I am trying to install a package from GitHub in support of deploying an app to shinyapps.io. If I download the package onto my machine, then it installs without issue. However, if I execute the command (in RStudio):
devtools::install_github(repo="rlmoore0/NMDeploy", subdir="NymbulCredentials")
then I get an error message: Error in (function (n) :
lazy-load database '../R/NymbulCredentials.rdb' is corrupt
This GitHub repo is public, so anyone can try running this command from within R/RStudio with devtools installed. I am on a Windows box running R 3.6.3, Win10.
Any suggestions?
I have looked at questions from other people with similar problems and based on the responses to them, I have tried:
Closing and restarting R/RStudio
Executing the command devtools::unload("NymbulCredentials")
Executing the command unloadNamespace("NymbulCredentials")
Removing the installation library the package is being installed to as referenced by .libPaths() and trying a fresh directory as the installation location.
The error still persists.
Related
I recently had to have a new laptop build at work and therefore reinstalled R and R Studio through a designated 'Company Portal' that did the software installation for me (this might be the problem). When now trying to install and load packages for use, I am running in to the below error for packages (in this case for tidyverse):
Warning message: In options(stringsAsFactors = TRUE) :
'options(stringsAsFactors = TRUE)' is deprecated and will be
disabled Error in setwd("C:/R/WorkingDir") : cannot change working
directory Execution halted Warning in install.packages :
installation of package ‘tidyselect’ had non-zero exit status
I also had this error when I first loaded R up that seems to be reappearing due any attempt to install packages:
Error in setwd("C:/R/WorkingDir") :
cannot change working directory
I've never had this issue before using RStudio across numerous machines and software versions, can anyone advise what is going on? My name (and thus C:\ user file path(s)) does have an apostrophe in it, if that matters - but it has never been an issue before. Is some key installation setting now misaligned and if yes, can anyone advise how to resolve?
I have an R environment in Anaconda with all the packages I need and use. I added a private repository from github onto my pc using the conda CLI. The project files includes an renv folder as well a .Rproj file. When I try to open any of the R scripts in the project without opening the Rproj file to open the project, they all show up as blank, and when I try to open a csv file it says it doesn't exist. When I open the Rproject, I cannot access any of my libraries or packages that I usually have access to in R. When I tried to install a package in this environment, I got the error:
warning in system(paste(cmd, "shlib-clean")) : 'make' not found warning in system(cmd) : 'make' not found error: compilation failed for package 'pkgload' * removing 'c:/users/poona/<repo name>/renv/staging/1/pkgload'
What's going wrong? I just want to access the scripts of the repository and load the packages of my usual environment. Any help is appreciated!
I'm trying to install jupyter_nbextensions_configurator on windows10 for python3.7 using conda.
An error occurred while installing the package.
I have already tried to reinstall anaconda,but the problem still occurs.
ERROR conda.core.link:_execute(700): An error occurred while installing package 'conda-forge::jupyter_nbextensions_configurator-0.4.1-py37_0'.
Rolling back transaction: done
LinkError: post-link script failed for package conda-forge::jupyter_nbextensions_configurator-0.4.1-py37_0
location of failed script: C:\Users\Cesare\Anaconda3\Scripts\.jupyter_nbextensions_configurator-post-link.bat
I want to know how to install it successfully.
I too was facing the same problem. Here's the output when I tried cloning the base environment [Running as Administrator]:
(base) PS C:\WINDOWS\system32> conda create --name django --clone base
WARNING: A conda environment already exists at 'C:\Users\Tanishk\anaconda3\envs\django'
Remove existing environment (y/[n])? y
Source: C:\Users\Tanishk\anaconda3
Destination: C:\Users\Tanishk\anaconda3\envs\django
The following packages cannot be cloned out of the root environment:
- conda-forge/win-64::conda-4.9.2-py38haa244fe_0
- defaults/win-64::conda-build-3.20.5-py38_1
- defaults/win-64::conda-env-2.6.0-1
Packages: 305
Files: 2077
Preparing transaction: done
Executing transaction: done
ERROR conda.core.link:_execute(698): An error occurred while installing package 'defaults::openssl-1.1.1h-he774522_0'.
Rolling back transaction: done
CondaError: Cannot link a source that does not exist. C:\Users\Tanishk\anaconda3\pkgs\openssl-1.1.1h-he774522_0\Library\bin\openssl.exe
Running `conda clean --packages` may resolve your problem.
