I recently tried to run an R script from command line that takes input from stdin:
echo "print this" | rscript.R
The input was generated from a python script. Suddenly, R crashed and ever since whenever I try and open Rstudio I get the following message:
rstudio_boost:::filesystem::read_symlink: Access is denied: /path
I have tried to reinstall Rstudio and R, but the problem remains. R runs fine from R GUI
Error from logs:
RStudio-Desktop\log\rsession-User:
04 Jul 2019 09:46:22 [rsession-User] ERROR Unexpected exception: rstudio_boost::filesystem::read_symlink: Access is denied: "C:/Users\User\My Documents"; LOGGED FROM: int __cdecl rstudio::r::session::RReadConsole(const char *,char *,int,int) c:\jenkins\workspace\ide\windows-v1.2\src\cpp\r\session\rstdcallbacks.cpp:387
After playing around, I finally decided to delete the whole folder in:
C:\Users\User\AppData\Local\Rstudio-Desktop
This fixed the problem. What the origin of the problem is remains a mystry...
Related
We've been trying to get spyder 5.1.5 to work inside a windows server environment (version 2012 and 2016). Before We've been using Spyder 3.3.6 successfully, but we want to upgrade to get "black/autopep8" autoformatting.
We have this distributed CIFS-like shared drive inside this server that has some strict permissions (but my username definitely has read/write/modify access at the level where the script is).
Unfortunately, when I try to get spyder to SAVE my code (and autoformat using black), I get the following (trimmed) error below. Basically spyder cannot save due to a permission error that doesn't make sense to me.
2021-12-08 18:23:52,005 [DEBUG] [spyder.plugins.completion.providers.languageserver.client] -> python Response error: {'code': -32602, 'message': "PermissionError: [WinError 5] Access is denied: '\\\\\\\\data-dist\\\\DataVol_014\\\\group\\\\users\\\\myusername\\\\myscript.py'", 'data': {'traceback': [' File "C:\\ProgramData\\Anaconda3\\envs\\py37-dev\\lib\\site-packages\\pylsp_jsonrpc\\endpoint.py", line 116, in consume\n self._handle_request(message[\'id\'], message[\'method\'], message.get(\'params\'))\n', '
Troubleshooting this further, I certainly have read/write/modify access to the following levels of this path:
\\data-dist\DataVol_014\group\users\myusername\myscript.py
\\data-dist\DataVol_014\group\users\myusername\
\\data-dist\DataVol_014\group\users\
However, I cannot even view this level:
\\data-dist\DataVol_014\group\
Still this does not make sense to me. I don't understand why spyder would need this level of access.
Any help/guidance is appreciated.
The error certainly goes away if I copy the script to a local drive on said server and the black autoformatting works as expected.
I'm working through some exercises in an Introduction to R book (A Dummy's Book to be specific). I am running Rstudio version 1.1.447. I Enter the following script
h <-hello
yourname <-readline("What is your name")
Print(paste(h,yourname))
I then press Ctrl+Shift+S. When I do this, I get the following error message:
Error in file(con, "w") : cannot open the connection
In addition: Warning message:
In file(con, "w") :
cannot open file 'C:/Users/efami/OneDrive/Documents/.active-rstudio-document': Permission denied
I am running RStudio in Administrative Mode in a Windows 10 system. I do not understand where this path is coming from (it's obviously picking up something in my environment) or how to get past this problem.
I still don't know why it picked up the directory shown in the error message but the trick is to override the directory by creating a .Renviron file. The instructions on how to do this are found here:
Change temporary directory
I am working on installing rethinkdb on my server, which is a CentOS 6.9 machine. I followed all the directions for a CentOS server as described here for CentOS 6. I got all the way to the command scl enable devtoolset-2 -- make but while building, on line 244 of 421, the line that reads build/release/obj/extproc/http_job.o I receive the error "CURLOPT_ACCEPT_ENCODING was not declared in this scope"
I don't know what this means or why the error is popping up, but I can't continue with the build process as long as that error is there.
I found an obscure article that mentions a fix for this particular error. If you go into the rethinkdb folder on the server, you can then navigate to src/extproc/http_job.cc and open that file in a text editor. You can then search for the function void set_default_opts and inside of that function is the line exc_setopt(curl_handle, CURLOPT_ACCEPT_ENCODING, "", "PROTOCOLS); If you just change CURLOPT_ACCEPT_ENCODING to `CURLOPT_ENCODING, you will be able to continue the build process. It worked for me.
I've installed Netbeans 6.0.1 on Mac OSX 10.5.8 successfully, but when I try start it up it doesn't and these log entries are logged in System.log:
Apr 7 19:35:09 henrique-ordines-macbook [0x0-0x9c09c].NetBeans 6.0.1[53785]: /Applications/NetBeans/NetBeans 6.0.1.app/Contents/MacOS/netbeans: line 48: ls: command not found
Apr 7 19:35:09 henrique-ordines-macbook [0x0-0x9c09c].NetBeans 6.0.1[53785]: /Applications/NetBeans/NetBeans 6.0.1.app/Contents/MacOS/netbeans: line 53: dirname: command not found
Apr 7 19:35:09 henrique-ordines-macbook [0x0-0x9c09c].NetBeans 6.0.1[53785]: /Applications/NetBeans/NetBeans 6.0.1.app/Contents/MacOS/netbeans: line 57: dirname: command not found
Apr 7 19:35:09 henrique-ordines-macbook [0x0-0x9c09c].NetBeans 6.0.1[53785]: Cannot read cluster file: /../etc/netbeans.clusters
Apr 7 19:35:09 henrique-ordines-macbook com.apple.launchd[65] ([0x0-0x9c09c].NetBeans 6.0.1[53785]): Exited with exit code: 1
I've repaired permissions using Disk Utility, as suggested on this thread, but I still get the same error.
If I log in as the Root user I can start it without problems.
How can I fix this?
As described on this thread, the "ls command not found" error was happening because my PATH environment variable had been broken somehow. By me in this case, uhem, anyway, once you've fixed that variable so that this part /usr/bin:/bin:/usr/sbin:/sbin:/usr/local/bin:/usr/X11/bin can also be in the PATH variable, Netbeans works again.
I recently installed R on a new Macbook Pro running Mountain Lion. When trying to plot data, I get a slew of ATSFontGetFileReference errors, and text doesn't render (though chart elements do):
> plot(study$outcome, study$predictor)
Jan 21 10:55:57 tatou.local R[35029] <Error>: ATSFontGetFileReference failed: error -50.
Jan 21 10:55:57 tatou.local R[35029] <Error>: ATSFontGetFileReference failed: error -50.
Jan 21 10:55:57 tatou.local R[35029] <Error>: ATSFontGetFileReference failed: error -50.
Jan 21 10:55:57 tatou.local R[35029] <Error>: ATSFontGetFileReference failed: error -50.
...
It seems like more of an OS X issue than R, since a little Googling has turned up similar errors for PDF renderers, etc. But I can't seem to find a solution...
All my fonts as well as R appear to be installed for use system-wide. Ideas?
I had the same error using RStudio on OSX 10.7.5. I saved my workspace and then quit RStudio. Upon opening it back up and loading my saved workspace, I ran the plot(x,y) command and there was no longer an error, with the text being displayed as expected.
I don't know your font situation exactly, but I just fixed this issue on mine. I have a mac, and helvetica is automatically installed as a .dfont file. I had downloaded a helvetica.ttf file and put it in 'mac_hd/libary/fonts'. Once I deleted this and restarted R I haven't had the problem. My error was verbatim ' ATSFontGetPostScriptName failed: error -984. ' (I included that to help with google results)