----------------------Original Question------------------------
Is there some sort of max number of files per partition?
Obviously there must be because each file is an inode entry
in the super-block, but what is the calculation/rule of thumb?
----------------------Responders-------------------------------
Oli Thor Atlason <olit@centrum.is>
Kevin Sheehan Kevin.Sheehan@uniq.com.au
Glenn Satchell Glenn.Satchell@uniq.com.au
Roger Graham rgraham@mitsui.oz.au
Leif Hedstrom <leif@infoseek.com>
Peter Bestel Peter.Bestel@uniq.com.au
Nate Itkin <Nate-Itkin@ptdcs2.intel.com>
S. D. Raffensberger sdr@rdga3.att.com
Peter Allan peter.allan@aeat.co.uk
Jochen Bern bern@TI.Uni-Trier.DE
Alan Ronemus aadrrd1@peabody.sct.ucarb.com
R A Lichtensteiger rali@meitca.com -or- rali@hri.com
David H. Brierley <dhb@ssd.ray.com>
F.L. Charles Seeger III seeger@cis.ufl.edu
Louis M. Brune brunel@delver.iterus.org
Paulo Licio de Geus <paulo@dcc.unicamp.br>
----------------------Collective Answer------------------------
Special thanks to Leif Hedstrom
The default is one inode for 2048 bytes of file system - when you do
a newfs it tells you how many per cylinder group. You can use the
-i option to newfs to vary how many are made.
To check the number of inodes
"/usr/ucb/df -i" on Solaris 2.x
To change:
SunOS 4.1.x Solaris 2.x
df(1V) -i df(1M) -e or -g
newfs(8) -i bytes/inode newfs(1M) -i bytes/inode
mkfs(8) nbpi argument mkfs_ufs(1M) -o nbpi=n
fs(5) fs_ufs(4)
Big known Exception: USENET Spool Filesystems. There you should have
one Inode per kB (instead of per 2 kB).
---------------
Many thanks. Next time, I'll read the man pages first.
-phil
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:10:33 CDT