Re: Problems raising fd limits in 1.3.78-1.3.80

sct@dcs.ed.ac.uk
Mon, 1 Apr 96 22:52 BST


Hi,

On Fri, 29 Mar 1996 23:26:41 -0500 (EST), David ``Joel Katz'' Schwartz
<stimpson@stimpson.igc.net> said:

> I used to have no problem raising the file descriptor and other
> limits in kernel versions 1.3.57-1.3.78. I could raise NR_OPEN, OPEN_MAX,
> and __FD_SETSIZE to 1024, NR_INODE to 4096, NR_FILE to 2048,
> __FD_SET_LONGS to 32, and then easily handle over 800 clients on my IRC
> server (igc.dal.net).

It is not recommended to increase __FD_SET's size. Unfortunately,
however large you make it, libc simply cannot cope with a select size
larger than 256 fd's.

As for the other parameters, you don't need to recompile the kernel.
In all post-1.3.57 kernels, you could achieve your inode and file
limits by (as root):
echo 4096 > /proc/sys/kernel/inode-max
echo 2048 > /proc/sys/kernel/file-max

Cheers,
Stephen.

--
Stephen Tweedie <sct@dcs.ed.ac.uk>
Department of Computer Science, Edinburgh University, Scotland.