()
You will notice the package giving me problem is different than yours, but the ERROR remains the same (ERROR conda.core.link).
Here's how I solved it:
Google the package mentioned in the ERROR. In my case it was openssl-1.1.1h-he774522_0
Open the result from Anaconda Files (should likely be the first search result):
anaconda / packages / openssl 1 - Files :: Anaconda Cloud
Ctrl+F to find the package in the list.
Download the compressed package to your local Downloads folder.
Backup the contents of the installed erroneous package *C:\Users\Tanishk\anaconda3\pkgs\openssl-1.1.1h-he774522_0* somewhere.
Copy the contents of the de-compressed downloaded package into the *C:\Users\Tanishk\anaconda3\pkgs\openssl-1.1.1h-he774522_0* folder.
Re-try in Anaconda prompt with whatever you got stuck at.
Note:
After step 7, I got another same error but for a different package this time (vs2015_runtime-14.16.27012-hf0eaf9b_3). I performed the same steps for this package too --> Google -- Download -- Replace old content. And I was able to run things successfully after that.
I suggest to google the package because if you search for the package name on Anaconda itself, it will ask you to login first.
I've recently installed R + Rstudio on my work-laptop and now I'm trying to install packages. I have full authorisation in the map that I'm downloading the packages to. However, I keep on getting the following error:
Error in install.packages : ERROR: failed to lock directory
‘\\hltfs001\Home$\jolien.jansen\Documents\R\win-library\3.6’ for
modifying. Try removing
‘\\hltfs001\Home$\jolien.jansen\Documents\R\win-library\3.6/00LOCK’
This error is accompanied by this error:
Warning in install.packages :
'lib = "\\hltfs001/Home$/jolien.jansen/Documents/R/win-library/3.6"'
is not writable
I've tried to install the packages in a different folder, which I'm also authorised to do, but that did not solve the issue.
I've googled the error and that gave me a couple of things to try.
At first I tried manually removing the LOCK folder, and restarting R. That did not help
Then I tried this code:
unlink("/hltfs001/Home$/jolien.jansen/Documents/R/win-
library/3.6/00LOCK", recursive = TRUE)
Which again removed the LOCK folder, but didn't help with the error either
Finally I tried:
install.packages("Rcpp", dependencies=TRUE, INSTALL_opts = c('--no-
lock'))
But again, that gave me the same error.
I had the same problem and I found that my problem is that my 00LOCK folder was located within an automatic-syncing Dropbox folder. So, I paused the Dropbox sync and then the problem went away after I deleted the 00LOCK folder. Apparently, the Dropbox syncing was interfering with the file locking mechanism of the installation process. So, once I finished installing the packages, I resumed my Dropbox syncing.
I have been trying to deploy my shiny application in shiny server which i have installed on Linux VM (RHEL 6.5, 64 bit) by following the instructions given in this site
The server is started successfully but i get the following when i try to access the sample application provided as a part of the shiny server.
Error in dyn.load(file, DLLpath = DLLpath, ...) :
unable to load shared object '/usr/lib64/R/library/Rcpp/libs/Rcpp.so':
/usr/lib64/R/library/Rcpp/libs/Rcpp.so: invalid ELF header
In addition: Warning message:
package ‘shiny’ was built under R version 3.3.0
Error: package or namespace load failed for ‘shiny’ Execution halted
I installed all the required packages without any error still i got this exception when i tried to access the sample application.
The following are the steps i followed to install shiny server.
Installed R. In my office EPEL repository is blocked since it is maintained by a third party organization. So I downloaded the rpm file for R from EPEL website and installed the same using the rpm command.
Next i downloaded the shiny package from the CRAN repository and installed the same using the following command
R CMD INSTALL shiny_0.13.2.tgz
Finally i downloaded and installed the shiny server RPM file.
The server was successfully installed and started without any errors, yet i get the above error when trying to access the application deployed in the server.
I googled a lot for this error but could not find much. Has anybody else faced this issue? Could someone help me out with this?
I found out the issue. I installed all the packages as root user but shiny server runs under a different username "shiny". In order to solve this I started an R session using the following command
sudo R
Then I installed the packages using their source, as my LINUX VM (office machine)did not have internet access. I used the following command for installing the packages from the source. First we need to untar the source and then run the following command.
require(devtools)
install('path-to-the-untared-folder')
The below link helped me in finding out the solution.
R - shiny server on Ubuntu