[lkml]   [1998]   [Apr]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRE: Improving Ext2 Reliability
On Tue, 14 Apr 1998, Jordan Mendelson wrote:

> The cache takes quite a long time to rebuild and I don't feel like wasting
> bandwidth to save 30 minutes :)

You've missed a major concern: a cache with corrupt data is a disservice
to your users. Squid does *not* verify that the data contained within the
cache is correct. Some very good points to this effect have been made in
the recent thread 'NT causes $10M loss' on cold.system. Consistent
metadata buys you nothing if you don't verify the data's valid. Fsck is
*only* meant to exit when a filesystem has valid metadata which the
sysadmin can use to clean up the mess after a crash (preferably restoring
from backup).

> I'm finally just scrapping 2.0.x and moving to 2.1.9x so I can at least get
> alt-sysrq and some better debugging messages. Hell, 2.1.9x has been more
> stable on my development boxes than the 2.0.33+ on my production machines.
> :)

Good luck! =)


To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to

 \ /
  Last update: 2005-03-22 13:42    [W:0.059 / U:3.668 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site