RStudio crashes after running glm snippet

I have a student whose session keeps getting aborted with a fatal error when running a pretty simple model. It only crashes when she saves the model into an object. So, the following runs fine:

glm(formula,data,family=poisson)

But the following aborts:

modelname<-glm(formula,data,family=poisson)

It DOESN't crash when rendering a Quarto document with the code and it doesn't crash when running the code inside the R GUI. It only crashes when running the code from RSTUDIO.

RSTUDIO version: 2025.09.0 Build: 387
R version: 4.5.1

Does the RStudio log file contain any error messages?

The student sent me the log files and I can see various instances of error 10053 and of Error 32

I believe that error 10053 typically indicates RStudio being unable to communicate with the R session. Does error 32 come with the words "broken pipe"? That would also signal communication problems. Is the student running locally installed copies of R and RStudio, or is either of them running on a server? Also, does changing the names of the various pieces of the formula (for instance, model1 <- glm(formulax, some_data, family=poisson), assuming that the formula is provided in a variable rather than explicitly typed out in the call) affect the bug?

Broken pipe is not a part of the error. I copied and pasted the complete error log at the end of this reply.

The bug is consistent independently of the names in the formula. It happens in all these instances and with different names for the data:

model1<-glm(y~x,data,damily=poisson)
model2<-glm(y~1,data,damily=poisson)
model3<-glm(y~x1+x2+...,data,damily=poisson)

But not wen running the model by itself and not storing it in an object:

glm(y~x,data,damily=poisson)

We have figured a temporal workaround that does run and doesn't cause r to abort:

saveRDS(glm(y~x,data,damily=poisson),"name.RDS")
model1<-readRDS("name.RDS")

Error logs:
[rsession-jbigelo2] ERROR system error 10053 (An established connection was aborted by the software in your host machine) [request-uri: /help/library/stats/html/family.html]; OCCURRED AT void __cdecl rstudio::session::HttpConnectionImpl::sendResponse(const class rstudio::core::http::Response &) C:\Users\jenkins\workspace\ide-os-windows\rel-cucumberleaf-sunflower\src\cpp\session\http\SessionHttpConnectionImpl.hpp:156; LOGGED FROM: void __cdecl rstudio::session::HttpConnectionImpl::sendResponse(const class rstudio::core::http::Response &) C:\Users\jenkins\workspace\ide-os-windows\rel-cucumberleaf-sunflower\src\cpp\session\http\SessionHttpConnectionImpl.hpp:161
2025-10-01T14:33:54.973895Z [rsession-jbigelo2] ERROR system error 32 (The process cannot access the file because it is being used by another process) [path: C:/Users/jbigelo2/AppData/Local/RStudio/notebooks/861AF89C-Assignment5/1/s/c66mkeuxcrs8j_t/_rs_chunk_plot_001.png, target-path: C:/Users/jbigelo2/AppData/Local/RStudio/notebooks/861AF89C-Assignment5/1/s/c66mkeuxcrs8j/00001e.png]; OCCURRED AT class rstudio::core::Error __cdecl rstudio::core::FilePath::move(const class rstudio::core::FilePath &,enum rstudio::core::FilePath::MoveType,bool) const src/cpp/shared_core/FilePath.cpp:1499; LOGGED FROM: void __cdecl rstudio::session::modules::rmarkdown::notebook::ChunkExecContext::onFileOutput(const class rstudio::core::FilePath &,const class rstudio::core::FilePath &,const class rstudio::core::json::Value &,enum ChunkOutputType,unsigned int) C:\Users\jenkins\workspace\ide-os-windows\rel-cucumberleaf-sunflower\src\cpp\session\modules\rmarkdown\NotebookExec.cpp:417
2025-10-01T14:33:54.973895Z

This is a known bug in 2025.09.0; we have just released a patch (2025.09.1+401) that fixes this and some other bugs.