 0 $is.bnd [1] TRUE $crs NULL attr(,"class") [1] "inla.mesh.segment" $segm$int $loc NULL $idx [,1] [,2] $grp [,1] $is.bnd [1] FALSE $crs NULL attr(,"class") [1] "inla.mesh.segment" $idx $idx$loc [1] 9 $idx$lattice NULL $idx$segm [1] 1 2 3 4 5 6 7 8 $crs NULL attr(,"class") [1] "inla.mesh"

[Finn Lindgren]

OK, that's a good start, as being able to run both inla() and
inla.mesh.2d() indicates the build itself is working.
I haven't changed the fmesher code in INLA in a while (although the
build methods and settings may have changed).

The "missing manifold" happens both when fmesher crashes and when the
generated mesh is empty (for whatever reason), e.g. due to the
boundary specification defining an empty mesh, or some other input
data problem.
So I'd first check the inputs your code is using for the mesh generator.

Finn

> To view this discussion on the web, visit https://groups.google.com/d/msgid/r-inla-discussion-group/4976afd0-1e93-4438-b932-b6a052c034e9n%40googlegroups.com.

[John M. Aiken]

blah. okay i think you are correct Finn. There was a mispecification in the input data. I had this same error message before and it was because of an installation issue. So i guess that is where my confusion came from. Thank you for your help.