Re: tail -f to a dir

[klmitch () MIT EDU (Kev)]

> I do not remember this being mentioned, but if it has sorry.
> I was playing today and I did a "tail -f /home/mydir" on RH 6.2
> and I get:
>
> tail: /home/mydir: Is a directory
> tail: tail.c:718: recheck: Assertion `valid_file_spec (f)' failed.
> Abort (core dumped)
>
> Sometimes I get a core, sometimes not. Same result as root, but always =
> creates a core as root though.

hmmm...Linux won't allow directories to be opened with open(), and returns
the error EISDIR.  It seems that tail fails to stop after that, and has an
assertion failure...while this would be a bug, there shouldn't be any way
to use it...

> So I tried this on my SGI running IRIX 6.55 and it went back to a prompt =
> like nothing happened.

I forget what IRIX does when you open() a dir...

> Then I tried it on Solaris 2.6 and it actually spit the contents of the =
> dir. in binary form, but when it was done - it went back to a prompt.

Solaris allows opening directories and returns the contents of the directory
on subsequent reads, just as you've discovered ;)

> Any insight other than don't tail a directory?

HTH...

--
Kevin L. Mitchell <klmitch () mit edu>