Using read_tsv twice crashes R, how to fix?

Question

I am trying to create a tibble from a multitude of TSV-Textfiles. I first created a list containing the paths to my textfiles called text_files. I then tried to loop through these textfiles to create one dataframe with all info.

for (file_path in text_files) {
  # Read in the TSV file
  tibble <- read_tsv(file_path, col_names = TRUE)
  
  # Extract file name for the TSV file
  file_name <- basename(file_path)
  
  # Add file name to tibble
  tibble$file_name <- file_name
  
  # Append the tibble to the all_data tibble
  all_data <- bind_rows(all_data, tibble)
}

Back in February my script still worked.Now it crashes and gives me a fatal error.

I have narrowed down the issue and see that it comes from "read_tsv", because I can get it to crash using just this:

file_path <- text_files[[1]]
tibble <- read_tsv(file_path, col_names = TRUE)

file_path <- text_files[[2]]
tibble <- read_tsv(file_path, col_names = TRUE)

As soon as I try to overwrite the tibble (or even save it as a new tibble), R crashes because of a fatal error. This happens only for a specific textfile - the ones with only 1 line of data work fine, and the ones with 9 lines of data cause the crash, but only if I read in another textfile with 9 lines of data before.

Does anyone know what the issue seems to be here, and more importantly how to fix it so I can read in all my data?

When playing around with it, I renamed file_path and tibble to other names, and got it to work once without crashing, but not reproducibly so.

Edit: Now it sometimes crashes even at the first try.