I've noticed this only on long running vims in my Linux server. I've narrowed down the E484s to lines of my vimscript that call system.

I just wonder if there is a way to deal with this because so far i have just been closing and restarting vim if this happens. But since the problem is so specific, it would be nice if it could be addressed so i can continue to use the running instance.

It's tricky to test and iterate on this because it takes at least 3 weeks before a vim instance gets into this state.

1 Answer 1


This happens when some system cron job cleans out "unused" files from /tmp. The solutions are:

  1. Find that cron job and stop it from rm'ing your Vim /tmp files.

  2. Recreate Vim's temporary directory.

Given the temporary directory name from the error message, you can recreate the directory manually with "!mkdir <dirname>", or you can use this command which you can put in your ~/.vimrc:

command! Mktmpdir call mkdir(fnamemodify(tempname(),":p:h"),"",0700)

That command was contributed to the vim_use list by Ben Schmidt.

  • This is wonderful, exactly the kind of manual workaround i need, thanks!
    – Steven Lu
    Aug 6, 2018 at 18:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.