In message <[EMAIL PROTECTED]>, Bernd Walter writes:
>On Fri, Mar 14, 2003 at 03:15:49PM +0100, Poul-Henning Kamp wrote:
>> In message <[EMAIL PROTECTED]>, Bernd Walter writes:
>> >The /dev/fd content looks strange:
>> >[160]cicely9# ls -al /dev/fd/
>> >total 33
>> >dr-xr-xr-x  2 root   wheel       512 Mar  6 16:59 .
>> >dr-xr-xr-x  5 root   wheel       512 Mar  6 17:00 ..
>> >crw--w----  1 ticso  tty      5,   1 Mar 14 14:50 0
>> >crw--w----  1 ticso  tty      5,   1 Mar 14 14:50 1
>> >crw--w----  1 ticso  tty      5,   1 Mar 14 14:50 2
>> >d-w-------  1 root   wheel       512 Feb 28 15:20 3
>> >d---------  1 root   wheel       512 Mar  6 16:59 4
>> 
>> What is strange about it ?
>
>After carefully rethinking - nothing.

That's why I'm against fdescfs and /dev/fd in principle:  They
are neither intuitive nor logical.

>I'd expected to see devnodes like we have without fdescfs and was
>surprised by the directories.
>I never used fdescfs before and needed /dev/fd/3 for a shell script.
>The script does not work with fdescfs, but of course this could be
>for a completly different reason.

You cannot access /dev/fd/3 until file desc #3 is actually open
so the order of arguments (and bugs in the shell) is important.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
[EMAIL PROTECTED]         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message

Reply via email to