sct> Hi,
Hi Stephen
sct> However, for nr_inodes to exceed max_inodes, we must be failing to reap
sct> the inode cache aggressively enough. What sort of load are you running
sct> under?
"two thirds of bugger-all"
9:19am up 6 days, 22:30, 16 users, load average: 0.03, 0.09, 0.08
16 users, all of them me. (rxvt's)
This machine is my personal workstation - it runs several rxvt's and
displays a netscape (which is executed on another machine with it's
display set to mine) - that's about it.
Oh, and it runs a little perl script every 5 minutes that checks to see
all our servers are still up, and pages me if one of them is dead.. I've
been getting false alarms lately because the machine is running out of
inodes so ping etc can't create new filehandles
(at least I'm assuming that's what's happening)
live:~# cat /proc/sys/fs/inode-nr
108480 13681
live:~# cat /proc/sys/fs/inode-max
131072
It seems to have stablised now though - it's been at this value for almost
3 days.
- Matt
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/