On Fri, Oct 06, 2000 at 03:43:59PM +0300, Peter Pentchev wrote:
> On a RELENG_4 machine with the world rebuilt on Sep 27, 'top' gave me
> the following output after sorting by the 'SIZE' field..
>
> last pid: 424; load averages: 0.17, 0.15, 0.10 up 0+00:29:43 15:39:23
> 46 processes: 1 running, 45 sleeping
> CPU states: 4.7% user, 0.0% nice, 1.2% system, 3.1% interrupt, 91.1% idle
> Mem: 19M Active, 202M Inact, 40M Wired, 80K Cache, 61M Buf, 240M Free
> Swap: 512M Total, 512M Free
>
> PID USERNAME PRI NICE SIZE RES STATE TIME WCPU CPU COMMAND
> 125 root 2 0 257M 560K select 0:00 0.00% 0.00% rpc.statd
> 244 mysql -2 0 12180K 11172K getblk 1:14 5.47% 5.47% mysqld
> 411 root 2 0 2116K 1580K select 0:00 0.00% 0.00% sshd
> 269 root 2 0 2116K 1580K select 0:00 0.00% 0.00% sshd
> [snip more processes]
>
> Aaaaaall right then.. what's the deal? :) Yes, I know that those 257
> megabytes are not *really* used, allocated, hogged and so on.. but why
> does statd ask for so much? :) And that's just 29 minutes after a reboot :)
Use The Source Luke. Taking a quick look at the source memory is only
allocated in two places. On malloc which doesn't account for this and
one f*ing hugh mmap of a file in /var to allow for extension of the
file. The code in -current mmaps 0x10000000 bytes of address space or
exactly 256MB. This is compleatly harmless as it not only doesn't use
any real memory, but since it's mmaping a file in /var it doesn't use
any swap either. Check it out for your self at:
/usr/src/usr.sbin/rpc.statd
-- Brooks
--
Any statement of the form "X is the one, true Y" is FALSE.
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
