[Mailman-Users] know to mkfs /var/spool with a higher inode
Phil Stracchino
alaric at babylon5.babcom.com
Mon Jun 18 20:41:24 CEST 2001
On Mon, Jun 18, 2001 at 02:04:26PM -0400, Bob Puff at NLE wrote:
> > Which operating system?
>
> Linux (Mandrake 7.x) on a i586.
>
> I have no "newfs". I do have a mke2fs. What is a suggested value for "bytes per inode" to handle the mail stuff we have been discussing? What is the default setting?
I quote again from 'man mke2fs':
-i bytes-per-inode
Specify the bytes/inode ratio. mke2fs creates an
inode for every bytes-per-inode bytes of space on
the disk. The larger the bytes-per-inode ratio,
the fewer inodes will be created. This value gen-
erally shouldn't be smaller than the blocksize of
the filesystem, since then too many inodes will be
made. [...]
Now, about a dozen lines prior to this, you'll find:
-b block-size
Specify the size of blocks in bytes. Valid block
size values are 1024, 2048 and 4096 bytes per block.
If omitted, mke2fs block-size is determined by the
file system size and the expected usage of the
filesystem (see the -T option).
So, your smallest allowed blocksize is 1024 bytes (1k), and the largest
legal number of inodes is one per block, which means the inode ratio and
blocksize resulting in the highest number of usable inodes is generated by
using -b 1024 -i 1024.
This is the setting I normally use for mail spools, news spools, /tmp,
etc -- anywhere you expect a large number of relatively small files to be
created.
You would do well to read through `man mke2fs` in detail and learn what
all the options do. Knowing how your tools work is a good
thing. Remember, always RTFM first.
--
Linux Now! ..........Because friends don't let friends use Microsoft.
phil stracchino -- the renaissance man -- mystic zen biker geek
alaric at babcom.com halmayne at sourceforge.net
2000 CBR929RR, 1991 VFR750F3 (foully murdered), 1986 VF500F (sold)
More information about the Mailman-Users
mailing list