Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!mit-eddie!uw-beaver!zephyr.ens.tek.com!tektronix!nosun!qiclab!onion!jeff
From: jeff@onion.pdx.com (Jeff Beadles)
Newsgroups: comp.unix.shell
Subject: Re: finding the filesystem of a file
Keywords: filesystem search find file
Message-ID: <1990Nov8.005050.16732@onion.pdx.com>
Date: 8 Nov 90 00:50:50 GMT
References: <6495@emory.mathcs.emory.edu>
Distribution: usa
Lines: 39

dt@mathcs.emory.edu (Dave Taylor {guest}) writes:
>Curious to know if anyone can tell me the fastest way to find out which
>filesystem a given file is on if the entire pathname is given.  I have only

Well, I didn't see the entire question, but if you're trying this from the
shell (This is comp.unix.shell, ya know... :-), and have a newer version of
df, then you might be able to do something like this:

% df /usr2/jeff/.newsrc

Filesystem    kbytes    used   avail capacity  Mounted on
/dev/ds08a    279922   32566  219363    13%    /usr2

From that, you could do something like:

------------------------snip here-----------------------------
#!/bin/sh
if [ -z "$1" -o ! -r "$1" -o $# -ne 1 ] ; then
	echo "Usage: $0 filename"
	exit 1
fi

## Do the df, print the 6th field of a line starting with '/'.
df $1 | awk '/^\// { print $6 } '

exit 0
------------------------snip here-----------------------------


Of course if your output of df is different, you may have to change the $6
to something else.

If you're trying to do this from 'C', then look at the statfs(2) man page.

Hope this helps!

	-Jeff
-- 
Jeff Beadles		jeff@onion.pdx.com