Re: corrupted cache

From: Jeffrey Borg <jeffrey@dont-contact.us>
Date: Thu, 14 Oct 1999 17:44:56 +1000 (EST)

You can change the reserved blocks on the fly with a existing system so
this is a 500Mb squid cache and you can see the results

[root@tux twig2beta2]# df
Filesystem 1024-blocks Used Available Capacity Mounted on
/dev/sda3 495927 443393 26919 94% /usr/local/squid
[root@tux twig2beta2]# tune2fs -m 0 /dev/sda3
tune2fs 1.10, 24-Apr-97 for EXT2 FS 0.5b, 95/08/09
Setting reserved blocks percentage to 0 (0 blocks)
[root@tux twig2beta2]# df
Filesystem 1024-blocks Used Available Capacity Mounted on
/dev/sda3 495927 443393 52534 89% /usr/local/squid
[root@tux twig2beta2]#

On Thu, 14 Oct 1999, Kendall Lister wrote:

> On Thu, 14 Oct 1999, Phillip Morgan wrote:
>
> > I have a rather embarrassing problem. On my 8gig IDE drive when
> > formatted unled linux 2.0.36 df shows the total blocks as 7956k. So I
> > set my cache size for Squid 2.2.stable4 to 7950 16 256.
> >
> > When the 8gig drive is full a maximum of 7548k blocks are used, but df
> > shows the disk as 100% full. Unfortunaetly over 400 megs have gone to
> > the great void, and squid never reached 98%, resulting in an out of
> > disk space error and squid started crashing.
>
> That 400 Mb looks a lot like 5% of 8 Gb, which is the portion of disk
> space reserved for super user access by the ext2fs. For a cache partition,
> you might well want to turn this down to 0%, by giving mke2fs the -m 0
> option.
Received on Thu Oct 14 1999 - 02:00:13 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:48:49 MST