On Mon, 2008-01-14 at 08:21 +0100, Andreas Kuntzagk wrote: > Am Sonntag, den 13.01.2008, 10:16 +0700 schrieb Fajar Priyanto: > > On Thursday 10 January 2008 23:21:55 techlists at comcast.net wrote: > > > Is there a switch in "find" (or some other command besides find) that'll > > > let you find files larger than a specified size? > > > > > > My file system is 88% full and I'd like to see where the biggest space > > > hoggers are. > > > > I also found this on the net: > > du /path/to/anywhere/* -hs | grep [0-9]M | sort -rn | head -20 > > This only shows you usage for directories less than 1GB. (and more than > 1MB) > To see all: > > du /path/to/anywhere/* -s | sort -rn | head -20 May I suggest (g)awk? That way you'll get all, not just 20, of what you want. du -s *|sort -rn|gawk --re-interval '/^[[:digit:]]{4,}\t/' - This shows dirs with block counts of 1000 or more. And then there is perl etc. Usually these threads get long as everyone jumps in with their personal favorite, including me here. :-) And smaller dirs can be identified with du -s *|sort -rn|gawk --re-interval '/^[[:digit:]]{,3}\t/' - BTW, I was surprised that the 4.* implementation defaults required the "--re-interval" switch. Hmmm. HTH -- Bill > > _______________________________________________ > CentOS mailing list > CentOS at centos.org > http://lists.centos.org/mailman/listinfo/centos